The Evergreen State College Institutional Repository

Spatial Patterns and Equity Implications of Wetland Mitigation in Western Washington

Item

Title (dcterms:title)

Eng Spatial Patterns and Equity Implications of Wetland Mitigation in Western Washington

Date (dcterms:date)

2017

Creator (dcterms:creator)

Eng McKellips, Trace M

Subject (dcterms:subject)

Eng Environmental Studies

extracted text (extracttext:extracted_text)

SPATIAL PATTERNS AND EQUITY IMPLICATIONS OF WETLAND MITIGATION
IN WESTERN WASHINGTON

by
Trace M. McKellips

A Thesis
Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
June, 2017

©2017 by Trace M. McKellips. All rights reserved.

This Thesis for the Master of Environmental Studies Degree
by
Trace M. McKellips

has been approved for
The Evergreen State College
by

________________________
Edward A. Whitesell, Ph. D.
Member of the Faculty

________________________
Date

ABSTRACT
Spatial Patterns and Equity Implications of Wetland Mitigation in Western Washington

Trace M. McKellips
Under the Clean Water Act, regulatory agencies require developers to offset wetland impacts
through wetland mitigation. Wetland mitigation research has thus far centered on the ability
to restore or create functioning wetlands, yet little attention has focused on the spatial
distribution of wetland relocation. With no spatial tracking system firmly in place, regulatory
agencies know little about the aggregate distribution of wetland losses and gains. Using an
environmental equity framework, this research examines if wetland mitigation transfers
wetlands and their ecosystem services from urban to rural environments within a threecounty area in western Washington State. In addition, this research examines socioeconomic
and racial equity within wetland mitigation. Using 2010 Census data, this research collected
population density data, socioeconomic indicators, and racial demographics within a ¾ mile
buffer of each impact site and its corresponding mitigation site. This research then tested for
a difference in mean values between these sites. Findings indicate that wetland mitigation
relocates wetlands and their ecosystem service benefits along a pronounced urban-rural
gradient. Population densities are, on average, 926 people per square mile greater near impact
sites than mitigation sites. Mitigation sites have higher median incomes and higher
percentages of minority populations. To address the difficultly of linking spatial data
between impact and mitigation sites, this research recommends that regulatory agencies
maintain a spatial database of all wetland mitigation projects in order to better link the
distribution of wetland losses and gains, analyze spatial trends in wetland relocation and
assess how this relocation relates to human populations.

Table of Contents
CHAPTER 1: INTRODUCTION……………………………………………….……..…..1
CHAPTER 2: WETLAND MITIGATION IN THE UNITED STATES.........................12
Introduction……………………………………………………………………….………....12
Wetlands in the Early Years of the Republic……………………………………….…….…13
Clean Water Act………………………………………………………………….……….…15
No-Net-Loss………………………………………………………………………….….…..19
Wetland Mitigation Sequence……………………………………………………….….…...20
Compensatory Mitigation……………………………………………………………..……..22
National Research Council Findings………………………………………………….……..25
Supreme Court Interpretations of the Clean Water Act……………………………….…….26
Conclusion…………………………………………………………………………….……..28
CHAPTER 3: LITERATURE REVIEW…………………………………………………31
Introduction………………………………………………………………………….……….31
The Rise of Ecosystem Services in Environmental Management……………….…………..32
Ecological Economics…………………………………………………………….………….33
The Problem with Currency in Trading Wetlands………………………………….………..35
Wetland Valuations and Ecosystem Services…………………………………………….….38
Spatial Influences on Wetland Values……………………………………………………….42
Environmental Equity within Wetland Mitigation…………………………………………..44
Need for Further Research…………………………………………………………………...46
CHAPTER 4: METHODS…………………………………………………………………49
Introduction…………………………………………………………………………………..49

iv

Choosing the Study Area……………………………………………………………………50
Acquiring Data………………………………………………………………………………54
Importance of Scale…………………………………………………………………………56
Spatial Analysis………………………………………………………………………..........57
Urban-Rural Equity…………………………………………………………………………59
Economic Equity…………………………………………………………………………....60
Racial Composition..………………………………………………………………………..61
Measuring Differences in Means and Statistical Significance………………………….......61
Limitations………………………………………………………………………………......62
CHAPTER 5: RESULTS……………………………………………………………….....67
Summary …………………………………………………………………………………....67
Complete Study Area Results ……………………………………………………………....67
Findings by Approach and Location…..................................................................................70
Mitigation Bank Results………………………………………………………………….…71
In-Lieu Fee Results……………………………………………………………………….…80
Permittee-Responsible Mitigation Results…………………………………………………..81
CHAPTER 6: DISCUSSION………...................................................................................82
Introduction……………………………………………………………………………….…82
Urban-Rural Equity………………………………………………………………………….82
Socioeconomic Equity……………………………………………………………………….84
Racial Equity………………………………………………………………………………...85
Limitations………….………………………………………………………………………..87
Future Research………………………………………………………………………….…..88

v

Conclusion…………………………………………………………………………………..89
Recommendations………..…………………………………………………………………91
WORKS CITED……….…………………………………………………………………..95
APPENDICES……………………………………………………………………………..104
APPENDIX A: Ecosystem Services by Category …………………………………………104
APPENDIX B: Economic Valuation Methods……………………………………………..105
APPENDIX C: Definitions of Compensatory Mitigation Methods………………………..106

vi

List of Figures

Figure 1. Human benefits from wetland ecosystem services……………………………...3
Figure 2. Wetland values along an urban-rural gradient………………………………......6
Figure 3. Permanent impacts to aquatic resources from 2010-2014.……………………..16
Figure 4. Approaches to compensatory mitigation………………………………………..22
Figure 5. Mitigation approaches and wetland relocation……………………………….....23
Figure 6. Acre range of wetland impacts from 2010-2014……………………………......43
Figure 7. Three-county study area…………………………………………………………51
Figure 8. Urbanization and population trends in the United States………………………..53
Figure 9. Workflow of spatial analysis ……………………………………………………58
Figure 10. Map of Snohomish mitigation bank impact and mitigation sites………………71
Figure 11. Map of Springbrook mitigation bank impact and mitigation sites …………….73
Figure 12. Map of Skykomish mitigation bank impact and mitigation sites………………75
Figure 13. Map of Columbia River mitigation bank impact and mitigation sites…………77
Figure 14. Map of East Fort Lewis mitigation bank impact and mitigation sites……….....79
Figure 15. Map of King County ILF program’s impact and mitigation sites………….......80

vii

List of Tables

Table 1. Population growth in the three-county study area.....................................................51
Table 2. Summary statistics of study area…………………………………………………...69
Table 3. Average relocation distance by approach and location………….…………………70
Table 4. Summary statistics for Snohomish mitigation bank………………………………..72
Table 5. Summary statistics for Springbrook mitigation bank………………………………74
Table 6. Summary statistics for Skykomishh mitigation bank……………………………....76
Table 7. Summary statistics for Columbia River mitigation bank…………………………...78
Table 8. Summary statistics for East Fort Lewis mitigation bank…………………………...79
Table 9. Summary statistics for King County in-lieu fee program…………………………..81
Table 10. Summary statistics for Permittee-Responsible Mitigation………………………..81
Table 11. Comparative summary statistics of differences in population densities…………..83

viii

ix

Acknowledgements
I owe my gratitude to so many people that assisted me throughout this project and
throughout my time in the Masters of Environmental Studies (MES) program. First and
foremost, Dr. Ted Whitesell served as my attentive and enthusiastic thesis reader. Ted was in
tune to my intellectual development throughout the process, listened thoughtfully in our
conversations, and gently nudged me in the directions I needed to go. His timely and
constructive feedback throughout the writing and research process improved the final product
immensely.
Mike Ruth helped me review my spatial analysis methods and provided important
points to improve my data management. As an adjunct faculty member, Mr. Ruth was by no
means obliged to assist me with my requests. Nevertheless, like with so many other students,
he offered his time with his already hectic schedule to help me overcome numerous
geospatial challenges.
I also want to express my gratitude to the Department of Ecology Wetlands Team at their
headquarters in Lacey for creating an internship and work space. All the Wetlands staff—in
particular, Dana Mock, Kate Thompson, and Teri Granger—were generous with their time in
orienting me to their wetland mitigation data. They also facilitated a number of introductions
to agency partners that were helpful. Lauren Driscoll deserves thanks for completing all the
necessary paperwork for my position. Thanks to Megan Webb from King County Mitigation
Reserves Program who provided GIS data for their In-Lieu Fee program. Many folks spent
time with data requests that did not end up in the final analysis due to incomplete spatial data.
The Army Corps of Engineers assisted me with multiple Freedom of Information Act
requests in a timely manner. The Washington State Department of Transportation, Clark
County, Snohomish County, and Pierce County also fielded requests. Across the board,
agency personnel handled my requests in a timely and professional manner.
I would like to thank all the MES core and adjunct faculty who work so hard for the
program and its students. MES Director Dr. Kevin Francis embodies the spirit of
inclusiveness, intellectual curiosity, and reflection that makes the MES program and The
Evergreen State College a great place to learn. I also want to thank Writing Assistant Michael
Radelich for his guidance during my early drafts. Thanks to the various members of my MES
cohort who offered written and verbal feedback this past year.
Finally, I owe my gratitude and love to my family. To my parents, who have always
encouraged me to follow my interests—even when they were unsure where that would lead.
And to my wife Katrina, who supported me the past two years during this important time in
my professional development. I aim to emulate her extraordinary dedication and resolve to be
the best she can be.

x

Acronyms
ACOE – Army Corps of Engineers
CWA – Clean Water Act
DOE – Department of Ecology
EPA – Environmental Protection Agency
FWPCA – Federal Water Pollution Control Act
FOIA – Freedom of Information Act
FWS – Fish and Wildlife Service
NEPA – National Environmental Policy Act
NLCD – National Land Cover Database
RHA – Rivers and Harbors Act
RIBITS – Regulatory In-lieu Fee and Banking Information Tracking System
USDA – United States Department of Agriculture
USGS – United States Geological Survey
WSDOT – Washington State Department of Transportation
WTP – Willingness-To-Pay

xi

CHAPTER 1: INTRODUCTION
Wetland value appears to be maximum when distributed spatially across a landscape that is
not dominated either by cities or agriculture, but one that balances nature and human
enterprises.
Mitsch & Gosselink (2000)

