AN EVALUATION OF DESIGN: LOW IMPACT DEVELOPMENT VS.
TRADITIONAL DESIGN ON A SITE IN LOS ANGELES, CALIFORNIA
By
Morgan Haffey

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Environmental Design – Master of Arts
2017

ABSTRACT
AN EVALUATION OF DESIGN: LOW IMPACT DEVELOPMENT VS. TRADITIONAL
DESIGN ON A SITE IN LOS ANGELES, CALIFORNIA
By
Morgan Haffey
Because of the growth in urbanism, professionals in the environmental industry are often
tasked with designing strategies to accommodate urban development while minimizing
environmental impact (Selbig & Bannerman, 2008). Converting the natural landscape into
residential or commercial developments can drastically alter its hydrologic characteristics (Selbig
& Bannerman, 2008). This study compares an existing undeveloped site along the Los Angeles
River (scenario 1) with two design scenarios. One design is the city's conceptual plan, referred to
as a Mia Lehrer and Associates Design (design scenario 3). The second design incorporates low
impact design elements, referred to as the Extended LID Design (design scenario 2). Using these
three scenarios and the EPA stormwater calculator, runoff, infiltration, and evaporation will be
measured to determine which design is most effective along these three parameters. The
Friedman’s One Way of Variance Test was applied to the treatments in order to determine which
design was most successful concerning stormwater management. The results show that the
design with LID controls are significantly greater than the existing conditions (p<0.005). These
results are important because they show that LID controls are beneficial to site design when
stormwater management is a concern. These findings also show that it is important to investigate
low impact development for future sites and can raise awareness within the engineering
community.

ACKNOWLEDGEMENTS

I would like to thank my major advisor Dr. Patricia Machemer, Professor at Michigan
State University School of Planning, Design, and Construction. Without her help, support, and
inspiration this thesis would not have been possible.
I would also like to thank the remaining people on my committee, Dr. Jon Burley and Dr.
Zeenat Kotval-Karamchandani, professors at Michigan State University School of Planning,
Design, and Construction, for your advice, guidance, and knowledge of statistical analysis.
Lastly, I would like to thank all of my friends and family for their support and
encouragement throughout my career as a university student.

iii

TABLE OF CONTENTS

LIST OF TABLES…………………………………………………………………………...…..v
LIST OF FIGURES..…………………………………………………………………………....vi
Chapter 1: Introduction ............................................................................................................... 1
Chapter 2: Literature Review ...................................................................................................... 3
2.1 Low Impact Development......................................................................................................... 3
2.1.1 What is LID?................................................................................................................. 3
2.1.2 International Definitions of LID ................................................................................. 3
2.1.3 The Evolution of LID ................................................................................................... 4
2.1.4 LID Design Elements ................................................................................................... 6
2.1.5 Urbanization ................................................................................................................. 7
2.1.6 Opportunities/Benefits ................................................................................................. 8
2.1.7 Constraints .................................................................................................................. 10
2.2 Design Elements of LID ......................................................................................................... 11
2.2.1 Rain Gardens .............................................................................................................. 11
2.2.2 Green Roofs ................................................................................................................ 13
2.2.3 Permeable Paving ....................................................................................................... 14
2.2.4 Infiltration Basin ........................................................................................................ 17
2.2.5 Constructed Wetlands................................................................................................ 19
2.3 Recent LID Studies ................................................................................................................. 20
2.3.1 MSU Low Impact Investigations .............................................................................. 25
2.4 LID Summary ......................................................................................................................... 28
Chapter 3: Methods .................................................................................................................... 29
3.1 Study Site ................................................................................................................................ 29
3.2 Design Scenarios Description ................................................................................................. 30
3.2.1 Existing Site (Design Scenario 1) .............................................................................. 30
3.2.2 Extended LID Design (Design Scenario 2) ............................................................... 32
3.2.3 Mia Lehrer and Associates Design (Design Scenario 3) ......................................... 33
3.3 Measurement Elements- Stormwater Management ................................................................ 34
3.4 Statistical Method to Test the Hypotheses ............................................................................. 35
Chapter 4: Results....................................................................................................................... 39
Chapter 5: Discussion ................................................................................................................. 43
5.1 Limitations .............................................................................................................................. 46
5.2 Policy and Practice Recommendations ................................................................................... 48
5.3 Conclusion .............................................................................................................................. 49
BIBLIOGRAPHY ....................................................................................................................... 51
iv

LIST OF TABLES

Table 1. The National Stormwater Calculator (SWC) Parameters ............................................... 39
Table 2. SWC Statistic Results ..................................................................................................... 40
Table 3. Friedman’s Test Results.................................................................................................. 41

v

LIST OF FIGURES

Figure 1. Schematic Design of a Rain Garden .............................................................................. 12
Figure 2. Schematic Design of a Green Roof ............................................................................... 14
Figure 3. Schematic Design of Permeable Paving ........................................................................ 17
Figure 4. Schematic Design of an Infiltration Basin..................................................................... 19
Figure 5. Schematic Design of a Constructed Wetland ................................................................ 20
Figure 6. Aerial Image of Existing Site (adapted from Google Earth) ......................................... 30
Figure 7. Map of Taylor Yard Parcels (adapted from The River Project, 2011.) ......................... 31
Figure 8. Conceptual Rendering of Extended LID Design ........................................................... 33
Figure 9. Conceptual Plan Designed by Mia Lehrer and Associates ............................................ 34
Figure 10. SWC Runoff, Infiltration, and Evaporation Analysis Results (adapted from SWC) .. 40

vi

Chapter 1: Introduction
The phrase “low impact development” (LID) indicates systems or practices that emulate
natural processes that result in “infiltration, evapotranspiration or use of stormwater in order to
protect water quality and associated aquatic habitats” (USEPA, 2016). LID is more closely
related to land development and finding a way to manage stormwater as close to its source as
possible. LID used in the re-development process is a way of recreating or preserving the natural
landscape, and creating effective and aesthetic site drainage that treats stormwater as a valuable
resource instead of sending it directly into city/county drains as waste (USEPA, 2016).
Managing stormwater on site is extremely important considering that traditional urban areas are
primarily impervious surfaces and lead to a more diverse range of pollutants, reduce pollutant
removal during overland flow, reduced infiltration, and increased peak flows (Selbig &
Bannerman, 2008). LID practices are a way of managing water to reduce the effect and impact of
the urban areas.
Stormwater runoff is a major source of pollution in urban areas (USEPA, 2016). Because
of urbanization, rainfall cannot naturally seep into the ground. Traditional piping systems are
designed to catch rainfall and transport it to bodies of water even though stormwater from urban
areas is not naturally filtered and carries trash, bacteria, and other pollutants (USEPA, 2016).
These pollutants come from many factors of the urban sprawl such as parking lots, buildings,
pavement, and gasoline from cars. In natural areas, rainfall is absorbed and filtered by soil and
plants. The goal of LID is to reproduce or emulate this natural process in order to manage
stormwater nearest to its source.

1

Through research it is clear that there are many benefits in using LID for design purposes.
Research lacks evidence of the concerns of LID and why people are hesitant to use these design
elements. This research focuses low impact development elements, it compares and assesses the
ecological benefits and discusses practice of LID designs on new development sites,
redevelopment sites, and with urban retrofitting. By comparing specific design variables in each
design scenario, calculating the results using the EPA National Stormwater Calculator, and
determining LID elements and strategies, this study provides evidential support for adopting LID
practices for designers, developers, researchers, and government agencies.

2

Chapter 2: Literature Review
2.1 Low Impact Development
2.1.1 What is LID?
Low impact development (LID) is a technique that emulates the natural hydraulic cycle
by using principles modeled after nature such as infiltration, filtering, storage, evaporation, and
detaining runoff close to its source (Southeast Michigan Council of Governments, 2008). With
the ever growing rate of population and urban communities, protecting the process of the water
cycle is highly important. The use of LID elements is one way to reduce the negative impacts of
urbanization on the natural hydraulic systems (USEPA, 2000). The objective for low impact
development is to maintain the natural condition that was present before development has
occurred. Since the increase in urbanization is critical to the environment, LID seeks to decrease
the amount of paving, pipe systems, and stormwater structures by using a more natural approach
such as grass swales, constructed wetlands, green roofs, rain gardens, and bioswales.
2.1.2 International Definitions of LID
As previously mentioned, the use of the term low impact development has different
meaning around the world. In the United States it is defined as, “a system or practice that mimics
the natural processes that result in the infiltration, evapotranspiration or use of stormwater in
order to protect water quality and associated aquatic habitat” (USEPA, 2016). In the United
States low impact development is more commonly associated with stormwater management
practices. In the United Kingdom, LID is defined as “development that through its low impact
either enhances or does not significantly diminish environmental quality” (Sylvawood, n.d.).
Overtime many people have expanded on the definition of LID. According to Heather
Sylvawood, a study by the University of West England defined LID as “integrally connected
3