For the first two centuries in the United States, Americans drained and filled wetlands
to make way for agriculture, cities, industry, and infrastructure. Initially, the vast majority of
Americans viewed wetlands as unproductive ecosystems with few benefits to local
populations (Hansen, 2006). Federal and state laws reinforced this prevailing attitude by
incentivizing the conversion of wetlands to other land uses (Vileisis, 1997). By the 1780s1980s, research estimated that Americans had converted a staggering 53% of wetlands in the
lower 48 states to other uses (Dahl, 1990).
While initially not well understood, wetland degradation has impacted aquatic
resources that have had adverse impacts on human populations (Vileisis, 1997). Wetlands
can be defined as “an ecosystem that depends on constant or recurrent, shallow inundation or
saturation at or near the substrate,” (NRC, 1995). Wetland habitats provide myriad benefits to
human populations. These human benefits derive from ecosystems and the processes can be
referred to as ecosystem services (MEA, 2005b). Ecosystem service benefits are not just the
sweet sounds of the songbirds that inhabit these systems, but real economic benefits.
Wetlands improve water quality by filtering toxic chemicals and impurities, reducing
expensive costs for stormwater treatment. Wetlands reduce peak flows of storms that cause
damages to public and private infrastructure. Additionally, wetlands provide cultural
ecosystem services such as educational opportunities, recreation, and spiritual values.

1

While understanding of wetland ecosystem services’ importance grew throughout the
20th century, their decline continued. President George H.W. Bush articulated a plan to
reverse this trend. In 1988, President Bush promulgated the country’s first national wetland
conservation strategy, calling for a “no net loss” of the country’s wetlands. The no-net-loss
plan however did not advocate halting all damage to existing wetlands. Rather, under the
powers of the Clean Water Act (CWA), the Army Corps of Engineers (ACOE) and the
Environmental Protection Agency (EPA) would regulate wetland impacts by requiring
developers to replace wetlands and their ecosystem services. This regulatory process is
known as wetland mitigation and is composed of a three-step sequence. Developers of a
project must first avoid and secondly, minimize and finally compensate for wetland impacts.
Clare, Krogman, Foote and Lemphers (2011) have critiqued this sequence as preferential
tothe final step of compensatory mitigation to achieve no-net-loss objectives and maintain the
country’s aquatic resources.
Compensatory mitigation uses an ecosystem-services framework to trade wetlands
from an impact site—where wetlands are damaged—to the mitigation site—where wetlands
are preserved, enhanced, restored, or created. In addition to acreage, regulators assess
specific ecosystem services (e.g., improved water quality, wildlife habitat) provided at the
impact site so there can be a commensurate transfer to the mitigation site. This arrangement
is predicated upon on the assumption that wetland ecosystem services have equal values
regardless of geographic location. This assumption, however, is flawed. A wetland’s multidimensional benefits, its position within a landscape, its hydrologic connection with other
aquatic resources, and its position in relation to human populations all interfere with an equal
transfer of ecosystem services from one location to another (Salzman & Ruhl, 2001).

2

In relation to human populations, wetlands provide ecosystem services at different spatial
scales. Figure 1 displays human benefits from wetland ecosystem services at individual,
community, and global scales. At the community and individual levels, proximity to
wetlands affects a group or individual’s ability to incur these benefits. Moreover, at the
community level, many benefits are evenly dispersed to the population at large, representing
local public goods (Mitsch & Gooselink, 2000).

Figure 1. Human benefits derived from wetland ecosystem services, adapted from Mitsch and
Gooselink, 2000.

As mitigation relocates wetlands across a landscape, this process produces outcomes
in which one population loses wetland functions and another population gains wetland
functions. But who is losing wetlands and who is gaining them? How far are wetlands being
relocated across the landscape? Acknowledging the host of benefits human populations
3

receive from wetlands, are there any issues of equity with respect to wetland relocation?
These questions are all inquiries into the nature of wetland mitigation, but only a handful of
previous researchers have addressed the spatial issue of wetland relocation and its potential
impacts on human populations.
These studies have identified one consistent trend: wetland mitigation relocates
wetlands from higher population densities to lower population densities (BenDor, T.K., &
Bruzovic, N., 2007; BenDor, T.K., Brozovic, N., & Pallathucheril, V.G., 2007; BenDor and
Stewart, 2011; Brass, 2009; King & Herbert, 1997; Robertson & Hayden, 2008; Ruhl &
Salzman, 2006). King and Herbert (1997) first identified this relocation of wetlands along an
urban-rural gradient. While this research uses the term “urban-rural gradient,” this term is not
a dichotomous distinction. Rather, the gradient falls along a continuum of urban, suburban,
peri-urban and rural environments with varying population densities.
Three potential factors may be driving wetland relocation away from urban areas.
These factors include the availability of land, economic incentives, and ecological
performance of wetlands in urban environments. First, the availability of land is an important
determinant for mitigation sites. Urban environments contain high percentages of built
infrastructure. Rural areas, by their nature, have more available land for potential wetland
restoration than urban areas.
As a second factor, economic incentives also influence wetland trading (Robertson,
2004). While regulated, wetland mitigation involves trading wetlands through a free-market
system. Developers and private businesses have no rational economic incentive for
conducting wetland mitigation on high-demand real estate. Rather, individuals acquire low-

4

priced land as they are permitted, acting in their own economic self-interest (Heal, 2000).
Cheaper land, in most cases, correlates to rural locations.
A third factor is based on ecological grounds; urban wetlands function poorly (Azous
& Horner, 2001; Kentula, Gwin, & Pierson, 2004). Burdened with an excess of pollution,
prior land use legacies, invasive plant colonization and habitat fragmentation, choosing
mitigation projects in urban areas often requires additional time and resources. One criterion
that potential mitigation sites are based upon is their likelihood to be self-sustaining after site
maintenance and monitoring ends. Excessive urban perturbations—or disturbances—
decrease the likelihood of self-sustaining sites. The National Research Council’s (2001)
influential review of compensatory mitigation recommended mitigation sites away from
urban areas with prior land use disturbances that could adversely affect mitigation site
performance.
Prioritizing rural site selection however has not been universally accepted. A
counterpoint to this logic argues that precisely because of urban disturbances, wetland
functions are more needed and their relative values are greater in urban environments (Ruhl
& Salzman, 2006). Mitsch and Gooselink (2000) conceptualize the relationship between
increasing wetland values and higher population densities in Figure 2.

5

Figure 2. Wetland values along an urban-rural gradient (Mitsch & Gooselink, 2000).
With more people in urban environments and a greater scarcity of wetlands, the
marginal value per unit of wetland increases. At an unknown tipping point, wetlands are
overwhelmed with the excesses of urban populations and their ecosystem functions collapse.
While this graph recognizes that the potential adverse urban influences overwhelming the
functional state of wetlands noted by the NRC (2001), it also conveys the increased marginal
value of urban wetlands. Wetland mitigation site selection exists with this tension between
balancing benefits to human populations and wetland functioning.
Attempting to quantify specific values of wetlands, however, continues to challenge
economists and ecologists (Boyd & Waigner, 2002). Not only do wetlands provide multidimensional benefits, their values are also dependent on the surrounding environment and
cultural value systems (Mitsch & Gooselink, 2000). For example, a rural wetland may be
valued for duck hunting, which is not feasible in an urban setting (Brass, 2009). The high
degrees of variability hinder an agreed-upon valuation of wetlands. While land use planners

6

and regulators understand the overall range of wetland ecosystem services, integrating these
values within developed landscapes remains an elusive goal.
Many land use challenges facing urban and urbanizing western Washington
communities relate to the scarcity of ecosystem services that wetlands provide. During rain
events, impervious surfaces limit rainwater infiltration of stormwater into underground
aquifers. Instead, roads and underground piping serve as the network of stormwater
conveyance, which transports stormwater from the asphalt straight into waterways. This
conveyance bypasses the process of water coming in contact with soil, which acts to purify
the water from many of chemicals it picks up along the way (Trombulak & Frissell, 2000).
When stormwater pipes flow into creeks and rivers, this system of conveyance exacerbates
peak flows and its corresponding flooding. Not only is this water lost for future human use,
this stormwater can be perilous for aquatic life, particularly for anadromous fish whose
spawning cycles are triggered by rain events (Alberti et al., 2007; Scholz et al., 2011). As a
habitat type, functioning wetlands moderate these aquatic impacts. Wetland ecosystem
services are not strictly confined to managing water, however. Human populations enjoy
wetlands for their many cultural ecosystem service benefits such as aesthetics, recreation, and
educational opportunities (Manuel, 2003).
Proximity to wetlands influences populations’ ability to incur and enjoy their
ecosystem service benefits. Given this spatial dynamic, wetlands should be dispersed in the
landscape so as not to exclude any social group. This dispersion is a central tenet of
environmental equity, defined as “the proportionate …distribution of environmental benefits
and risks among diverse economic and cultural communities. It ensures that the policies,
activities and the responses of government entities do not differentially impact diverse social

7

and economic groups” (DOE, 2013 p. 1-2). There is good reason to apply an environmental
equity lens to wetland mitigation. First, research has argued that equitably distributing green
infrastructure projects could promote urban poverty alleviation (Dunn, 2010). Second, a
growing literature has analyzed the inequitable distribution of parks and open spaces among
different social groups (Jennings, Gaither, & Gragg, 2012) but few have extended this
analysis to wetlands.
With an average of 8,000 acres of permanent impacts to non-tidal wetlands annually
(IWR, 2015), the success of current wetland mitigation hinges on the ability to relocate
wetlands from one location to another. This relocation results in a transfer of wetland
ecosystem services that could affect human populations near the impact sites—where
wetland impacts occur—and mitigation sites, where wetlands are restored, preserved, or
created (Mitsch & Gooselink, 2000). Unfortunately, this spatial redistribution of wetlands
and its relation to human populations remains poorly understood (BenDor, Brozovic, &
Pallathucheril, 2008). Principally, state and federal regulators do not maintain a spatial
database to track where wetland losses and gains are occurring. Rather, regulatory personnel
manage wetland mitigation projects on a project-by-project basis without knowledge of
aggregate spatial patterns. As impact and mitigation sites are dispersed throughout the built
environment of human populations, the distribution of wetland mitigation losses and gains
and their relation to communities represent a significant knowledge gap.
Several academic studies have collected base data that examines wetland mitigation
and human populations. The geographic study areas in previous analyses included Florida
(King & Herbert, 1997; Ruhl & Salzman, 2006), North Carolina (BenDor & Steward, 2011),
a three-county area in northeast Illinois (BenDor, Brozovic, & Pallathucheril, 2007;

8

Roberston & Hayden, 2008) and a three-county area in central Oregon (Brass, 2009). These
past study areas represent a very small sample of wetland mitigation projects in the United
States.
Spatial relocation of wetland mitigation and its equity implications on human
populations lay the foundation of this research project. To better understand spatial patterns
of wetland relocation through the mitigation process, this study examines a three-county
study area of Clark, King, and Snohomish counties in western Washington. These counties
have urban-rural population gradients that serve as an apposite natural laboratory to examine
site selection patterns.
The central research question in this thesis examines if wetland mitigation relocates
wetlands and their ecosystem service benefits from urban to rural areas. In addition, this
research examines socioeconomic equity and racial equity in wetland mitigation. To
accomplish this, this research analyzed 139 wetland mitigation projects across three counties
in western Washington State with diverse urban-rural gradients. Using a ¾ mile buffer
around each impact site and its corresponding mitigation site, this research analyzed
differences in the human populations affected by wetland mitigation projects. The results
from this study conclude that wetland mitigation in western Washington relocates wetlands
in the following manners:


Along a pronounced urban-rural gradient



From lower to higher income populations



From lower to higher percentages of minority populations

Results from this study advance our understanding of wetland mitigation’s spatial
redistribution of wetlands and its equity implications on human populations. Viewed in

9

tandem with previous studies, these spatial tendencies advance our knowledge of diminishing
urban wetland resources. While Western Washington manages a growth boom, planning for
urban landscapes that balance human populations and ecological resilience remains a major
challenge for decision-makers at the city, state, and federal levels (Godschalk, 2004). As
market-based strategies for trading ecosystem services expands, research needs to critically
analyze both its ecological and social impacts. Improved knowledge of the process of
wetland mitigation and the degree to which it is relocating ecosystem services across
landscapes will help regulatory bodies examine their guidelines and the potential adverse
effects on urban environments. Increased understanding of these spatial and social
characteristics should inform future guidelines that instruct wetland mitigation site selection.
This research project proceeds in the following manner: Chapter 2 provides a brief
overview and history of wetland management in the United States. This background chapter
includes a brief history of widespread wetland degradation, important regulations, shifts in
public perceptions of wetlands’ utility, efforts to preserve and protect aquatic resources, the
development of wetland mitigation, how mitigation has evolved over the past four decades,
and Supreme Court rulings on mitigation jurisdiction. Chapter 3 reviews the academic
literature relating to the spatial influences and societal impacts of wetland mitigation,
synthesizing relevant studies in ecosystem services, economic valuations, and the trading of
environmental goods and services. Chapter 4 reviews methods used in this study to determine
the extent of wetland relocation in western Washington, including data acquisition,
geographic information systems (GIS) analysis using ESRI software, and statistical analysis
using JMP software. Chapter 5 presents results from this study. These results include maps of
spatial distribution and tests of statistical significance in determining differences between

10

impact and mitigation sites, and variance by mitigation type. Chapter 6 discusses the
findings in the preceding chapters, how these results compare to previous studies, and the
implications for wetland mitigation. This chapter also surveys the limitations and constraints
of the present study. The thesis ends with a concluding section, synthesizing the main points
found in this research and stating recommendations for future research.

11

CHAPTER 2: WETLAND MANAGEMENT IN THE UNITED STATES
Researchers and planners have yet to construct systems that enable them to address the basic
question as to whether wetland mitigation contributes to social disparity and inequity.
BenDor, Brozovic, & Pallathucheril (2008)
Introduction
The central question in this research asks if wetland mitigation relocates wetlands and
ecosystem services from urban to rural locations. Corresponding to this question, this
research examines the equity implications of local populations surrounding impact and
mitigation sites. To understand how wetland resources are being relocated throughout the
landscape, this chapter reviews past wetland management in the United States.
Over the past 250 years, over 100 million acres of wetlands have been drained and
converted for other uses (Hansen, 2006). By acre, the majority of these historical wetland
drainages and fills were to increase agricultural production, but also included drainages to
make way for urban and rural development, transportation infrastructure, and industry.
Throughout the 20th century however, increased knowledge of wetlands’ importance created
a movement to improve management of our country’s aquatic resources. The CWA advanced
a paradigm shift in federal policies, reversing trends of widespread wetland conversion.
Halting wetland loss, however, did not happen overnight. Due to a number of technical
challenges and oversight limitations, the first decades of wetland mitigation often failed to
replace wetland functions and their ecosystem services (NRC, 2001; Turner, Redmond, &
Zedler, 2001). To improve site performance, shifts in guidelines have recommended greater
percentages of off-site mitigation. The effects of these guidelines—greater wetland
relocation—reinforce the need to develop an integrated geospatial system to track and
analyze the redistribution of important wetland resources.

12

This chapter first chronicles wetland management in the early years the republic,
marked by widespread degradation that coincided with the country’s burgeoning population.
The second section examines the landmark legislation of the Federal Water Pollution Control
Act of 1972, later amended as the Clean Water Act of 1977, which established the wetland
mitigation process. The third section describes the national goal of no-net-loss of wetlands
and how compensatory mitigation efforts have been used to achieve this goal. The specific
approach of mitigation available and their potential effects on wetland relocation are
examined. Fourth is a section describing the mitigation sequence developed to reduce the
impacts to aquatic resources. Since the final step of the sequence—compensatory
mitigation—is the only step that relocates wetlands, the fifth section details the complexity
and nuances of this step. In 2001, the National Resource Council (NRC) assessed how well
compensatory mitigation was performing and made recommendations for future efforts.
Given the report’s influence on current mitigation practices, the sixth section reviews these
findings. The last section reviews how the Supreme Court interpreted two important cases
involving the CWA and wetlands. As wetland management has progressed to the present,
wetland loss has slowed considerably. For its part, mitigation relies on the relocation of
wetlands across landscapes to help achieve national no-net-loss objectives. To better
understand why maintaining the country’s aquatic resources remains imperative, this
research first surveys wetland degradation in the early years of the American republic.
Wetlands in the Early Years of the Republic
Throughout the Unites States’ nearly 250-year history, Americans’ relationship with
wetlands has changed dramatically, from policies encouraging the draining and filling of
wetlands to the current regulatory environment seeking to increase total wetland acreage.

13

Complex tradeoffs between economic gains from wetland impacts and wetland ecosystem
service benefits have been debated at each level of wetland management, from local to
federal. While this research presents only a tip-of-the-iceberg account of wetland history, for
a comprehensive account of wetland policy and history, see Vileisis (1997).
In the early years of the republic, government policies encouraged wetland drainage
through incentive programs to increase agriculture and harness previously inaccessible land
for urban and rural development. Known as the Swamp Land Act, Congress passed the first
major piece of wetland legislation 1849. The Swamp Land Act ceded federally owned
wetlands to the states. In turn, states could sell these lands in order to fund levee construction
and building drainage infrastructure to decrease flooding, a perennial problem for
communities built on the banks of undammed rivers. As Vileisis (1997) notes, this piece of
legislation also brought one of the first public debates about defining wetland boundaries,
functions, and benefits to the national stage (p. 73-74). Nevertheless, early settlers poorly
understood wetland benefits at this time. In the view of most settlers, wetlands hindered
progress (Dahl, 1990). Understanding that drained wetlands provided rich agricultural land
strongly incentivized wetland draining and filling. In total, this program converted an
estimated 26 million hectares of wetlands to non-wetland uses (Mitsch & Gooselink, 2015).
Not until the Rivers and Harbors Act of 1899 (RHA, 1899) did the federal government begin
to regulate dredge and fill operations.
The initial intent of the RHA was to ensure navigability of U.S. waterways. Regulated
by the Army Corps of Engineers (ACOE), this agency granted permits to ensure dredge and
fill operations did not block navigable routes. The jurisdiction of the RHA did not extend to
“waters of the United States.” Rather, jurisdiction was confined to navigable waters—a much

14

narrower geographic area than the future CWA. During the 20th century, the ACOE also was
charged with building hydropower, dam, and levee infrastructure—all duties that
dramatically altered riverine and riparian wetlands (Vileisis, 1997). Thus, while the ACOE
managed water, protecting wetland resources was not their top priority.
Through pressure from another federal agency, the Fish and Wildlife Service (FWS),
the ACOE began addressing environmental degradation through the RHA in 1967 (Hough &
Robertson, 2009). With growing recognition of rampant industrial pollution and polluted
water resources, the ACOE instated a public interest review to assess a proposed project’s
suitability. This new review assessed projects not only for effects on the navigability of
waters, but also for effects on “fish and wildlife, conservation, pollution, aesthetics, ecology,
and the general public interest” (quoted in Downing, Winer, & Wood, 2003, p. 477). Even
with these new policies, momentum was building for stronger environmental protections at
the federal level.
Clean Water Act
Despite a presidential veto, Congress passed the Federal Water Pollution Control Act
Amendments of 1972 (FWPCA, 1972), creating the strongest legislation to date to protect
aquatic resources. This agreement came just five years after the ACOE had adopted more
stringent review measures, soliciting worry that the new regulations would simply be
duplicating regulations in the RHA (Hough & Roberston, 2009). Nevertheless, the FWPCA
found purchase with Congress as consciousness grew around the impacts industry and
development were having on the nation’s aquatic resources. The stated goal of the FWPCA
was to “restore and maintain the biological, chemical, and physical integrity” of navigable
waterways (FWPCA, §230(1)). While still maintaining the same geographic reach of

15

navigable waterways, the purpose of the legislation expanded to protect the important
ecosystem services of aquatic resources.
In 1977, Congress made key amendments to the FWPCA that still guide wetland
mitigation today. These amendments also marked when the law attained its modern-day title,
the Clean Water Act. (Henceforth, this study refers to the FWPCA by its colloquial title, the
CWA). It is important to note that the initial FWPCA did not contain “wetlands” anywhere in
the legislation. The 1977 amendments changed the jurisdiction from navigable waters to
“waters of the United States,” which included adjacent wetlands, isolated wetlands, and
tributaries of major rivers. This expansion can be attributed to the growing understanding of
hydrological connections—often underground—between wetlands and other aquatic
resources like lakes, rivers, and aquifers. Including wetlands within §404(1)(b) greatly
increased the regulatory scope of the CWA. Figure 3 shows the disproportionate impacts to
non-tidal wetlands compared with other aquatic resources.