with land management and as much as describing physical development, LID also describes a
form of livelihood” (Sylvawood, n.d.). The study also states that LID is a “multi-featured and
intrinsically integrated form of development” (Sylvawood, n.d.). In the United Kingdom, LID
encompasses a variation of environmental elements “including visual quality, wildlife habitat
protection, air quality, and land consumption as we as stormwater volume and quality” (Wang et
al., 2017). Therefore it seems there is no simple definition of low impact development because
of how detailed the topic is and how it is implemented in different regions of the world.
2.1.3 The Evolution of LID
Since low impact development was first introduced, it has evolved due to use of different
terminology, descriptions, and practices and principles. Because of these differences the
evolution of LID has caused an increase in confusion, miscommunication, and misunderstanding
(Fletcher et al., 2017). The development and use of terminology in disciplines such as landscape
architecture, planning, and engineering is used in more of an informal manner, which is driven
by local and regional perspectives and understanding (Fletcher et al., 2017). This results in the
use of different terms that in time can be used to define similar concepts in different parts of the
world. Over time this will lead to overlaps, contradictions, and confusion (Fletcher et al., 2017).
For example, according to Urban Drainage Multilingual Glossary publish in 2004, which
includes drainage definitions in English, French, German, and Japanese, there are many terms
that have similar definitions between the four languages. But the glossary also demonstrates that
some terms and concepts could not be accurately translated from their original language
(Fletcher et al., 2017).
A good comparison is the use of terminology and language in the medical field. Medical
terminology maintains a common reference point meaning that all terminology is broken down
4

into concepts and relationships (Bronnert, Masarie, Naeymi-Rad, Rose, & Aldin, n.d.). Having
this efficiency within the medical field allows for improved communication between medical
offices and fields (Fletcher et al., 2017). Medical terminology is based around Nomina
Anatomica, or anatomical nomenclature, which removes the obstacle of confusion when
assessing new findings (Fletcher et al., 2017). This is along the same reason that plants are
classified by their taxonomy. Each type of plant is categorized in a kingdom, subkingdom, super
division, division, subdivision, class, subclass, order, family, and genus (USDA & Natural
Resources Conservation Service, 2017). Although plants have “common names”, scientific Latin
names are used to help describe both the genus and species of plants around the world. This is
similar to the use of anatomical nomenclature in the medical field.
The topic of urban drainage dates back to at least 3000 BC and has had a focus on the
movement of water away from urban areas (Fletcher et al., 2017). In recent years the topic of
sustainability has become very popular and over time has produced new terms (Fletcher et al.,
2017). Terms such as low impact development, sustainability, sustainable urban drainage
systems, water sensitive urban design, and best management practices are fairly new within the
last 20-30 years (Fletcher et al., 2017).
The term low impact development (LID) is most commonly used in North America and
New Zealand and was first used in 1977 in a report on land use planning in Vermont, USA
(Fletcher et al., 2017). The approach of LID originally was to “minimize the cost of stormwater
management by taking a design with nature approach” (Barlow 1977). The original intent of LID
was also to complete a “natural” hydrology system by the use of site layout and control measures
(Fletcher et al., 2017). This is a balance between pre-development runoff, infiltration, and
evapotranspiration volumes, which would be implemented through a “functionally equivalent
5

hydrologic landscape” (Fletcher et al., 2017). After some time, the widespread design
community had evolved the interpretation of LID from its original meaning to a broad set of
practices and designs that treated stormwater (Fletcher et al., 2017).
Today, since the definition of low impact development has evolved, most states have their
own best management practices manual that define what LID is and how it should be used. Most
countries have similar definitions, but it is not uncommon that the terminology varies. For
example in New Zealand the use of LID implies that the impact of the design is far lower than
that of the original practice (Fletcher et al., 2017). Since the definitions and terminology of LID
varies with location it is important that researchers and professionals communicate in order to
ensure that they are referring to the same concepts.
2.1.4 LID Design Elements
There are many different design elements related to low impact development and it is
important to understand the function of each of them in order to incorporate the most appropriate
element in site specific design. Different sites have varieties of issues concerning stormwater
management. Rain gardens for example are capable of increasing groundwater recharge and are
appropriate for sites with contamination as well as excess runoff (Dietz, 2007). Green roofs are
another LID design element that research has shown positive results with retention rates between
60% and 70% of precipitation (Dietz, 2007). Permeable paving is a great alternative to concrete
and asphalt paving because of its ability to infiltrate stormwater and decrease the rate and
amount of runoff in hardscaped areas (Dietz, 2007). Therefore, it is crucial that one understands
the different types of design elements associated with LID in order to design with stormwater
management in mind.

6

2.1.5 Urbanization
In the Unites States there are roughly 400 metropolitan areas, and the top 100 of them
occupy 12% of the nation’s landmass, generate 68% of our jobs, 75% of our national GDP, and
are home to 65% of the population (Katz, 2017). Clean water is an important part of sustaining
life, but urbanization is one of the many factors that are threatening our water resources.
According to Southeast Michigan Council of Government (2008), problems related to
stormwater runoff are most prevalent where urbanization has occurred. Conventional land
development alters the land and effects the water cycle (Southeast Michigan Council of
Governments, 2008). Altering one part of the water cycle causes changes to the other
components of the hydraulic cycle (Southeast Michigan Council of Governments, 2008). Due to
a large increase in impervious surfaces such as roads, buildings, and parking lots there is an
abundance of rainfall that runs off instead of soaking into the soil (National Asphalt Pavement
Association, 2016). The high percentage of impervious surface cover decreases the landscape’s
ability to absorb water (University of New Hampshire, 1995). According to The National
Asphalt Pavement Association or NAPA, the United States has more than 2.7 million miles of
paved roads and highways (National Asphalt Pavement Association, 2016). As water flows off
of impervious surfaces and runoff increases, the amount of groundwater recharge decreases
(Southeast Michigan Council of Governments, 2008). Impacts of stormwater runoff include
increased flooding and property damage, degradation of the stream channel, less groundwater
recharge and dry weather flow, impaired water quality, increased water temperature, loss of
habitat, and decreased recreational opportunities (Southeast Michigan Council of Governments,
2008). Katz (2017), a centennial scholar who focuses on the challenges and opportunities of
global urbanization, states that by 2050, 70 percent of the global population will live in an urban

7

area. With this large of an increase in urbanization it is important that we explore the benefits
and concerns of low impact design and how it has a positive impact on urban sprawl.
2.1.6 Opportunities/Benefits
Previous studies explore the beneficial uses of LID at many different scales from site
specific to entire watersheds. However, there is still a debate concerning the unexplored gaps of
low impact development and the effectiveness it has towards stormwater management issues
(University of New Hampshire, 1995). Even though LID is a fairly new concept, one benefit is
that it can be applied to new developments, urban retrofitting, and redevelopments in order to
help communities find a balance between public safety, economic development, and ecological
protection (University of New Hampshire, 1995). Since the goal of LID is to emulate the
natural landscape, pre-development hydrology measures of runoff rate and run off volume are
used in the development of LID elements (University of New Hampshire, 1995). In reality, the
amount of water that leaves a site, whether it is a new development or a redevelopment, should
match the “same rate, quality, and quantity of water that existed in the predevelopment
condition” (University of New Hampshire, 1995).
A project in Ingham County, Michigan concerning Lansing’s most polluted pipe outlet is
being redesigned to incorporate many LID elements to replace the storm piping system that is
currently on site (“Ingham County Drain Commissioner Plans Massive Urban Retrofit,” n.d.).
This project is a great example of the use of LID elements and their positive impact on the
environment. For this “massive urban retrofit” the drain commissioner has resorted to green
infrastructure to replace the gray infrastructure because the inclusion of all LID elements is a
fraction of the cost of replacing the storm pipes (“Ingham County Drain Commissioner Plans
Massive Urban Retrofit,” n.d.). The LID features that are incorporated into the future design of
8

this urban retrofit are constructed wetlands, rain gardens, bio-rention areas, ponds and biodigestive green walls (“Ingham County Drain Commissioner Plans Massive Urban Retrofit,”
n.d.). Currently the piping system collects runoff for around 800 acres and rushes the
contaminated water directly into the Red Cedar River causing it to flood (“Ingham County Drain
Commissioner Plans Massive Urban Retrofit,” n.d.). According to project planners, it is expected
that the “retrofit will reduce pollutant loading by 95%” (“Ingham County Drain Commissioner
Plans Massive Urban Retrofit,” n.d.).
Another benefit to low impact development is that it does not require large amounts of
stormwater piping and infrastructure. Its goal is to reduce the amount of “downstream structural
practices” and concentrate on maximizing “soil filtration/infiltration, biological uptake of water
and nutrients, and cultivation of useful microbe populations” that are found in natural soils in
order to transform compound contaminants found in stromwater (University of New Hampshire,
1995). Integrating low impact development strategies instead of underground piping systems
into future designs is an effective way to deal with issues concerning runoff reduction, water
quality and quantity treatment, and flood control (University of New Hampshire, 1995). Many
studies indicate that it is possible for LID controls to reduce hydraulic impacts of development
instead of traditional stormwater systems (Selbig & Bannerman, 2008).
Another advantage of LID is that it addresses stormwater with smaller, cost effective
features that are scattered throughout an entire development site, which can ultimately replace
the ineffective and costly solution of traditional pipe and pond management (University of New
Hampshire, 1995). LID elements that this research will address are rain gardens, constructed
wetlands, green roofs, permeable pavement, and infiltration basins. Attempting to identify