Figure 3. Permanent impacts to aquatic resources from 2010-2014 (IWR, 2015).
16

Protecting aquatic resources required offsetting impacts to wetlands through wetland
mitigation, a complex regulatory system to ensure the CWA replaces essential ecosystem
services. Mitigation under the CWA follows a three-step sequence—avoid, then minimize,
and lastly, compensate—to moderate the impacts to wetland resources. The sequence order
prioritizes each step before moving on to the next step. With this reasoning, regulators prefer
the first step, avoidance, above all else. If avoidance cannot happen, regulators favor the next
option, minimizing wetland impacts. Only after exhausting these first two options can
compensatory mitigation be considered an option. While this process ostensibly limits the
role of compensatory mitigation, research has argued that the mitigation sequence leans too
heavily on this final step (Clare et al., 2011).
With this new legislation, the ACOE no longer solely held the regulatory reigns.
Rather, the ACOE would oversee the day-to-day permit application process to impact
wetlands—known as §404(1)(b)—while the EPA would oversee compliance and issue
compliance guidelines. In addition, if the EPA had the power to exercise veto powers over
ACOE decisions the EPA disagreed with, limiting the broad discretion the ACOE previously
held. This new oversight role by the EPA created inter-agency tension, as the EPA and
ACOE struggled to bilaterally manage mitigation programs (Hough & Robertson, 2009).
The ACOE, familiar with management under the RHA, resisted adopting an
organizational mentality that strongly protected aquatic resources. After the CWA passed,
many assumed the law would lead to a rejection of permit applications that damaged
wetlands. As it turned out, the ACOE denied few permits. When the EPA exercised its veto
power, the ACOE initially resisted this oversight. Before the passing of the CWA, the ACOE
had sole jurisdiction to regulate the RHA. The CWA changed this dynamic, with the EPA

17

overseeing ACOE decisions. Even after an out-of-court settlement from National Wildlife
Federation v. John O. Marsh Jr. (1981) in which the ACOE agreed that EPA mitigation
guidelines were binding, the ACOE released an internal guidance document days later stating
the contrary, that the EPA guidelines were advisory only (ACOE, 1984). Despite
disagreement, the EPA did little to exercise its veto authority. Rather, the two agencies failed
in their respective capacities to curb wetland impacts. The ACOE showed little propensity to
deny permit applications while the EPA failed to use its veto power to challenge the ACOE’s
decisions (GOA, 1988). While the agencies struggled to come together with a shared
purpose, developers had little regulatory clarity to follow.
Turning the broad objectives of the CWA into an effective regulatory mechanism
proved to be an ambitious task. Chief among the challenges was aligning all federal
agencies—not just the ACOE—with dubious histories in wetland management to follow the
CWA goal of protecting the integrity of U.S. waters. The Department of Agriculture (DOA),
for instance, had long subsidized wetland drainage to increase agricultural acreage and
productivity. From 1940–1977, an estimated 23 million hectares of wetlands were converted
through the DOA’s Agricultural Conservation Program (Mitsch & Gooselink, 2015).
Incentives continued after Congress passed the CWA, creating a situation in which one
federal agency subsidized wetland drainage and another agency that penalized it. In 1985, the
Food Security Act cut these agricultural subsidies. Known as “swampbuster” programs, these
initiatives helped unify federal agencies in the protection of wetlands.
Nevertheless, the 1970s and 80s marked a slow start to curbing wetland conversion.
Even as the DOA halted their wetland conversion programs, the ACOE resisted a strong
interpretation of protecting wetland resources. It would be another five years until the agency

18

came to an agreement on the intent and administration of the CWA. This period also aligned
with President George H.W. Bush’s national wetlands goal.
No-Net-Loss
In the late 1980s, President George H.W. Bush’s wetland initiative brought wetlands
and their ecosystem service benefits into the national spotlight. While President Nixon and
President Carter had issued Executive Orders directing federal agencies to increase wetland
protection, President H.W. Bush upped the ante for American wetland protection by
proposing to reverse the net loss of wetlands. President Bush adopted his pro-conservation
attitude after national wetland inventories estimated that Americans had converted over 50%
of wetland resources (Tiner Jr., 1984). Another assessment estimated annual wetland loss
between the 1950s and 1970s at 439,000 acres per year (Frayer, Monahan, Bowden, &
Graybill, 1983). In light of these assessments, the National Wetlands Policy Forum in 1987
set forth a new agenda to protect wetlands. The top recommendation advised a “no-net-loss”
national policy. Through halting wetland conversion and investing in wetland restoration, the
United States could set a trajectory for long-term wetland gain (Hough & Roberston, 2009).
Speaking at a Ducks Unlimited gathering, President Bush embraced this no-net-loss
framework and beckoned his countrymen to support strong environmental protections.
I want to ask you today what the generations to follow will say of us 40 years from
now. It could be they'll report the loss of many million acres more, the extinction of
species, the disappearance of wilderness and wildlife; or they could report something
else. They could report that sometime around 1989 things began to change and that
we began to hold on to our parks and refuges and that we protected our species and
that in that year the seeds of a new policy about our valuable wetlands were sown, a
policy summed up in three simple words: "No net loss." And I prefer the second
vision of America's environmental future.
Bush, G.H.W., 1989

19

The no-net-loss goal set forth the goal that for every acre of wetland damage, at least
one acre had to be replaced. To achieve no-net-loss, federal regulations needed a robust
method to account for wetland loss in order to increase wetland acreage elsewhere. Wetland
mitigation under the CWA fit this framework of tracking wetland loss and gains to achieve
no-net-loss standards.
After nearly two decades of disagreement about jurisdiction within the CWA, the
ACOE and EPA jointly published a memorandum of understanding (MOA) in 1990 (EPA,
2017a). The MOA ended the conflicting agency goals and clarified the mitigation sequence
still practiced today. The choices made in the mitigation sequence determine the extent in
which wetland mitigation will relocate wetlands across the landscape. Since the 1990 MOA,
these preferences have not been stagnant. Rather, updated mitigation guidelines have resulted
in higher proportions of wetland relocation. The next section describes the complex
regulatory framework known as the mitigation sequence, which forms the basis for
maintaining wetland resources and for wetland relocation though permitted wetland impacts.
Wetland Mitigation Sequence
The mitigation sequence follows three distinct steps. This first step to mitigate
wetland impacts is avoidance. If alternatives for a project exist without damaging wetland
resources, CWA guidelines instruct developers to seek these alternatives. Avoiding impacts
altogether would be the most efficient way to protect existing wetland resources. Research in
Canada however, which has a similar regulatory framework to the United States, critiqued
the efficacy of the avoidance policy (Clare et al., 2010). This research attributed lack of
avoidance measures to the lack of clarification on what “avoidance” means, not prioritizing

20

high-value wetlands, undervaluing wetlands in economic valuations, and an over-confidence
in the capabilities of restoration ecology to restore and create wetlands.
In between avoidance and compensation rests the second step of minimization. If
wetland impacts cannot be avoided, then developers should seek to reduce wetland impacts
to the extent possible. These steps, outlined in Section H of the §404(1)(b) guidelines,
resemble best management practices (BMP) when dealing with dredged or fill material.
Examples of minimization include covering materials to prevent erosion, changing the timing
of project work to avoid spawning or nesting seasons, and using appropriate technology such
as employing mats under heavy equipment to avoid compaction (Gardner, 2005). Like the
first step of avoidance, minimization has received little attention as an alternative to
compensatory mitigation (Hough & Roberston, 2009). As Clare et al. (2010) argue, “The
language that allows compensation if avoidance or minimization ‘is not practicable’ becomes
a de facto loophole in its non-specificity, allowing developers to skirt the intent of the law
and move directly to compensation,” (p. 169).
As a means to protect existing wetland resources, regulators should prioritize the first
two steps of the mitigation sequence. These two steps rely on naturally-occurring
hydrological cycles and other wetland functions already established. As has been
documented, wetland restoration and creation have often produced mixed results (Zedler,
1996). As one EPA employee remarked, “In my view, nature has a remarkable track record
in creating wetlands, and developers do not,” (as quoted in Vileisis, 1995, p. 324). While
advances in restoration ecology have improved site success, weak implementation of the first
two steps represents a missed opportunity for wetland protection. Rather, mitigation has been

21

structured around compensatory mitigation, the third—and least preferable—step in the
sequence.
Compensatory Mitigation
As the name implies, compensatory mitigation requires developers to compensate for
their damage to wetlands. This final mitigation sequence step requires restoration,
establishment, enhancement, or preservation to replace wetland functions. Since mitigation
requires more than a 1:1 replacement of wetlands, compensatory mitigation should create a
net gain of wetland acreage. In compensatory mitigation, developers can choose between
three primary mitigation methods—also referred to as approaches—to fulfill their mitigation
requirements. These approaches include permittee-responsible mitigation (PRM), In-Lieu
Fees (ILF), or mitigation banking. Figure 4 displays the different approaches to
compensatory mitigation.

Figure 4: Approaches to compensatory wetland mitigation.
Permit applicants can achieve these requirements on-site, which compensates for
impacts at the same location as the wetland impacts, or off-site, which relocates

22

compensatory efforts in a different location. This choice of mitigation site selection features
prominently in determining wetland relocation. Figure 5 provides a graphic to visualize how
each mitigation approach relocates wetlands across a landscape.

Figure 5. Wetland mitigation approaches and wetland relocation (BenDor et al., 2007).
The second avenue allows developers to pay a commensurate fee for third-party
mitigation. Third-party mitigation can take two forms, mitigation banking or in-lieu fee (ILF)
programs. Mitigation banks have one large mitigation site, called a mitigation bank, which
mitigates for multiple impacts. Developers purchase wetland “credits” before a development
project begins. For each mitigation bank, regulatory agencies determine the total credits
within the mitigation bank, each credit’s monetary value and define the service area, the
geographic area where wetland impacts can occur. When regulators approve a project, the
credits are debited into the banking ledger, which tracks all impacts and credit usage. For a
potential developer, buying wetland credits for a project can be a cost-effective and
timesaving choice. For conservationists interested in protecting wetland resources, wetland
banks allow strategic site selection of one large restoration site that can provide a variety of
wetland ecosystem benefits (NRC, 2001). With one large mitigation site and many impact

23

sites, wetland relocation across the landscapes is more pronounced. If a wetland bank is
located adjacent to high population densities, this should provide a host of wetland ecosystem
services to adjacent populations. If the wetland bank is located in an area with a low
population density, relatively fewer people may benefit from this bank’s ecosystem services.
ILF mitigation, on the other hand, differs from wetland banks in its form of currency.
While wetland banks use a credit currency, an ILF permittee pays a fee commensurate with
the wetland impacts to a government agency or non-profit with restoration expertise. These
programs however have struggled to link payments with wetland mitigation (ELI, 2006).
Without clearly defined objectives, ILF funds often went to expenditures other than wetland
mitigation (Hough & Robertson, 2009). At one point, the ACOE and EPA even considered
eliminating ILF as a form on compensatory mitigation (ELI, 2006). Nevertheless, the initial
shortcomings have been identified, adjustments have been made to better link ILF with
targeted mitigation, and ILF continues to be an option. To date, the DOE has approved three
ILF programs in Washington.
Despite their differences, wetland banks and ILF programs share many attributes.
Both of these two forms were designed to improve outcomes for mitigation sites. Both
approaches reduce temporal loss in wetland mitigation. Temporal loss refers to the lag time
when a wetland is impacted and the time it takes to restore wetland functions (i.e. ecosystem
services) at a mitigation site. Both approaches involve off-site mitigation that allow multiple
impact sites to go toward one, larger mitigation site. Both approaches will likely grow as
recommendations have shifted from on-site to off-site mitigation (NRC, 2001).
In the first decades after the CWA, wetland mitigation failed to replace wetlands
ecosystem services across landscapes. As a way to track progress in wetland mitigation, the

24

National Resource Council (NRC) began a comprehensive study, per the request of the
ACOE and EPA. In a telling decision of the regulatory agencies’ priorities, the report only
assessed compensatory mitigation, leaving out assessments for the first two steps in the
mitigation sequence. The report’s first conclusion did not mince words. “The goal of no net
loss of wetlands is not being met for wetland functions by the mitigation program, despite
progress in the last 20 years,” (NRC, 2001, p. 2). The NRC attributed mitigation’s no-netloss failures to myriad organizational and procedural shortcomings.
National Research Council Findings
Maintaining the ecological functions of compensatory mitigation sites to be selfsustaining over time remains one of the greatest challenges for wetland mitigation (Zedler,
1996). To address this challenge, the NRC (2001) recommended a watershed approach. A
watershed approach identifies the host of biotic and abiotic features of the landscape to be
considering when selecting a mitigation site. These include climate, topology, hydrology and
soil conditions. In the context of this thesis, one critical recommendation warns against
selecting mitigation sites in “seriously degraded or disturbed sites” (p. 5). Increasing levels of
urbanization yield more disturbed sites. Thus, this recommendation gives preference to rural
areas of low population densities over urban areas with high population densities. Further,
the report details that mitigation sites with floral communities not yet fully established are
susceptible to the perturbations of population growth and human influences. In Washington
State, this approach has been adopted by the DOE in their guidance, “Selecting Wetland
Mitigation Sites Using a Watershed Approach” (Hruby, Harper & Stanley, 2009).
The NRC also identified other shortcomings of compensatory mitigation, many of
which make it difficult to assess wetland relocation across landscapes. First, unclear

25

performance standards impede compliance with mitigation requirements. If a mitigation site
exhibits reduced ecosystem service benefits compared to an impact site, this divergence
complicates assessing the functional equivalency of wetlands. This lack of equivalency
between impact and mitigation sites also hampers assessing the redistribution of ecosystem
service benefits to human populations. Current understanding of the complex nature of
wetland dynamics prevents certainty about the transfer of ecosystem services.
Finally, tracking how compensatory mitigation programs affect wetland resources
across the landscape remains low. The NRC (2001) recommended maintaining a database to
properly track mitigation progress. In 2007, the ACOE developed the Regulatory In-lieu fee
Bank Information Tracking System (RIBITS). This database captures ILF and wetland
banking programs, although PRM is absent. The impetus for developing the system was to
provide developers easily accessible information by which to know if using any mitigation
service areas were contained within their proposed development. RIBITS could also present
a spatial representation of where wetland impact and mitigation sites are located. Another
database called ORM2 has also been developed to track all types of §404(1)(b) permits.
While the ACOE tracks impacts with ORM2, site location data within the protected database
does not contain coordinate data for mitigation sites, preventing geospatial analysis (R.
Haines, personal correspondence, April 26, 2017). The ACOE shares database information
only through Freedom of Information Act (FOIA) requests.
Supreme Court Interpretations of the CWA
While much of the discussion in this section has centered on the ACOE and EPA, the
Washington State Department of Ecology (DOE) also plays a prominent role in

26

compensatory mitigation by sharing permit responsibilities with the ACOE. Recent court
decisions have limited federal authority of the CWA, creating a larger role for state agencies.
This shift first began with the Supreme Court hearing of Solid Waste Agency of Northern
Cook County (SWANCC) v. Army Corps of Engineers (2001).
In a 5-4 decision, the Court ruled that the ACOE overreached their jurisdiction by
applying isolated wetlands used by migratory bird species to the CWA. While waters
adjacent to rivers and other interstate water bodies were within their jurisdiction through the
Commerce Clause, the Court pointed out that regulating isolated wetlands—wetlands without
direct hydrologic connections to other aquatic resources—misinterpreted the original text of
the CWA. The Court’s decision questioned the broad interpretation of waters of the United
States, bringing the question of applying navigable waterways back into the spotlight. This
questioning went against actions governing the previous decades, which sought to protect the
biological and hydrological integrity of U.S. waters (Downing et al., 2003).
Rapanos v. United States (2006) further limited federal oversight of isolated
wetlands. While the Court failed to issue a majority opinion (4–1–4), Justice Kennedy’s lone
interpretation has been most influential. With four Justices narrowly interpreting CWA
jurisdiction to include wetlands with a surface connection to navigable waters and the other
four Justices interpreting tributaries and adjacent wetlands, Justice Kennedy took the middle
of the road, citing the term “significant nexus” used in SWANCC v. Army Corps of
Engineers. In his words:

27

Wetlands provide the requisite nexus, and thus come within the statutory phrase
“navigable waters” if the wetlands, either alone or in combination with similarly
situated wetlands in the region, significantly affect the chemical, physical, and
biological integrity of other covered waters more readily understood as “navigable.”
When, in contrast, wetlands’ effects on water quality are speculative or insubstantial,
they fall outside the one fairly encompassed by the statutory term “navigable waters.”
Justice Kennedy opinion, Rapanos v. United States, 2006, p. 779-780.
Challenging regulators to examine this nexus acknowledges the complex interactions
wetlands have with underground aquatic resources. Indeed, Craig (2008) argues the nexus
standard should lead to integrating ecosystem services within CWA regulation. In particular,
this nexus highlights the regulating service of filtering sediment and pollutants from the
surface waters. While not explicitly defining the term, Justice Kennedy’s words underscore
the importance of wetlands in combatting nonpoint source pollution. The CWA originally
identified point-source pollution, or pollution coming from one source that can be
geographically isolated. CWA amendments in 1987 included non-point sources as well. As
the name implies, non-point source pollution cannot be traced back to one location. Rather,
non-point source pollution comes from myriad sources throughout the landscape. Examples
include accumulated chemicals from stormwater, pesticides from agricultural runoff, or
urban areas with high percentages of impervious surfaces. Capturing non-point source
pollutants before entering the Puget Sound or Columbia River is a critical ecosystem service
that wetlands provide in urban environments in western Washington.
Conclusion
Since European colonization of the United States, wetland resources have been
drained and filled to increase economic productivity. Chief among these converted uses have
been agriculture and urban development. Government programs consistently incentivized
wetland conversion, such as when the federal government granted wetlands to state