9

“sensitive resources” on site is important when distinguishing which LID elements are
appropriate on site.
Because low impact development can be applied at many different scales (University of
New Hampshire, 1995) it is viewed favorably and shows versatility. This scale spans from
watershed level down to individual site design. Resource conservation, pollution prevention, and
decentralization of runoff are three factors that can be addressed at a watershed scale.
Minimizing cut and fill, reducing impervious surfaces, and strategic timing of runoff are
elements that can be dealt with at a site level (University of New Hampshire, 1995). By
analyzing the use of LID from large scale to small scale issues, one can see that this is a
beneficial factor when considering the use of LID principles.
While there are many benefits of low impact development, there are also concerns that
need to be addresses. With a lack of research, it is difficult to determine why people are hesitant
to incorporate LID when developing a site design.
2.1.7 Constraints
Communities may be hesitant to incorporate LID into designs because municipal decision
making often happens in elective cycles rather than within long term planning (Woolson, 2013).
However, while low impact development may show short term stormwater improvement, but is
more productive when considering long term effectiveness. Cost concerns are a large issue with
low impact development and the materials that are required. People are not willing to pay for a
more expensive alternative even though there is research that shows LID is beneficial to the
environment and serves a long term purpose when considering stormwater management
(Woolson, 2013). Although research has evaluated LID controls individually and has proven that

10

LID controls are successful in reducing stormwater runoff volumes and improve water quality,
there are very few studies that evaluate large scale LID concepts that incorporate multiple LID
elements (Selbig & Bannerman, 2008).
Another constraint is the LID is less familiar and therefore communities may be hesitant
to adopt. Often communities adopt planning and design approaches, methods, and techniques
that are familiar and known. To engage in innovative approaches takes risk, education, and
additional energy.
2.2 Design Elements of LID
2.2.1 Rain Gardens
One type of low impact development feature is a rain garden. They provide effective
ways to collect and harvest rainwater as well as beautify an area (Grant & Giraud, 2015). Since
California has mostly dry seasons with minimal rain, it is important to collect as much rainwater
as possible. Although rain gardens are more popular in wetter climates they are efficient for
California’s Mediterranean climate (Grant & Giraud, 2015). Water from a site is channeled by
swales, curb openings, rain gutters, etc., and is diverted into the rain garden area where it soaks
into the ground and waters the vegetation (Grant & Giraud, 2015). Rain gardens that are
designed correctly hold water for a short period of time in order to retain water long enough for it
to percolate into the ground, which over time will maximize the amount of groundwater recharge
(Grant & Giraud, 2015). This is one solution to keep water on site instead of diverting it to
drains, rivers, and ponds. The goal of rain gardens is to allow water to slowly percolate into the
ground acting as a small bioretention pond. “Plants and soil microorganisms break down organic
compounds and remove pollutants such as phosphorus, nitrogen, and hydrocarbons” that are

11

collected from surfaces such as parking lots, roofs, driveways, and industrial sites (Grant &
Giraud, 2015).
Figure 1 demonstrates the fundamental parts of a constructed rain garden which includes,
native moisture tolerant plant materials, native soils, appropriate pond water depth, and a
perforated underdrain design (Southeast Michigan Council of Governments, 2008). Rain gardens
are suitable for large scale and small scale sites and are beneficial to both residential and
commercial design (Southeast Michigan Council of Governments, 2008). While being suitable
for large and small scale sites they are also suitable for a variety of soils ranging from sandy
loams to clay soils(Jaber, Woodson, LaChance, & Charriss, 2012). Common benefits of rain
gardens have been found between researchers and include “less stormwater runoff, slower runoff
rates, less pollution in the runoff, more water to replenish groundwater supplies, and improved
landscape” (Jaber et al., 2012).

Figure 1. Schematic Design of a Rain Garden

12

2.2.2 Green Roofs
Vegetated roofs, otherwise known as green roofs, are thin layers of vegetative materials
that allow for the rooftop to serve as a green space (Southeast Michigan Council of
Governments, 2008). There are three types of vegetated roofs that all contain different layers and
thickness of materials. The three types of variations are intensive, semi-intensive, and extensive
(Southeast Michigan Council of Governments, 2008). An intensive green roof “utilizes a large
variety of plants that include trees and shrubs”, which requires four inches or greater of layers
(Southeast Michigan Council of Governments, 2008). This type of green roofs requires a lot of
maintenance and are often found in park like settings (Southeast Michigan Council of
Governments, 2008). Extensive vegetated roofs have a limited variety of plant selection because
the substrate layer is shallow (Southeast Michigan Council of Governments, 2008). Unlike the
intensive variation, extensive green roofs require little maintenance once the are established. This
specific type is commonly used for an environmental benefit, such as storm runoff (Southeast
Michigan Council of Governments, 2008). All of these variations have layers that contain
“waterproofing, synthetic insulation, non-soil engineered growth media, fabrics, synthetic
components, and foliage” (Southeast Michigan Council of Governments, 2008). Green roofs can
be applied to many different scales of design including commercial, urban areas, industrial,
residential, and recreational (Southeast Michigan Council of Governments, 2008).
Figure 2 demonstrates the different layers that are associated with a vegetated roof. These
layers can be different thicknesses depending on the type of green roof. Each layer has a
different purpose. The root structure protects the roof construction from damaging any roots
(Southeast Michigan Council of Governments, 2008). The waterproof membrane protects the
structure from extra moisture(Southeast Michigan Council of Governments, 2008). Next is the
13

protective layer, which is a “specially designed perforation resistant protection mat” that stops
mechanical damage of the root barrier and roof construction during installations (Southeast
Michigan Council of Governments, 2008). The purpose of the drainage layer is to allow excess
run off into water outlets. Depending on the design of the green roof many drainage layers also
serve as a means of “water storage, enlargement of the root zone, space for aeration of the
system and protection for the layers below it” (Southeast Michigan Council of Governments,
2008). The filter layer separates the plant and substrate layers from the drainage layers in order
to collect “small particles, humic and organic materials” for the availability of the plants
(Southeast Michigan Council of Governments, 2008).

Figure 2. Schematic Design of a Green Roof
2.2.3 Permeable Paving
One type of infiltration technique is permeable paving, also known as pervious pavement.
Pervious pavement is a technique that uses stormwater infiltration, storage, and structural
pavement that has a permeable surface (Southeast Michigan Council of Governments, 2008) and
dates back to the 1960’s (EPA, Woodlands, Thelen, Howe, & Associates, 2017). The idea of
porous pavement was introduced in order to “promote percolation, reduce storm sewer loads,
reduce floods, raise water tables, and replenish aquifers” (EPA, Woodlands et al., 2017). Many
14

porous pavement sites have been constructed since then and have both succeeded and failed in
various climate situations (EPA, Woodlands et al., 2017). Most of the failures are due to the fact
that silt and other materials entered the site and clogged the pavement seams (EPA, Woodlands
et al., 2017). As time has passed many variations of pavers and pavement have been created.
According to Cahill Associates, who have implemented more than 200 porous asphalt pavement
sites since the 1980’s, have reported no failures because of proper design and construction
methods (EPA, Woodlands et al., 2017). Therefore it is important that when implementing low
impact development elements on a site, intense research must be performed in order to design for
successful site specific performance. Underneath these layers is a storage reservoir that collects
the run off from the above layers. Pervious pavement can be used for all variations of scale
including “parking lots, entire streets, walking paths, sidewalks, playgrounds, plazas, and
recreation courts” (Southeast Michigan Council of Governments, 2008). Different variations of
pervious paving are porous asphalt, pervious concrete, permeable paver blocks, and reinforced
turf/gravel (Southeast Michigan Council of Governments, 2008).
Figure 3 is an illustration of the layers that make up pervious pavement. The top layer,
pervious pavement, allows water to pass through and slowly infiltrates through the next couple
layers of large aggregate, coarse aggregate, and uncompacted subgrade (Southeast Michigan
Council of Governments, 2008). Depending on the site and scale different designs will call for
varying amounts of layers. As the water passes through these layers is it cleansed of some
pollutants and larger particles that cannot pass through the small openings in the pervious
pavement. Pervious pavement can be used for more than a best management practice (BMP), it
can also be used for safety reasons. Because pervious pavements allow water to pass through, it
decreases the chances of hydroplaning (Southeast Michigan Council of Governments, 2008).
15

Porous asphalt is a “standard bituminous asphalt where the fines have been screened and
reduces, allowing water to pass through small voids” (Southeast Michigan Council of
Governments, 2008). Porous asphalt can be used in any location that is appropriate to use
standard asphalt. It is typically poured directly on the gravel subbase with a thickness of 2.5
inches (Southeast Michigan Council of Governments, 2008). Pervious concrete is another
variation of pervious pavement. It is very similar to porous asphalt in that fact that it is also
created by reducing the number of fines in a mix in order to establish voids (Southeast Michigan
Council of Governments, 2008). Porous concrete is much more ridged in appearance compared
to traditional concrete due to the larger fines which makes this easy for water to pass through
(Southeast Michigan Council of Governments, 2008). According to the LID Manual for
Michigan, porous concrete has proven to be an effective stormwater management technique if it
is installed correctly. Permeable paver blocks are made up of “interlocking units that provide
some portion of surface area that may be filled with a pervious material such as gravel”
(Southeast Michigan Council of Governments, 2008). Most permeable paver blocks are often
concrete and serve as an aesthetic element. They can be used in many different type of designs,
but they are most popular in “plazas, patios, parking areas, and low-speed streets” (Southeast
Michigan Council of Governments, 2008). Reinforced gravel/turf is another variation of pervious
pavement types and also consists of an interlocking structural unit that has voids for turf grass to
grow (Southeast Michigan Council of Governments, 2008). This makes it suitable for traffic and
parking (Southeast Michigan Council of Governments, 2008). The reinforced turf includes
concrete or plastic underlain by a gravel layer serving as a drainage system (Southeast Michigan
Council of Governments, 2008). These different types of permeable pavings are great options
when considering stormwater management solutions, but there are common concerns within the

16

previous research. These concerns include clogging and failure, cost prohibitive, maintenance
costs are high, functionality in cold climates, and adequate stability and structure for truck traffic
and heavy loading (“Eisenberg, Bethany, Lindow, Kelly Collins, and Smith, David R.,” 2017).

Figure 3. Schematic Design of Permeable Paving
2.2.4 Infiltration Basin
An infiltration basin is a vegetated depression where stormwater runoff is stored until it
slowly infiltrated into the soil (RWRA, n.d.). The purpose of an infiltration basin is primarily to
enhance water quality by removing pollutants, but it can also serve as a means for managing
flooding and channel erosion control (RWRA, n.d.). This specific LID element applies mostly to
sites that contain soils with a reasonable infiltration rate and the water table is fairly low in order
to prevent pollution of groundwater (RWRA, n.d.). Unlike a conventional piping system that
requires an outflow, LID infiltration basins do not require this component due to the fact that
outflow is through the surrounding soil (New Jersey Stormwater Manual, 2004). Ideal sites to
incorporate the use of infiltration basins are usually medium-density residential or commercial
sites that contain an impervious cover of 36%-66% (RWRA, n.d.).

17

A study done by the Wisconsin Department of Natural Resources (WDNR) and the U.S.
Geological Survey (USGS) compared two infiltration basins to determine whether using LID
techniques or conventional stormwater systems was more productive when aiming to reduce
runoff volumes and improve water quality (Selbig & Bannerman, 2008). Comparisons of
pollutant loads was analyzed between both infiltration basin designs in order to evaluate the
benefits of low impact design (Selbig & Bannerman, 2008). This study along with many others
indicated that developments adopting the low impact development approach produce less runoff
than conventional sites (Brander, Owen, & Potter, 2004).
Figure 4 demonstrates a typical design for an LID infiltration basin. When designing an
infiltration basin consideration should be given to the soil characteristics, depth to the
groundwater table, sensitivity to the region, and runoff water quality (New Jersey Stormwater
Manual, 2004). It is important that the soils are permeable and not compacted during the design
phase because this will alter the infiltration rate and cause contamination, clogging, or flooding
(New Jersey Stormwater Manual, 2004). Basins should only occur where the surrounding slopes
are less than ten percent, and the basin floor needs to be as level as possible for “uniform
spreading” of stormwater runoff (New Jersey Stormwater Manual, 2004).

18

Figure 4. Schematic Design of an Infiltration Basin
2.2.5 Constructed Wetlands
Constructed wetlands are manmade wetlands that simulate the natural processes in order
to properly treat stormwater, wastewater, agricultural waste water, and improve water quality of
point and non-point sources of water pollution (Davis, n.d.). Much like the other LID elements,
constructed wetlands have layers. The layers consist of wetland vegetation, soils, and “microbial
assemblages” that improve water quality (USEPA, 2000). Wetlands are most commonly used in
the process of revitalizing ecosystems and as water treatment systems (Davis, n.d.). Although
constructed wetlands can be used at a variety of scales, they are most successful when designed
for a larger site. A benefit of this element of LID development is that constructed wetlands can
have an infinite lifetime (Davis, n.d.). Depending on the wastewater loadings and the capacity of
the wetland to remove waste, a wetland can operate for over 20 years with little to no loss in
effectiveness (Davis, n.d.).
Figure 5 demonstrates how a wetland is constructed and its different layers. Typically a
constructed wetland requires an inlet device and an outlet device. The use of marsh plants and

19

water tolerant vegetation is required so that they can use the nutrients and minerals that are
cleansed from the water that is collected resulting in higher water quality (Davis, n.d.).

Figure 5. Schematic Design of a Constructed Wetland
2.3 Recent LID Studies
A recent study conducted in May 2007 in Waterford, CT analyzed the stormwater runoff
and pollutants of a traditional subdivision design compared to a low impact subdivision design.
The traditional design includes 17 lots that were built using current codes, regulations, and
construction practices (Dietz & Clausen, 2008). The traditional design included a curb and gutter
system with a 8.5 meter asphalt road, landscaping, and turf that were similar to new subdivisions
in the area (Dietz & Clausen, 2008). The roof runoff was directed toward lawn areas and
driveways, and the total impervious surface of the subdivision after construction was 32% (Dietz
& Clausen, 2008). The low impact subdivision included 12 lots and incorporated several
“pollution prevention measures” (Dietz & Clausen, 2008). This site replaced the 8.5 meter
asphalt road and curb and gutter with a 6.1 meter Ecostone paver road and grassed swales (Dietz
& Clausen, 2008). Other LID controls that were integrated into the design include bioretention,
rain gardens, permeable pavers for driveways, and houses were constructed in clusters which
reduced lawn size and maintenance (Dietz & Clausen, 2008). The total impervious surface after
construction was completed was 21%, significantly lower then in the traditional design. The
20

results of this study show that a large increase in runoff volume was due to the increase in
impervious surfaces in the traditional subdivision design (Dietz & Clausen, 2008). It also
demonstrated that in the LID subdivision the annual stormwater runoff volume did not change as
the impervious surfaces increased (Dietz & Clausen, 2008). There is not a change in stormwater
runoff volumes due to the low impact controls used throughout the subdivision design.
An additional case study on a project in Houston, Texas resulted in the development of an
apartment complex that would have been canceled if it were not for the use of low impact
development strategies. Queenston Manor Apartments sit on a 7.2 acre site in Houston,Texas.
The developer originally planned the apartment complex on this site, but the economic model
required that nine apartment buildings be constructed in order to generate the appropriate
revenue (Convergent Water Technologies, 2014). Originally all of the sites detention was
accounted for offsite in surrounding developments, but the county determined that it was no
longer available for use and the project came to a halt. Engineers at EHRA saved the project by
incorporating LID as well as a FocalPoint High Performance Modular Biofiltration System
(HPMBS) (Convergent Water Technologies, 2014). A HPMBS is a “combination of a high
performance, open cell underdrain, a clog-proof bridging mesh, bridging stone, and a high
performance biofiltration media that flows at a rate of over 100” per hour” (Convergent Water
Technologies, 2014). This system allowed for all of the complex’s common areas and courtyards
to serve as drainage areas. The objective of the LID design was to decrease peak flow, which in
time would decrease the total detention volume (Convergent Water Technologies, 2014). By
introducing the LID approach to design, the developer was able to proceed with the project due
to the fact that the area dedicated to the detention facility in the traditional design was
eliminated. This allowed for an increase in area for more apartment buildings, which permitted
21

the developer to meet the revenue requirements (Convergent Water Technologies, 2014).
Designers at EHRA redesigned the project to include “porous pavers in parking stalls, a directly
infiltrated underground detention system, vegetated swales and vegetated depressions which
drain through a series of small FocalPoint HPMBS” (Convergent Water Technologies, 2014).
This resulted in a decrease in surface storage, which allowed space for two additional apartment
buildings. This was equivalent to 48 apartment units (Convergent Water Technologies, 2014).
The success of a combination of LID controls allowed for this project to be implemented.
The Ipswich River Watershed developed three low impact development case studies as
part of a demonstration project “designed to showcase practices that can help improve low-flow
and water quality conditions” (The Massachusetts Department of Conservation and Recreation &
Ipswich River Watershed Association, 2009). The case studies asses, quantify, and demonstrate
the benefits of LID controls.
The first case study was the Partridgeberry Place, a residential development in Ipswich,
Massachusetts. Twenty houses were clustered together on 0.2 acre lots surrounded by woodland
areas. The LID controls that were included in the design were significantly smaller setbacks to
property lines, a common septic tank, minimal pavement due to clustering, narrow roadways,
short driveways, rooftop stormwater drains which infiltrate directly into the ground, rain gardens,
grass pavers, grass swales, and native vegetation (The Massachusetts Department of
Conservation and Recreation & Ipswich River Watershed Association, 2009). A computer
program was developed in order to process site measurements such as runoff patterns, rate of
runoff, and average rainfall (The Massachusetts Department of Conservation and Recreation &
Ipswich River Watershed Association, 2009). The computer program also produced results that
were compared to four design conditions. The four design conditions were the pre-developed
22

condition, the LID subdivision, a cluster only subdivision, and a traditional subdivision design
(The Massachusetts Department of Conservation and Recreation & Ipswich River Watershed
Association, 2009). By producing runoff patterns for each site and comparing them, the study
“characterized how effective the LID features” were at reducing runoff compared to traditional
design. The results showed that for the pre-developed site the peak runoff rates were the lowest
for all storm sizes (The Massachusetts Department of Conservation and Recreation & Ipswich
River Watershed Association, 2009). The LID subdivision and the cluster only design yielded
similar results in that the peak flow runoff rates were slightly higher than the pre-developed site.
The results also showed that the traditional subdivision design produced significantly more
runoff and higher peak flow rates compared to all other design scenarios (The Massachusetts
Department of Conservation and Recreation & Ipswich River Watershed Association, 2009).
Another case study related to theIpswich Watershed is the Silver Lake Beach LID
Retrofit. It is a 28 acre pond in Wilmington, MA, but is frequently closed due to high amounts of
E.Coli bacteria from polluted stormwater runoff (The Massachusetts Department of
Conservation and Recreation & Ipswich River Watershed Association, 2009). LID controls that
were implemented in this case study include planted swales, permeable pavement, and
bioretention cells. Sampling wells were placed in the parking lot to evaluate the concentrations
of chemicals in the water that is collected from asphalt runoff (The Massachusetts Department of
Conservation and Recreation & Ipswich River Watershed Association, 2009). Samples were
collected for five months prior to the start of construction and for one year after construction was
completed. The results show that overall the combination of “LID retrofits” helped to reduce the
number of beach closures in the swimming area (The Massachusetts Department of Conservation
and Recreation & Ipswich River Watershed Association, 2009). The study also concluded that
23