28

governments to sell in order to combat frequent flooding. With policies such as these, an
estimated 53% of the United States’ wetland resources were converted in a span of 200 years
(Dahl, 1990). Industrialization also began severely polluting our nation’s waterways. In
response to the degradation of U.S. waters, Congress passed the landmark legislation in 1972
that paved the way for the CWA. However, not until the Swampbuster programs of 1980’s
did federal policy align to prioritize wetland protection.
The CWA charges the ACOE and the EPA with protecting the “biological, chemical,
and physical integrity” of U.S. waters. Taking this broad language and turning it into an
agreed-upon regulatory program has proved challenging. First, the ACOE and EPA
themselves have disagreed upon jurisdiction and procedure. The Supreme Court and lower
courts continue to interpret the CWA, altering wetland management. Nevertheless, wetland
mitigation currently administered under the CWA presents an intricate regulatory program to
protect aquatic resources and their underlying ecosystem services. While a complete freeze
on damaging wetland resources is considered incompatible with economic development,
wetland mitigation has developed to offset the adverse impacts to wetland resources. The
mitigation sequence has three distinct steps, but it is the last step of compensatory mitigation
that assumes the most prominent role in practice. Compensatory mitigation has evolved
during the past decades, from an initial preference for on-site wetland mitigation to a
preference for off-site mitigation. This off-site mitigation relocates wetlands and their
benefits from one location to another.
This thesis research examines this aspect of off-site, compensatory mitigation.
Nationwide, the CWA impacted an average of 13,300 acres of wetlands annually from 20072014 (IWR, 2015). With the preference for off-site mitigation, this causes thousands of

29

wetland acres to be relocated annually. Over time, what effects might aggregate relocation
have on the local ecology and, in turn, local human populations? If wetland losses occur
within one specific land use type (e.g., urban areas), this may reduce the resilience of the
local environment.
This chapter presented a brief history of wetland management in the United States,
with particular attention to specifics of wetland mitigation regulated under the CWA. The
following chapter will present a review of the pertinent scientific literature, including a focus
on ecosystem services, wetland valuations, spatial influences of wetland values, and the
societal impacts of wetland mitigation. This review will examine theoretical and applied
research on wetlands and their ecosystem services. Understanding these two aspects of
wetlands and ecosystem services will contextualize the ensuing analysis.

30

CHAPTER 3: LITERATURE REVIEW
The projected continued loss and degradation of wetlands will reduce the capacity of
wetlands to mitigate impacts and result in further reduction in human well-being.
Millennium Ecosystem Assessment (2005)
Introduction
Wetland mitigation regulated under the Clean Water Act (CWA) permits thousands
of acres of permanent wetland impacts each year (IWR, 2015). To compensate for these
impacts, mitigation creates, enhances, preserves, or restores wetlands, often at a different
location than the wetland impacts. Accordingly, human populations near these sites gain or
lose wetlands and their ecosystem service benefits depending in part on their proximity to
impacts and mitigation projects. However, only a handful of studies have examined spatial
dynamics and environmental equity of wetland loss and gain through CWA mitigation.
The purpose of this chapter is to review relevant literature centered upon
environmental equity in the relocation of wetlands and their ecosystem services. Ecosystem
services play a crucial part in this research; they are the very reasons why human populations
benefit from wetlands and why regulatory agencies across the country spend time and
resources centered on wetland management. Understanding the ecosystem services
framework that environmental managers now use will contextualize how this research
applies this framework to wetland mitigation. This literature review will first examine the
rise to prominence of ecosystem services and efforts to accurately assess their benefits to
human populations. As recognition of ecosystem services’ importance has grown, economic
markets have attempted to integrate ecosystem services into commodity markets. The second
section details how wetland mitigation uses this framework to commodify and trade
ecosystem services within a neoliberal economic system across spatial and temporal scales.

31

In particular, this chapter surveys the challenges of trading complex habitat types like
wetlands across these scales.
The third section examines different methods of wetland valuations. This research
presupposes benefits to human populations living near wetlands. Researchers and economists
use a wide variety of valuation techniques to measure these benefits. While these valuation
techniques can lack both precision and accuracy (Boyer & Polasky, 2004), understanding
different valuation frameworks will increase understanding of the complexity of valuing
wetlands with multiple ecosystem services and the nuance of how wetlands’ position in a
human-influenced landscape can alter these valuations.
The final two sections will examine past research influential in developing methods
for this study. These sections will review previous studies on the equitable distribution of
wetlands within mitigation. Specifically, this section surveys environmental equity within
three parameters: urban-rural equity, socioeconomic equity, and racial equity. The final
section positions this research within these previous studies and articulates the need for
further studies to increase our understanding of wetland mitigation spatial dynamics and its
relation to human populations.
The Rise of Ecosystem Services in Environmental Management
While ecosystem services has risen to prominence within a conceptual framework in
natural resource management, researchers still disagree about a precise definition. The
Millennium Ecosystem Assessment (MEA), the most comprehensive global analysis of
ecosystem services to date, states simply that ecosystem services are the “benefits humans
obtain from ecosystem services,” (MEA, 2005c, p. v). The scope of this definition
encompasses an extraordinary breadth of the earth’s processes. For example, the MEA

32

recognizes that ecosystem services include provisioning services such as food, fiber, water,
and genetic material; regulating services such as climate regulation, water purification,
natural hazard regulation, hydrology regulation and erosion control; cultural services such as
spiritual, recreational, aesthetic and educational benefits; and supporting services such as
nutrient cycling and soil formation. [See Appendix A for a more complete list of ecosystem
services types.]
Increased interest in ecosystem services stems from the recognition that human
activities threaten critical regulating and provisioning ecosystem services requisite for human
health and well-being (MEA, 2005c). Attempting to put a price on these services yields
immense values. A landmark study (Costanza et al., 1997) estimated these services on a
global scale, approximating their values to be in the range of US$16-54 trillion annually. To
put these numbers in context, global gross national product (GNP) at this time was US$18
trillion. As the authors readily admitted, their study presented many assumptions and
conceptual limitations, extrapolating values from existing literature on small-scale ecosystem
valuations to account for the entire planet’s land mass. Nevertheless, the article stimulated
immense interest in ecosystem service valuations and remains the most cited article in the
field of ecological economics (Costanza, Stern, Fisher, He, & Ma, 2004). The high estimates
garnered further interest in capturing the value of earth’s processes and functions in
economic valuations. Integrating the earth’s dynamic ecosystems into an economic
framework remains ecological economists’ primary challenge.
Ecological Economics
Ecological ecologists combine elements of two fields—neoclassical economic theory
and natural systems—but lack of consensus has prevented a unified set of tenets for

33

practitioners to follow study (Bockstael, Freeman, Kopp, Portney, & Smith, 2000; Dorman,
2004; Morino-Saul & Roman, 2012). On one hand, neoclassical economists apply
foundational principles such as cost, benefit, supply and demand, and monetization of goods
and services. Marginal, or incremental values, follow linear relationships when adding units
of value (Heal, 2000). On the other hand, natural systems follow few of these principles.
Instead of using a reductionist approach, ecologists embrace the complexity of
interconnected relationships and feedbacks. Natural systems are also observed to be
nonlinear, with links between human well-being and ecosystems “indirect, displaced in space
and time, and dependent on a number of modifying forces” (MEA, 2005a, p. 2). In general,
ecological-economists accept non-market (i.e., non-monetary) values, such as the intrinsic
value of nature and human rights. They also believe in the non-substitutability of natural
capital, often referred to as “strong sustainability (Merino-Saum & Roman, 2012). In the
context of wetlands, strong sustainability principles regards human-made features that mimic
wetlands as inferior, such as water treatment plants that filter pollutants.
Different approaches within the two disciplines have also divided the field. On one
side, neoclassical economists traditionally value natural resources for the physical products
harvested, such as food or timber. On the contrary, Costanza et al. (1997) recognized this
approach misses the valuation of natural capital, which provides not just a one-time payoff,
but a continual flow of services spanning generations. While attempting to capture the value
of these services, this methodology does not deviate from traditional neo-classical
monetization into a cost-benefit framework. Once these services are properly valued, proper
measures can be taken to account for—or to internalize the negative externalities of—
ecosystem degradation (Merino-Saum & Roman, 2012).

34

Using economic valuations as a tool to conserve natural systems and ecosystem
services has elicited strong debate on its merits (Lele, Springate-Baginski, Lakerveld, Deb, &
Dash, 2013). On one side, ecosystem service valuation proponents argue that the process
corrects for externalities, items or consequences not accounted for in a cost-benefit analysis
of development projects (MEA, 2005b). Properly valuing these services accounts should
reduce negative externalities by accounting for lost services and supports government
regulatory bodies in prioritizing their protection with increasing resource scarcity (Daily et
al., 2000). Wetland mitigation is embedded within these ecosystem service valuation
principles.
A counter perspective argues that far from enhancing environmental protection, these
valuations precipitate their decline (Robertson, 2004). While comprehensive valuations
underpin ecosystems importance, valuations also pave the way toward commodification.
Economic valuations are then framed within the context of neoliberalization and free trade. A
cornerstone of capitalism rests upon the free flow of capital in the forms of goods and
services; mitigation attempts to trade these ecosystem services within this economic system
through environmental trading markets, or ETMs (Heal, 2000). These markets now extend to
carbon sequestration, clean air regulation, and biodiversity preservation (Walker, Brower,
Stephens & Lee 2009) in addition to wetland mitigation.
The Problem with Currency in Trading Wetlands
For a functioning market to be in place, users must be able to trade these goods with
an agreed-upon currency. Therein lies another challenge in setting up a market for ecosystem
services. Salzman and Ruhl (2001) contend that any currency must be traded equally across
type, space, and time. The authors refer to this interchangeability across these scales as