the four different types of permeable paving allowed for infiltration at a rate ranging from 49
inches per hour to 10,000 inches per hour (The Massachusetts Department of Conservation and
Recreation & Ipswich River Watershed Association, 2009). By testing different permeable
pavers the case study proved that permeability is a successful element of LID design, but also is
successful in reducing the amount of pollutants in stormwater runoff.
The third case study that is affiliated with the Ipswich Watershed is the Silver Lake
Neighborhood LID Retrofit. This case study entails a three acre residential neighborhood that
borders the lake. The goal was to determine successful LID elements that can decrease the
amount of runoff from rooftops, driveways, and streets within the neighborhood (The
Massachusetts Department of Conservation and Recreation & Ipswich River Watershed
Association, 2009). Twelve rain gardens and two areas of permeable pavers were used in the
front of homes in the public right of way along both sides of the street. Runoff from rooftops,
driveways, and roads was redirected into the rain gardens and permeable pavers to allow for
infiltration (The Massachusetts Department of Conservation and Recreation & Ipswich River
Watershed Association, 2009). An important aspect of this case study is the importance of
educating the homeowners about LID, the rain gardens, and the permeable pavers. Community
members were informed of the study and educated on how to maintain and protect the low
impact development features. Rain gauges were installed to measure and monitor the runoff
volumes from the neighborhood to the lake (The Massachusetts Department of Conservation and
Recreation & Ipswich River Watershed Association, 2009). In order to measure concentrations
of pollutants such as phosphorus, nitrogen, metals, petroleum hydrocarbons, and bacteria,
additional equipment was used to capture samples from storm drains (The Massachusetts
Department of Conservation and Recreation & Ipswich River Watershed Association, 2009). The
24

results concluded that the combination of rain gardens, permeable pavers, and vegetated swales
reduced the volume of stormwater runoff that flows directly into the lake. The results also state
that the sample size was too small in order to determine if the use of rain gardens and permeable
pavers reduces the amount of pollutants in stormwater runoff. It is suggested that the conclusion
is based on a limited data set and needs to be further investigated (The Massachusetts
Department of Conservation and Recreation & Ipswich River Watershed Association, 2009).
It is clear that recent research and case studies prove that low impact development is a
successful way to decrease runoff rates and increase infiltration volumes. It is unclear if LID is a
good alternative for reducing the amount of pollutants in stormwater runoff. Whether or not a
combination of LID elements is more successful than the use of a single LID control is unclear
and will require further research.
2.3.1 MSU Low Impact Investigations
In recent years graduates at Michigan State University (MSU) have been highly
interested in the topic of low impact development. Their studies have inspired me to further
investigate the topic of LID and its positive effects on the environment. In 2016 a similar study
was completed by Hongwei Tian on a site in Grand Rapids, Michigan. The study investigates the
“LID performance by examining four different design scenarios”, two designs include LID
controls and two designs lack LID controls (Tian, 2016). The four different sites include the
existing site, LID with Cloud Design, LID Design, and a traditional design. The design scenarios
were evaluated using eleven variables that were selected in the areas of energy use, climate
change, stormwater management and ecosystems (Tian, 2016). The eleven variables include
impervious surfaces, permeable pavement, green space, average tree water consumption, total
shadow area, number of trees, runoff, soil infiltration, evaporation, field sparrow habitat
25

sustainability index, and fox squirrel habitat sustainability index (Tian, 2016). These eleven
parameters were analyzed using the Friedman’s One Way of Variance Test in order to compare
the design scenarios using rankings. After analyzing these parameters, the results show that the
designs with LID elements are statistically better than the existing design, but it cannot be
determined that the design scenarios with LID elements are better nor worse than a traditional
design.
Students at MSU also completed a related study, Metrics in Master Planning Low Impact
Development for Grand Rapids, Michigan, that investigates the metrics that demonstrate the
effects that low impact development has on urban sustainability issues (Burley, Li, Ying, Tian, &
Troost, n.d.). The metrics include “reduction in stormwater volume, increase in stormwater
quality, increase in songbird habitat sustainability, increase in vegetation biodiversity, reduction
in water requirements by woody vegetation, increase in latent soil productivity, increase in
vegetation adaptation to climate change, increase in visual quality, improvement in microclimate
diversity, reduction in landscape maintenance and energy inputs, and walkability” (Burley et al.,
n.d.). LID controls that are incorporated into the design include green roofs, rain tanks and
cisterns, permeable pavement, bioretention and rain gardens, dry and wet swales, and constructed
wetlands (Burley et al., n.d.). The goal of the proposed design was to improve stormwater
management treatments. A team of students, professors, and varying professionals designed a
combination of LID controls and assessed the stormwater runoff, infiltration, and
evapotranspiration rates using the U.S. EPA National Stormwater Calculator to document the
change volumes (Burley et al., n.d.). For this study, climate change was also examined by
measuring variables such as trees, shade, and land use changes. In order to evaluate tree water
consumption the Simplified Landscape Irrigation Demand Estimation (SLIDE) was used (Burley
26

et al., n.d.). This method calculates the water demand for water-conserving irrigation plans based
on plant species (Burley et al., n.d.). After analyzing all of the parameters, stormwater, tree water
consumption, change in land cover, habitat sustainability, visual quality, and soil productivity,
the results show that the master plan developed by the MSU team members is significantly better
than traditional approaches and significantly better than the existing site (Burley et al., n.d.).
Another study completed by a student at MSU concerns low impact housing in River
Rouge, Michigan. This research investigates the positive and negative effects of using a
landscape based approach to design versus an architectural based design approach. The
landscape based design incorporates mixed-use areas, residential areas, commercial areas,
stormwater treatment strategies, open spaces, closed wetlands, and open wetlands (Wang et al.,
2017). This approach “orchestrates the structure of the environment based upon the composition
of the landscape and then the needs of the greater environment” (Wang et al., 2017). The
buildings and circulation are then laid over in “designated zones determined by the organization
of the landscape plan/design” (Wang et al., 2017). This means that the design of the sites comes
first, where as in architectural based design, the landscape comes last in the design process. An
architectural design includes the placement of structures, and then incorporating circulation
patterns and landscape in “leftover spots” (Wang et al., 2017). In this study a visual quality test is
used to examine and compare treatments to measure environmental visual quality and
stormwater runoff quality. Different LID controls and combinations of LID controls were
assessed for each scenario and include bioswales, bioretention, open wetlands, and a
combination of bioretention and constructed wetlands. After evaluating the parameters the results
“indicate that the landscape based housing development has significantly better
visual/environmental quality and that a bioretention water treatment area combined with a
27

constructed wetland” has a positive effect on water quality by removing approximately 96.3% of
phosphorus in the onsite runoff (Wang et al., 2017).
2.4 LID Summary
Although communities, developers, and designers are considering the use of LID
elements, there is limited research concerning the effectiveness of low impact development
versus traditional design, including research on the effects of integrated LID techniques. This
research evaluates designs with LID elements to determine if there is a slower rate of stormwater
runoff and lower energy use compared to a traditional design. Furthermore, this research assesses
the combined use of a series of low impact development elements throughout an entire site.
Previous researchers have focused on the evaluation of one specific LID technique such as just
green roofs, or just rain gardens. Although previous studies have evaluated a single LID element
at different scales, research on the cumulative benefits of multiple LID elements for a single site
is needed. More research is needed on the evaluation of a combination of LID techniques on
revitalized environments.
In summation this research seeks to explore the positive effects of low impact
development techniques on stormwater runoff rates for a specific site using EPA measures and
NCRS (Natural Resources Conservation Service) soil data to compare three design scenarios.

28

Chapter 3: Methods
This research will analyze the use of a combination of low impact development elements
for three different design scenarios. Each design will incorporate a different combination and
size of LID elements. After calculating the six different variables (average annual runoff, days
per year with runoff, percent of wet days retained, smallest rainfall with runoff, largest rainfall
without runoff, and maximum rainfall retained) and comparing the scenarios against different
LID combinations, the Friedman One-Way Analysis of Variance Test will be applied in order to
compare the design scenarios, and determine if their differences are statistically significant.
3.1 Study Site
The study site is approximately 44 acres and is adjacent to the Los Angeles River (See
Figure 6). The site’s soil type is D, which represents high runoff potential. This soil type is
classified as urban land or commercial land (NRCS, 2017). The hydraulic conductivity of the soil
is approximately 0.01 inches per hour, which implies that the soil has a high amount of runoff.
Soils with lower conductivity produce more runoff (USEPA, 2014). The site’s topography is
particularly flat and has an average slope of approximately 5%. The site’s precipitation and
evaporations data are collected by the Downtown Los Angeles and the University of Southern
California rain gage and weather station. The climate change scenario is classified as “no
change”. The land cover and LID controls vary based on the design scenario being analyzed. The
wet dry event threshold is 0.10 inches on site. “The Center for Watershed Protection
recommends using a runoff threshold of 0.10 inches because impervious areas of the watershed
are assumed to generate runoff beginning at approximately 0.10 inches of rainfall” (MDEQ,
2006).