35

fungibility. While this research centers itself on issues of spatial equity, this section addresses
issues of equity that also present themselves when traded across different types and temporal
scales. When examining the currency of natural capital, the ecosystem services themselves
influence their fungibility. With one specific ecosystem service, market users are more likely
to assume a roughly equal trade.
For carbon sequestration, the currency is measured in metric tons of carbon dioxide
(CO2). Specialists can measure how much CO2 a stand of trees sequesters, which can be
protected as an offset for emissions CO2 elsewhere. Notwithstanding other benefits provided
by trees, CO2 sequestration represents a currency that is measured and has approximately the
same value over space and time. With multi-benefit habitats, their fungibility becomes
questionable. For example, research suggests that trading land properties for biodiversity
protection fails to properly understand the ecological and spatial diversity in these land
values (Walker et al., 2009). Wetland mitigation faces similar challenges.
Currencies used for wetland mitigation fail the fungibility test on all three parameters
of type, space, and time. For wetland type, Cowardlin, Carter, Golet, and LaRoe (1979)
classified the county’s diverse wetland types. Each wetland type is a highly-adapted system
that has proved difficult for humans to relocate (Turner, Redmond, & Zedler, 2001). These
types depend on the inputs, or controls, that relate to its geomorphic location, water source
and hydrodynamics (Hruby et al., 2009). In some cases, relocation precludes the possibility
that wetland type can be matched in the diverse landscape of Washington. Teasing apart the
intricate, inter-connected web of relationships and ecosystem services within each wetland
type to determine their value and equal transfer remains an elusive, if not impossible task.
Trading across temporal scales also poses serious limitations due to the lag time that

36

often exists between an impacted wetland and its restoration or creation. During a
development project, wetlands are often damaged before restoration begins, creating a netloss in wetland functions during that time. To compensate, the Washington Department of
Ecology (DOE) requires a higher fee in this scenario to address this time lag (Department of
Ecology, 2013a). Uncertain ecological trajectories also pose a problem for trading wetlands
(Zedler & Calloway, 1999).
Research continues to examine not only if relocating wetlands can be achieved, but
also if practitioners can control with any degree of accuracy the types of ecosystem services
supplied by mitigation efforts. Generally, there is a five or ten-year monitoring period are
required by the DOE and the United States Army Corps of Engineers (ACOE). How
wetlands and their ecosystem services develop after this monitoring and maintenance period
remains poorly understood, as few researchers have conducted longitudinal studies on site
performance. While this research focuses on spatial equity in wetland mitigation, temporal
equity is a relevant issue, particularly with unknown trajectories that affect the flow of
wetland ecosystem services,
The final ingredient for a currency to function is the ability of actors to trade
commodities equally across space. This dimension proves the most problematic for wetland
mitigation and is the primary focus of my research. Before examining spatial components of
wetland ecosystem trading, this study reviews attempts to quantify wetland ecosystem
services through different valuations. As the studies indicate, populations’ spatial proximity
to wetlands and wetlands’ place within the landscape both influence valuations.

37

Wetland Valuations and Ecosystem Services
Per acre, wetland systems such as estuaries, floodplains, and tidal marshes are among
the most valuable and productive habitat types in the world (Costanza, Farber, & Maxwell,
1989). Society places value on wetlands because they provide a wide array of ecosystem
services that benefit populations at localized, regional and planetary scales (Mitsch &
Gooselink, 2000). While these services provide stability for economic well-being, the extent
of human influence around the globe threatens the delivery of many of these services on
which humans depend (MEA, 2005a).
One response to this threat has been to conduct wetland valuation studies. Many
researchers concede these studies do not produce values that can be applied to wetlands in
other locations (Brander, Florax, & Vermaat, 2006). Rather, the threshold for successful
valuations should be if they assist decision makers in choosing policy actions and land use
recommendations (Boyer & Polasky, 2004). As research indicates however, wetland
valuations fail to produce clear findings, with studies often producing wildly variable figures.
This variation stems from two primary causes. First, variation exists in wetlands themselves
due to their performed functions and place within the landscape (Mitsch & Gooselink, 2000).
The second reason rests in the shortcomings of valuation techniques themselves (Boyer &
Polasky, 2004). These valuation methods include contingent valuation, travel cost, hedonic
pricing, production functions, and replacement cost, among others (Brande et al., 2006). See
Appendix B for a full list of valuation methods and descriptions.
The hedonic method measures the value of a good—in this case, wetlands—by using
existing prices as a proxy. Housing and land sales prove to be an effective measure. In urban
areas, proximity to wetlands positively corresponds to wetland values in the three studies
reviewed by Boyer and Polasky (2004). Research in Portland, Oregon showed an increase of
38

$436 in housing prices when houses were moved from 1.6 kilometers to 300 meters to the
nearest wetland (Mahan, Polasky, & Adams, 2000). Studies in rural areas have lacked
conclusive results. Reynolds and Regalado (2002) found wetland type to be a determining
factor whether proximity yielded positive or negative values. The preference of shallow
ponds over forested wetlands, for example, suggests rural residents may prefer hunting and
aesthetic values to other benefits. Loss in agricultural production as well may cause rural
landowners to prefer non-wetlands lands to wetlands. After all, settlers have been converting
wetlands for agricultural purposes since European settlement began in earnest in the 1700’s
(Dahl, 1990). While the hedonic valuation method can show trends, scale limits these studies,
as they measure values only in close proximity to wetlands.
A hedonic study of a three-county area in North Carolina, for example, found that
proximity up to ¾ of a mile to natural wetlands steadily increased property values by roughly
$3,100 (Kaza & BenDor, 2013). When examining restoration projects through the state-run
Ecosystem Enhancement Program (EEP) however, land values varied. Interestingly, land
within .0125 miles of EEP sites had average values $15,500 less than land not in proximity to
EEP sites, suggesting a negative relationship. Given the assumed social benefits of wetlands,
results from this study contradict conventional thinking. A key—and I would argue,
flawed—assumption in this study supposes that these sites were not initially chosen based on
land values, but rather suitability for wetland restoration or preservation. This assumption
repudiates research that indicates private entrepreneurs in wetland mitigation identify profit
as the primary reason for opening a mitigation site (Kaplowitz & Bailey, 2008). As many
EEP sites were bought directly from private mitigation companies whose profit depends on a
difference between the amount recouped for restoration credits and the initial land price,

39

initial land prices may have had everything to do with where wetland mitigation sites are
located.
Replacement cost is another valuation technique that estimates the price to substitute
a good that is no longer available. Many municipal planners use this method in determining
how best to provide safe drinking water. In the Puget Sound, long-protected upper
watersheds in the Cascade Mountain Range and from Mount Rainier provide clean,
consistent drinking water with few filtration costs (Seattle Public Utilities, 2013; Tacoma
Public Utilities, 2008). New York City infamously faced this decision in the 1990’s, with
development in the nearby Catskills Mountains threatening to decrease water quality
standards so that costly filtration plants would be needed. Instead, planners chose to increase
protection for the watershed and associated wetlands, as the $6-8 billion estimated
replacement cost of building purification plants dwarfed the cost of watershed protection to
increase the watershed’s natural water infiltration and purification ecosystem services
(Chichilnisky & Heal, 1998). This large-scale watershed protection preserved existing
wetlands, forests, and their respective ecosystem service benefits for rural populations, while
also supplying the most-populated city in the country with its drinking water.
For fish and wildlife, economists can use production methods analysis to estimate
how much a particular wetland increases fish production. Unfortunately, economists rarely
apply this type of research to urban settings, despite good reasons to do so (Boyer & Polasky,
2004). For example, a growing body of research indicates that polluted urban waterways may
be severely disrupting salmon populations as they enter fresh water system to spawn, the
final stage in their life cycle (Scholz et al., 2011). While isolating the effects of increased
riparian wetlands would be difficult to isolate in an urban setting, applying production

40

methods analysis could improve understanding of the link between degraded urban rivers and
the economic—or production—loss due to a river’s riparian wetland loss.
Researchers have also applied contingent valuation to urban wetlands. Contingent
valuation uses hypothetical values and asks respondents if they would be willing to pay that
amount for a given service or good (Brander et al., 2006). By using different amounts,
researchers average respondents’ preferences to determine willingness-to-pay (WTP). This
methodology has three primary limitations. First, dealing with hypotheticals may not
accurately predict if respondents would actually pay (Boyer & Polasky, 2004). Saying one
would pay $40/acre for restored wetlands is one thing; handing over the money is quite
another. This method may artificially increase WTP. A meta-analysis of wetland valuation,
for example, found that contingent valuation methods yielded greater estimates than other
methodologies (Brander et al., 2006).
Second, incomplete knowledge may also hinder contingent valuation of wetlands.
This ignorance has led to the opposite effect, a decrease in wetland values, as indicated in a
different meta-analysis (Woodward & Wui, 2000). Given wetlands’ complexity and their
dispersed benefits, contingent valuation studies often ask to value just one specific ecosystem
service such as improved water quality. This process also undervalues wetlands (King,
1998).
Third, socioeconomic factors influence perceived valued of wetlands, as indicated by
(Brander et al., 2006). If mitigation relocates wetlands spatially, this will likely result in a
population with different socioeconomic characteristics. In turn, these socioeconomic
differences will attach different perceived values to wetlands. For example, a low-income

41

community may produce starkly different estimates than an upper-middle class community,
despite performing the same functions.
While instructive in noting public perceptions, contingent valuation’s limitations
restrict its application in wetland relocation. Noting the wide variability, the NRC (2001)
recommended mitigation guidelines absent from valuation studies based on human
perceptions. These studies lack of precision and accuracy prevents their integration in the
mitigation process. Nevertheless, these studies point to differences in perceptions of wetlands
and wetland uses among populations. These differences are important to deliberate when
examining environmental equity occurring within the spatial relocation of wetlands and their
ecosystem services because of wetland mitigation.
Spatial Influences on Wetland Values
Proximity to wetlands influences perceived values of these systems among human
populations (Brander et al., 2006). Known as spatial discounting, this theory states that
resources located farther away from populations will decrease its perceived value (Perrings &
Hannon, 2001). Accordingly, resources located in close proximity to human populations are
more highly valued. Wetland mitigation, therefore, has the power to increase or decrease the
perceived value depending on wetlands relative proximity to human populations. Research
by Manuel (2003) indicated that small, urban wetlands are valued for aesthetics and
contribute to perceptions of place among local populations. This research indicates that local
residents value cultural ecosystem services—aesthetics, education, recreational, spiritual, and
the landscape as a sense of place—more than regulating and provisioning services whose
benefits are more dispersed and less understood by the surrounding population.
Manuel (2003) recognized that the size of wetlands influence local perceptions of

42

wetlands. With small wetlands, the author recognizes their size is unlikely to galvanize a
community over potential impacts. As Figure 6 indicates, the vast majority of projects
requiring mitigation are less than one-tenth of an acre.