29

Figure 6. Aerial Image of Existing Site (adapted from Google Earth)
3.2 Design Scenarios Description
This research will analyze and estimate the stormwater management runoff of three
design scenarios for the study site including the existing site (design scenario 1), extended LID
design (design scenario 2), and the preliminary design in the LA River Master Plan (August
2014) developed by Mia Lehrer and Associates (design scenario 3). Design scenario 1 is an
abandoned rail yard along the Los Angeles River and is in the process of being revitalized.
Design scenario 2 is a master plan design created at Michigan State University (MSU) in a
graduate level course that integrates extensive LID elements specific to this research. Design
scenario 3 was completed by Mia Lehrer and Associates, a firm hired by the city of Los Angeles
to design parcel G2 of the Taylor Yard Site. Each site is thoroughly described below.
3.2.1 Existing Site (Design Scenario 1)
The existing site (Figure 6 and Figure 7) is an abandoned rail yard located in Los
Angeles, California alongside the Los Angeles River and occupies approximately 247 acres.
30

According to “The River Project”, it is the largest undeveloped parcel along the river, which
presents “extraordinary opportunities” for river revitalization plans (The River Project, 2011).
Within the existing site there are seven different parcels. This study will focus on parcel G2.

Figure 7. Map of Taylor Yard Parcels (adapted from The River Project, 2011.)
“Parcel G2 is currently owned by Union Pacific and is the operating facility for the
existing but abandoned rail lines. It is a contaminated site due to the industrial uses. Parcel G2
contains the majority of the rail lines along the Los Angeles River. This parcel is under a
feasibility study in order to incorporate future habitat restoration, water quality remediation,
flood mitigation, wetlands restoration, and recreation uses. This parcel is approximately 44
acres” (The River Project, 2011. pg 1).
The existing site has been used for various maintenance, fueling, and industrial
operations which began in the 1930’s and continued until 2006 (Environmental Management
Group, 2014). These operations include diesel shops, machine shops, a roundhouse, two
turntables, underground and above ground storage tanks, a service track, and miscellaneous
buildings (Environmental Management Group, 2014). Once the site was permanently closed in
31

2006, all above ground structures that remained on the site except for various concrete slabs,
footings, and foundations were demolished (Environmental Management Group, 2014). In
previous years many soil and groundwater investigations have taken place on site and identified
many chemicals in the soil including petroleum hydrocarbon, arsenic and lead (Environmental
Management Group, 2014). Since this site has been identified as a “crown jewel” of Los
Angeles, it is in the process of being rezoned for uses other than industrial, and the site continues
to undergo remediation and evaluation (Environmental Management Group, 2014).
3.2.2 Extended LID Design (Design Scenario 2)
Design scenario 2 (Figure 8) focuses on the revitalization of the Los Angeles river front
in parcel G2. In parcel G2 the design includes many LID elements including rain gardens,
bioswales, constructed wetlands, infiltration basins, and permeable paving. The city of Los
Angeles is eager to develop parcel G2 in Taylor Yard because of its placement on the river.
Design Scenario 2 studies the option of narrowing the river due to the low average amount of
water that is usually in the river. Because of the existing concrete ditch that was originally
designed to handle flooding, it is nearly impossible for the river to help clean and filter
stormwater runoff. The U.S. Army Corps of Engineers has performed an Integrated Feasibility
Report (IFR) for the Los Angeles River Ecosystem Restoration Study, which aims to restore
natural riparian ecosystem values amongst an 11 mile portion of the river (Environmental
Management Group, 2014). The purpose of the IFR is to transform the space into a “20th century
single purpose river channel into a 21st century multi-purpose infrastructure that incorporates the
natural environment, public use, recreation, and flood control” solutions (Environmental
Management Group, 2014). This design incorporates significant green space for activities,
connections to the nearby state park, walkable systems for the surrounding communities, and
32

connections across the river. All of these elements contribute to the concept of a “21st century
multi-purpose” public space with a focus on environmental quality.

Figure 8. Conceptual Rendering of Extended LID Design
3.2.3 Mia Lehrer and Associates Design (Design Scenario 3)
Design scenario 3 (Figure 9) was designed by Mia Lehrer and Associates after being
hired by the City of Los Angeles. Their design is strictly conceptual and includes site
remediation, water quality improvements, abundant parkland and open space, river amenities,
community gateways, and integration of habitat elements (Environmental Management Group,
2014).

33

Figure 9. Conceptual Plan Designed by Mia Lehrer and Associates
3.3 Measurement Elements- Stormwater Management
For these three scenarios stormwater runoff, infiltration, and evaporation rates will be
analyzed by using the National Stormwater Calculator (SWC). The SWC is a software tool that
evaluates runoff, infiltration, and evaporation for small scale sites in the United States (USEPA,
2014). The goal of SWC is to estimate the amount of stormwater runoff generated from a
specific location under different development and control scenarios (USEPA, 2014). The
calculations are based on historical findings of average rainfall, soil conditions, soil drainage
rates, topography, different types of land cover, and different LID controls (USEPA, 2014). Part
of the calculation process allows you to incorporate different LID elements into each design and
see how they affect the outcome of the sites stormwater runoff rates. The SWC’s “computational
engine” is run by the EPA Stormwater Management Model (SWMM), which is highly
recommended for hydrology modeling and analysis (USEPA, 2014)The SWC focuses on
34

informing professionals in the development industry on how they can meet preferred stormwater
targets and can be used to assist in answering the following questions (USEPA, 2014):
1. What is the largest daily rainfall amount that can be captured by a site in either its
pre-development, current, or post-development condition?
2. To what degree will storms of different magnitudes be captured on site?
3. What mix of LID controls can be deployed to meet a given stormwater retention
target?
4. How well will LID controls perform under future meteorological projections made by
global climate change models?
Data on 1) site location, 2) the site’s soil type, 3) the site’s soil drainage rate, 4) the site’s
topography characteristics, 5) hourly rainfall data from a nearby rain gage, 6) evaporation rate
from a nearby weather station, 7) climate change scenario selection, 8) the site’s land cover for
each design scenario, and 9) LID control measures for each design scenario is input into the
SWC model to predict accurate runoff rates, infiltration rates, and evaporation rates. The
calculator results include average annual rainfall, average annual runoff, days per year with
rainfall, days per year with runoff, percent of wet days retained, smallest rainfall with runoff,
largest rainfall without runoff, maximum rainfall retained, and a pie chart representing
percentage of runoff, infiltration, and evaporation rates for each design scenario.
3.4 Statistical Method to Test the Hypotheses
The null hypothesis for this research states that if the SWC is applied to each design
scenario, then there will not be a significant difference is the amount of runoff for each design
scenario. Therefore LID elements have no positive effect on stormwater management practices.
The research/alternative hypothesis states that if the SWC is applied to each design scenario,
35

then at least one scenario will show that LID elements have a positive effect on stormwater
management practices because the runoff amount will be lower than at least one other scenario.
For this research, the Friedman One Way of Variance Test will be applied in order to
determine which design is most successful when considering LID elements as a feature to control
stormwater management on site. The Friedman One Way of Variance Test is a non-parametric
statistical test used to evaluate the treatments’ values based on ranks (Daniel, 1978.). Friedman’s
Test is used “between groups when the dependent variable being measure is ordinal” (Statistics,
2013).
The first step in completing the Friedman’s Test is to convert the results into rankings. To
do this, the observations in each block (b) are ranked separately so that each block contains a
separate set of ranks (k). A block is one of the eight statistical results adapted from the SWC.
These are the eight computational components that the SWC computes once the data is inserted.
Each design was ranked by six blocks and three different treatments.
The second step is to determine the null hypothesis and the research hypothesis. For this
research, the null hypothesis (H0) is that all of the design scenarios have identical effects, and the
research hypothesis (H1) is at least one design scenario has a larger value than at least one other
design scenario.
The third step is to obtain the sums of the ranks (R) in each column. If all treatments have
identical effects, than we would expect the sums to be fairly similar in size. When there is one
sum that is sufficiently different from the others the null hypothesis can be rejected.
The fourth step is to find the computational chi-squared value.

36

Equation 1: Computational Chi-squared Value
𝑘

12
𝑥𝑟2 =
∑ 𝑅𝑗2 − 3𝑏(𝑘 + 1)
𝑏𝑘(𝑘 + 1)
𝑗=1

Where:


b is the number of blocks



k is the number of treatments



R is the sum of ranks for reach treatment

The fifth step is to determine the level of risk, or alpha (Îą). In this research Îą is equal to
0.005. Alpha is the percent chance that the null hypothesis is correct. Since alpha is equal to
0.005, this means that there is a 99.5% chance that the research hypothesis is true. Using the
book, Applied Nonparametric Statistics, Daniel provides a table (Daniel, 1978, p.452) that
2
contains the chi-square values of 𝑥(1−α)
with k-1 degrees of freedom (Daniel, 1978). If 𝑥𝑟2 is
2
greater than or equal to the tabulated value of 𝑥(1−α)
with 2 degrees of freedom, then the null

hypothesis will be rejected (Daniel, 1978).
The sixth step is to determine which design scenarios are better than others using the
multiple comparison test that is associated with Friedman’s test. When the Friedman’s test is
applied to research and leads us to reject the null hypothesis, we are curious as to where the
differences are relevant (Daniel, 1978). The multiple comparison procedure will determine
where the differences in the research are apparent.

37

Equation 2: Multiple Comparison Test
𝑏𝑘(𝑘 + 1)
|𝑅𝑗 − 𝑅𝑗′ | ≥ 𝑧√
6
Where:


Rj and Rj’ are two sums of the different treatments’ ranks



z is the tabulated value provided by a specific table in Daniel’s book (Daniel,
1978, p.397) and corresponding to Îą/k(k-1)

38

Chapter 4: Results
In order to test the research hypothesis, that is the SWC is applied to each design
scenario, then at least one scenario will show that LID elements have a positive effect on
stormwater management practices by decreasing the amount of runoff on site, measurements on
site parameters were recorded. Table 1 shows all of the data input for the stormwater calculator
for each design scenario.

Table 1. The National Stormwater Calculator (SWC) Parameters
Table 2 indicates the results of eight different variables for the three design scenarios,
where average annual rainfall and days per year with rainfall are constant. The sites will be
compared based on the six remaining variables. According to Table 2 and the statistical results
computed by the SWC, the average annual runoff for the three design scenarios is 12.86, 7.07,
and 8.92 inches. The days per year with runoff for each design are 18.43, 3.92, and 5.64 days.

39

The percent of wet days retained is 14.89, 69.66, and 56.48 days. The smallest rainfall with
runoff for each design is .1, .49, and .4 inches. The largest rainfall without runoff is .22, .88,
and.44 inches. And the maximum rainfall retained is .4, 1.49, and .87 inches.

Table 2. SWC Statistic Results
Table 2 is a representation of each design scenarios results and their percent of runoff,
infiltration, and evaporation for an annual average rainfall.
When comparing the existing site to design scenario 2 and 3, one can see in Figure 10
that the percent of runoff was reduced by 38% and 26%, while infiltration increased by 33% and
25%.

Figure 10. SWC Runoff, Infiltration, and Evaporation Analysis Results (adapted from SWC)

40

In Table 3, the design scenarios have been ranked from 1-3 where 1 means the most
effective stormwater management design and 3 meaning the least effective stormwater
management design.

Table 3. Friedman’s Test Results
The sum for design scenario 1 is R=18, design scenario 2 where R=6, and design scenario
3 where R=12. We then need to find r2. For design scenario 1 where R2=324, for design scenario
2 where R2=36, and for design scenario 3 where R2=144.
Given the ranks calculated, the number of blocks being 6 and the three design treatments,
the chi-squared value was determined to be 12.
Therefore:
Equation 3: Results of Computational Chi-squared Value
𝑥𝑟2 =

12
(182 ∗ 62 ∗ 122 ) − 3(6)(3 + 1) = 12
(6(3)(3 + 1)

The computational chi-squared is 12.
2
The value of 𝑥0.995
with 2 degrees of freedom is 10.597, which is smaller than the

computational chi-square of 12. Therefore, in this research, there is at least one design that is
statistically better than at least one other design scenario.
Using the table from Daniel’s book (Daniel, 1978, p397), and knowing that α is equal to
0.005 and k is equal to 3, we can calculate that z is equal to 2.4.
41

Therefore:
Equation 4: Results of Multiple Comparison Test
18(3+1)

2.4 √

6

= 8.313

The rank totals were Ra (Design scenario 1-Design scenario 2) = 12, Rb (Design scenario 2 –
design scenario 3) = 6, Rc (Design scenario 3 – Design scenario 1) = 6. Thus we can conclude
that design scenario 2 is better than design scenario 1, but neither better nor worse than design
scenario 3. We can also conclude that design scenario 3 is better than design scenario 1, but
neither better nor worse than design scenario 2.

42

Chapter 5: Discussion
Before I began my research involving low impact development, I was unsure of its
benefits, but always believed there was an environmentally friendly opportunity for designers to
incorporate stormwater management techniques into urban development. Although this research
investigates and analyzes the use of LID elements, it opens up a platform for further research.
This thesis also inspires designers, planners, and environmentalists to pose questions about the
rules and regulations of design in the 21st century. Should professionals be required to use tools
such as the National EPA Stormwater Calculator in order to prove that a plan is designed to the
best of their ability while keeping the health of the people and the environment in mind? After
completing this research I believe that this is one of the most important questions to pose.
The calculations indicate that design scenario 2, which includes green space and a
combination of LID controls, can successfully reduce runoff volume by retaining rainfall on site.
Design scenario 3 is similar because of its large amount of green space, but lacks LID variables.
Although you can see the positive impact of LID elements in the quantifiable research, it is
important to keep other aspects of design in mind that were not measured using statistical
analysis. Aspects such as the value of aesthetics and educational components are more difficult
to measure, but are extremely important to successful developments that encourage and teach
people about the environment. They are important because there is a lack of knowledge of
environmental issues. Using educational components is a way to inform people of LID benefits
and constraints. By using simple diagrams, such as the diagrams previously shown in the
literature review, people can be educated on the process of each LID element and how each one
works. Once people understand the process, benefits, and constraints, they can use these
elements in an aesthetic manner. Therefore, the LID elements are beneficial, as well as providing
43

aesthetic appeal for visitors. On site educational components are a great way for landscape
architects, planners, engineers, and designers to influence visitors and share their knowledge of
ways to be environmentally friendly. Given the rate of urbanization along with increase in
impervious surfaces sharing this knowledge from community to community is important. It is
also a way for visitors to inform themselves of the history of a site. This history is important to
professional designers because it influences the way that they approach their designs. For
example, a designer would approach a site that was previously a forested area differently than
they would approach a site that was once an industrial park. Investigating all parameters of a site
will ensure that the design resolves the sites existing issues. The history of the site is also
important to the community. It is a way for community members to learn about the past and to
incorporate history into future designs through education. Education can be taught on site in
many different ways. Some ways include educational signage, historical landmarks, sculptures,
and interactive educational components. All of these factors are just as important as the
quantifiable research and raises many questions as to what factors designers should be focus on.
This research analyzes the average rainfall on site as well as the average amount of
runoff. Even though the average of these parameters shows that Design Scenario 2 and Design
Scenario 3 are similar when considering stormwater runoff, infiltration, and evaporation
amounts, the research does not show what would happen in a large storm event. In a large rain
storm it may be more obvious that LID elements are required to design for successful on site
stormwater management, but further research is required in order to determine those results.
There are many factors that need to be analyzed in order to determine if LID is an appropriate
solution for site specific developments. These factors include soil conditions, slopes, elevations,
and climate. These factors are important to consider because they influence the approach that
44

designers have in solving the sites issues. These factors are also important because they can
impact a designer’s decision of which LID elements, if any, are appropriate on site and how they
should be implemented. It is possible that future research may show that LID controls are site
specific solutions and they are not a good alternative to stormwater management for every single
design.
As previously mentioned there are many benefits and constraints to the use of LID
elements. Throughout this research I have documented what I think the most important barriers
in the research of LID are and that need to be further explored. This includes government
limitations, code limitations/regulations, cost, preconceived notions, interaction between
disciplines, and most important unfamiliarity. Code regulations are one reason that it is difficult
to require use of LID elements in site design. This is because the process of submitting permits
can be tedious for professionals, so professionals tend to avoid this process. Cost and
preconceived notions are a reason that communities and clients have a hard time accepting the
use of LID elements. As previously mentioned, LID elements can be expensive, but it is difficult
to show or convince people that the expense in the beginning will benefit over a longer period of
time. People have difficulty seeing the benefits when there is a lack of results from the
beginning. Interaction between disciplines is critical in order to design properly. Different
disciplines needs to work together to influence communities and government agencies that LID
is in fact a productive feature for the environment. Unfamiliarity is a lack of education within
communities and government agencies. It is possible that some professionals also lack the
knowledge of LID results and influences. By requiring professionals to attend conferences and
performing continuing educational practices, professionals will be more knowledgeable of LID
and their benefits and constraints. When engineers, landscape architects, and planners work
45

together the result is often more efficient. Rethinking the standards of stormwater management
includes cooperation from all participating disciplines and an understanding of each disciplines
roll. It also includes an investigation of regulatory standards of stormwater management systems
and designs. The use of LID elements and proof of their success is going to require people to
look past the traditional standards and be open to alternative options such as LID. Successful
design requires a mixture of many different approaches, not just the approach that only meets the
require standard. That is failure to design with the health, safety, and welfare of the community
in mind.
It is obvious that the sites with larger amounts of green space reduce the amount of runoff
and increases infiltration rates. One can also see that the site with LID controls reduces the
amount of runoff even more, as well as increasing the amount of infiltration. The difference
between the results of design scenario 2 and design scenario 3 indicates that a combination of
LID controls and an abundance of green space are crucial to reducing runoff rates, increasing
infiltration rates, and positively managing evaporation rates. However, research is still needed
on how different levels of LID technique integration effects model outputs and more importantly
how different levels of LID techniques effect urbanizing communities and their residents
including their perceptions of sustainability and the value of natural habitat particularly in urban
spaces. Additionally, the practice of LID and its potential to be implemented would benefit from
studies that compare stormwater model outputs, costs associated with stormwater management,
and the cost of these lid practices.
5.1 Limitations
With any research come limitations. Unfortunately the main limitation for this research is
the site that was selected. Taylor Yard was originally selected due to the intense drought that
46

southern California was experiencing. While the site offered a great way to model the
importance of collecting, filtering, and reusing water. However, due to the phasing process the
design for the site was not fully developed. This resulted in the use of a conceptual design. The
EPA calculator was more difficult to determine areas with a conceptual design. For this current
thesis, the solution was to provide estimates for all three sites in order to compare them equally.
It is obvious that design scenario 1 is mostly made up of impervious surfaces, but design
scenario 2 and 3 were more difficult to determine areas such as paving, green space, buildings,
LID elements, meadow, and forest.
A significant limitation of this research is using only the SWC, to improve understanding
and to better compare design alternatives further research should incorporate more variables and
models that would include other parameters such as energy use, habitat analysis, and possibly
even climate change. Although the EPA calculator relayed important information about low
impact development, these parameters could all be compared against each other using multiple
tests including Friedman’s Test. That would provide a better understanding of their benefits,
concerns, and how they impact design.
If future research were to compare additional parameters such as energy use, it would
make this research more viable. Energy use is an important parameter and can be determined by
analyzing the sites canopy coverage. In urban areas canopy coverage is extremely important
because it affects environmental elements such as shade, cooling, improve air quality, and reduce
stormwater runoff (Bartens, Day, Harris, Wynn, & Dove, 2009). The American Forests
recommend that a city should target 40% canopy coverage.

47

Another parameter that would further develop this research is the evaluation of current
and future climate change. In order to determine climate change the shadow casting and areas on
site will need to be determined. Because urban areas are the main generation of greenhouse
gases, it is determined that they have an increasing impact on climate change (Tian, 2016).
Climate change includes more intense weather such as frequent storming and intense changes in
temperature (Tian, 2016). By determining increasing the shadow area climate change can be
slowed down. LID applications can reduce and control stormwater runoff, as well as absorb heat
by providing shade area (Tian, 2016).
A final parameter that would further develop this research is to perform an ecosystem
analysis. By finding critical or important species on site, their habitat suitability can be calculated
for each design scenario. In order to do this the Habitat Suitability Index Model will need to be
applied. This model provides information regarding impact assessment and habitat management
(Sousa, 1983). This will determine which types are cover and site amenities are appropriate for
increase in reproduction. Ultimately one design will be more ideal than the others depending on
what type of cover each species requires.
5.2 Policy and Practice Recommendations
This research is an example of an application, The EPA National Stormwater Calculator,
and how it has a positive effect on design intentions. From this research it is recommended that
landscape architects, planners, engineers, and designers should be required to use such
applications in order to provide parameters that can be compared and contrasted. It is important
that the design phase includes measurable attributes that can show why different design decisions
were implemented. For example, in a new site development designers may implement different
concepts that affect stormwater management, ecosystems, climate change, or even energy use
48

and these applications can help determine what design decisions are best from an environmental
standpoint.
Regardless of how much, it is obvious that the use of LID elements effects stormwater
management. If designers were required to use the SWC application, they could further prove
why their design is more appropriate than the existing site. The research determines that it is
important that landscape architects, engineers, planners, and designers encourage each other to
go one step further in the design process and incorporate the use of these available applications
in order to raise more questions in the design field. Opening more questions within the design
field will give the design process more recognition and credibility.
5.3 Conclusion
In summation one can conclude that LID controls have the most positive affect on the
environment when combined with green space. It also investigates the use of LID controls and
their positive and negative effects on stormwater management practices. The purpose of this
research was not only to show the difference in use on sites with and without LID controls, but to
help pose questions for future research regarding the use of technology programs such as the
EPA National Stormwater Calculator. By posing other questions regarding LID controls, it is
possible that this can influence the standards and regulations of design requirements. Not only
does this research show that a combination of LID controls and green space have the most
positive influence on stormwater management, it also challenges landscape architects, engineers,
and designers to incorporate quantitative and qualitative research into each design in order to
prove that each site was designed to the best of their ability. As previously mentioned, some of
these variables include aesthetics, educations, and history. These qualitative variables are
important for communities to understand the comprehensive reasoning behind designs and helps
49

community members to better understand the site itself. By also providing clients with an
abundance of information regarding quantitative results such as average runoff rates before and
after the implementation of LID controls, one can see how low impact development can
positively affect urbanized areas.
Although this research provides initial results regarding LID controls, it sets a platform
for future research and how LID controls are site specific. Future research will help people to
understand the importance of stormwater management and its effect on the environment.

50

BIBLIOGRAPHY

51

BIBLIOGRAPHY

Bartens, J., Day, S. D., Harris, J. R., Wynn, T. M., & Dove, J. E. (2009). Transpiration and root
development of Urban trees in structural soil stormwater reservoirs. Environmental
Management, 44(4), 646–657. http://doi.org/10.1007/s00267-009-9366-9
Brander, K. E., Owen, K. E., & Potter, K. W. (2004). Modeled impacts of development type on
runoff volume and infiltration performance. Journal of the Ameri, 40(4), 961–969.
http://doi.org/10.1111/j.1752-1688.2004.tb01059.x
Bronnert, J., Masarie, C., Naeymi-Rad, F., Rose, E., & Aldin, G. (n.d.). Problem-Centered Care
Delivery: How Interface Terminology Makes Standardized Health Information Possible.
Retrieved from http://library.ahima.org/doc?oid=105588#.WQJoQvnyuUk
Burley, J. B., Li, N., Ying, J., Tian, H., & Troost, S. (n.d.). Metrics in Master Planning Low
Impact Development for Grand Rapids Michigan.
Convergent Water Technologies. (2014). LOW IMPACT DEVELOPMENT QUEENSTON
MANOR APARTMENTS. Retrieved from https://www.convergentwater.com/focalpointLow-Impact-Development/focalpoint-case-studies/low-impact-development-vs-traditionalstormwater-design
Davis, L. (n.d.). Constructed Wetlands Handhook, 1.
Dietz, M. E. (2007). Low Impact Development Practices: A Review of Current Research and
Recommendations for Future Directions. Water, Air, and Soil Pollution, 186(1), 351–363.
http://doi.org/10.1007/s11270-007-9484-z
Dietz, M. E., & Clausen, J. C. (2008). Stormwater runoff and export changes with development
in a traditional and low impact subdivision, 87, 560–566.
http://doi.org/10.1016/j.jenvman.2007.03.026
Eisenberg, Bethany, Lindow, Kelly Collins, and Smith, David R. (2017), (March).
Environmental Management Group. (2014). Mitigated Negative Declaration Taylor Yard River
Parcel G2 Project, (August).
EPA, Woodlands, T., Thelen, E., Howe, F., & Associates, C. (2017). 3/3/2017 History, 2–3.
Fletcher, T. D., Shuster, W., Hunt, W. F., Ashley, R., Arthur, S., Trowsdale, S., … Viklander, M.
(2017). SUDS , LID , BMPs , WSUD and more – The evolution and application of
terminology surrounding urban drainage. Urban Water Journal, 12(7), 525–542.
http://doi.org/10.1080/1573062X.2014.916314
Grant, S., & Giraud, D. (2015). Coastal California Rain Gardens, (October), 1–7.
52

Ingham County Drain Commissioner Plans Massive Urban Retrofit. (n.d.), 95.
Jaber, F., Woodson, D., LaChance, C., & Charriss, Y. (2012). Stormwater management: Rain
gardens.
Katz, B. (2017). Where Will 70 % Of The Population Live In 2050 ?, 4–7.
MDEQ. (2006). Design of Stormwater Filtering Systems.
National Asphalt Pavement Association. (2016). Engineering Overview, 14–15.
New Jersey Stormwater Manual. (2004). Standard for, (February), 1–11.
NRCS. (2017). Los Angeles County , California , Southeastern Part, 2–5.
RWRA. (n.d.). Minimum Standard 14.01: Infiltration Basin.
Selbig, W. R., & Bannerman, R. T. (2008). A comparison of runoff quantity and quality from
two small basins undergoing implementation of conventional- and low- impactdevelopment (LID) strategies: Cross Plains, Wisconsin, water years 1999–2005, (Lid), 57.
Sousa, P. J. (1983). Habitat Suitability Index Models: Field Sparrow, (82), 14.
Southeast Michigan Council of Governments. (2008). Low Impact Development Manual for
Michigan, 169–192.
Statistics, L. (2013). Friedman Test in SPSS Statistics.
Sylvawood, H. (n.d.). Low Impact Developments – Defining a Vision. Retrieved from
https://sylvawoodunderthestars.wordpress.com/2015/06/16/low-impact-developmentsdefining-a-vision/
The Massachusetts Department of Conservation and Recreation, & Ipswich River Watershed
Association. (2009). Ipswich River Targeted Watershed Grant Fact Sheet : Three LowImpact Development Case Studies. Retrieved from
http://www.mass.gov/eea/docs/dcr/watersupply/ipswichriver/twg-lid.pdf
The River Project. (2011). Map of Taylor Yard Parcels, 3–5.
Tian, H. (2016). EVALUATING LOW IMPACT DEVELOPMENT PERFORMANCE ON THE
SITE IN GRAND RAPIDS , MICHIGAN By Hongwei Tian A THESIS Submitted to
Michigan State University in partial fulfillment of the requirements for the degree of
Environmental Design - Master of Arts.
University of New Hampshire. (1995). Low Impact Development, (Lid), 1–24.

53

USDA, & Natural Resources Conservation Service. (2017). Classification _ USDA PLANTS.
Retrieved from https://plants.usda.gov/classification.html
USEPA. (2000). GUIDING PRINCIPLES FOR CONSTRUCTED TREATMENT
WETLANDS : Treatment Wetlands : Office of Wetlands, Oceans and Watersheds,
(October).
USEPA. (2014). National Stormwater Calculator User ’ s Guide, (January).
USEPA. (2016). Urban Runoff : Low Impact Development, 3–4.
Wang, M., Hyde, R. Q., Burley, J. B., Allen, A., Wang, M., Hyde, R. Q., … Allen, A. (2017).
Low-impact housing : River Rouge , Michigan Low-impact housing : River Rouge ,
Michigan. Housing and Society, 42(3), 193–206.
http://doi.org/10.1080/08882746.2015.1121679
Woolson, B. E. (2013). Barriers to Implementing LID.

54