Figure 6. Acre range of national wetland impacts from 2010-2014 (IWR, 2015).
As wetland mitigation approaches one-half century in the United States, these small
impacts will likely continue. These small-scale impacts add up to thousands of wetland
impacts every year. From 2010-2014, regulatory agencies granted an average of roughly
8,000 acres of permanents impacts to non-tidal wetlands nationwide (IWR, 2015).
Recent studies have taken a broader and more critical look at spatial dynamics of
wetland mitigation that may affect local populations, yet this remains an understudied area of
research. King and Herbert (1997) were the first researchers to analyze wetland impact sites
and mitigation sites in regards to human populations. This first study of Florida Department
43

of Transportation (DOT) sites recognized a strong urban-rural shift in wetland mitigation
placement. Nearly ten years later, Ruhl and Salzman (2006) continued research on Florida
wetland mitigation sites and incorporated population densities into their analysis. Overlaying
population density with site location, their analysis found that mitigation bank areas had low
population densities while impact sites had much higher densities. BenDor, Bruzovic and
Pallathucheril (2007) studied the impacts of wetland mitigation in four counties in the greater
Chicago area, assessing if mitigation type and size correlated with population densities.
Mitigation banking in particular moved wetlands along a strong urban-rural gradient. This
study also examined demographic data to note environmental equity within the types of
people living near impact and mitigation sites. These researchers analyzed differences in
ethnic and racial percentages between impact and mitigation sites to measure racial equity.
To examine socioeconomic equity, the researchers measured average household income and
households in poverty. These types of analysis assessing the potential impacts on human
populations of a widespread government program were what President Clinton intended
when he signed an executive order to address environmental justice within government
programs.
Environmental Equity within Wetland Mitigation
Executive Order No. 12898 (1994) from President Clinton asked that “each Federal
agency shall make achieving environmental justice part of its mission by identifying and
addressing, as appropriate, disproportionately high and adverse human health or
environmental effects of its programs, policies, and activities on minority populations and
low-income populations in the United States” (p. 1). Nevertheless, current regulatory

44

guidelines do not address if cumulative impacts of wetland mitigation augments social
disparity (BenDor, Brozovic & Pallathucheril, 2008).
Researchers and regulatory agencies have examined ecological functions in relation
to human populations. Due to the high levels of human perturbations that may augment
project costs, regulatory agencies recommend mitigation sites away from urban areas (NRC,
2001). Washington State, in turn, has followed these recommendations by using watershed
boundaries as parameters for site selection. Hruby et al. (2009) with the Washington
Department of Ecology (DOE) specify that using a watershed approach does not set any
limits on the distance of wetland relocation, as watershed basins can cover large areas.
However, altered hydrologic regimes found disproportionately in urban areas may cause
increased flooding, eutrophication of local waters, poor water quality, bank erosion and loss
of habitat (p. 14).
Current regulatory recommendations for wetland placement fail to take human
populations into account (BenDor et al., 2008). Wetlands likewise have thus far failed to
garner the attention that the equitable distribution of parks and other green spaces have. In
practice, site selection with parks and wetlands starkly contrast each other. While
environmental equity advocates seek to maximize urban green spaces for underserved
communities (Jennings et al., 2012), the low fungibility of wetlands prevents adopting this
same framework; previous land legacies may lead to poor site wetland functioning. For this
reason, site selection for wetlands recommends low population densities to achieve
ecological maximum benefits (NRC, 2001).
Nevertheless, the previous studies that examined urban-rural equity within wetland
mitigation also examined socioeconomic and racial equity. Results do not indicate any clear

45

trends across the studies. BenDor and Stewart (2011) found a movement from more White
and higher income populations to populations with higher percentages of minorities and
lower incomes in North Carolina. These findings were consistent with earlier work from the
same author, BenDor et al. (2007), which found more White populations and higher incomes
population living near impact sites. However, Brass (2009) found different findings in
Oregon, with populations living near mitigation banks having higher incomes and higher
percentages of White populations. The average distance between the sites also varied,
ranging from 13.5-31.2 miles. The researchers hypothesized that relocation distance may be
attributed to whether mitigation is managed at the local, state, or federal level (BenDor &
Steward, 2011). While these studies captured valuable data, this limited range represents a
significant data gap nationwide in wetland mitigation. Further studies conducted at the state
and county level will help fill the gap in spatially analyzing mitigation trends to advance our
understanding of environmental equity in wetland relocation.
Need for Further Research
With the small number of geographic areas analyzed for wetland mitigation trends,
replicating methodologies from previous research will enhance baseline spatial and
socioeconomic data for wetland mitigation. Choosing the paired t-test method by BenDor,
Brozovic and Pallathucheril (2007) offers a succinct analysis of differences between
mitigation and impact sites. Using data from King, Snohomish and Clark counties—three
counties in Washington State that have pronounced urban-rural gradients—this research will
examine if wetlands are relocated from urban-to-rural environments. In addition, this
research will examine racial and socioeconomic equity within the distribution of wetland
impact and mitigation sites.

46

No one has yet to complete this type of study in Washington State. As ecosystem
service markets gain acceptance and expand to account for adverse environmental impacts,
regulators and researchers must examine how this system affects both local ecology and
human populations. This research examines how these three Washington counties are
administering their wetland mitigation plans and if this regulatory process is being
implemented equitably to human populations across spatial scales.
While the regulatory framework of the Clean Water Act (CWA) has enabled impacts
to wetlands to occur for decades, determining the spatial relationship between wetland losses
and gains has received little attention. Examining the spatial distribution of wetlands is
important because of wetlands’ many localized benefits for human populations. In the
urbanized environment of the Western Washington, flood mitigation, storm abatement, and
improved water quality all benefit the ecological health of sensitive aquatic resources.
Wetlands also provide food and fiber, regulate temperature, and provide aesthetics (MES,
2005). Mitsch and Gooselink (2000) recognized that urban areas and their populations
benefit greatly from these services due to the relative scarcity of wetlands. Despite the
increased marginal value of wetlands in urban areas, guidelines have directed site selection
away from urban areas (NRC, 2001). The NRC made these recommendations based on the
relative low success rate of urban mitigation sites, noting that previous disturbances in urban
areas often alter soil composition or hydrology to prevent wetland conditions from reestablishing. Perturbations such as invasive plant colonization also increase the likelihood of
continued maintenance costs. As conceptualized by Mitsch and Gooselink (2000), wetland
functions may have a tipping point, where previous land use legacies or continuing

47

perturbations may cause wetland function to cross a tipping point and overwhelm their
functional viability.
Therein lies the tension within this urban versus rural wetland mitigation site
selection. On one hand, urban environments stand to benefit the most from wetlands and their
ecosystem services. On the other hand, ecological conditions, land value prices, and the
availability of land all favor mitigation site selection in rural areas. By comparing population
densities surrounding impact and mitigation sites, this study will examine if wetlands are
being relocated along an urban-rural gradient in Western Washington. By comparing
socioeconomic and racial composition surrounding impact and mitigation sites, this study
examines how different socioeconomic and racial groups are distributed near impact and
mitigation sites. Results from this study will assist in closing the data gap in understanding
how wetland mitigation relates to human populations.

48

CHAPTER 4: METHODS
No actor in the [mitigation] banking process takes steps that would allow us to test the policy
implications of the phenomenon—i.e., tracks the redistribution of wetlands, estimates the
effects thereof on ecosystem service values, notifies the affected public, and provides
opportunity for public input.
Ruhl & Salzman (2006)
Introduction
This research surveys the spatial and socioeconomic characteristics between wetland
impact sites and mitigation sites within three counties in Western Washington. Specifically,
this research asks if wetland mitigation relocates wetlands and their ecosystem services from
urban to rural environments. Closely linked to the spatial analysis is understanding what
types of populations are losing and gaining wetlands. While regulatory agencies emphasize
ecological functioning of wetlands, this framework overlooks the distribution of wetlands
among human populations. Environmental justice literature has linked green spaces to
improved human health (Jennings et al, 2012). This research extends this logic to wetlands to
examine how site selection may be equitably or inequitably redistributing wetlands. While
only a handful of studies have looked at socioeconomic characteristics of populations, results
have been mixed across sites. This research aims to increase the knowledge of spatial
dynamics in wetland mitigation projects and examine how wetland relocation may affect
human populations.
This chapter describes the methods used to survey the spatial distribution and
socioeconomic attributes of populations affected by wetland mitigation. The organization
follows the chronological sequence in which the research was conducted. First, this section
describes the criteria for choosing the study area. Second, the details of data acquisition from
state, federal and county agencies are outlined. Third, this study establishes rationale for the
49

many decisions of spatial scale that were made within this research. Next, methods of the
geospatial analysis are described. These analyses include determining the urban-rural
gradient, the relocation distance between impact and mitigation sites, and the socioeconomic
characteristics of the populations surrounding these sites. After the geospatial analysis
methods are defined, the challenges of acquiring data are reported. The final section outlines
the limitations within the dataset.
Choosing the Study Area
Three counties were selected from Western Washington to be included in the study
area. These counties are King, Snohomish and Clark. King and Snohomish are adjacent to
each other, are located within the Puget Sound Basin within the Seattle greater area. Clark
County borders the Columbia River within the Portland, Oregon greater metro area. Both
Seattle and Portland have undergone significant population growth in recent decades; this
growth has influenced changes in land use changes throughout these counties. See Figure 7
for map of the study area.

50

Figure 7. Three-county study area. This map illustrates the study area within western
Washington and western United States.

51

These counties were selected based on two primary criteria. First, each county had a
distinct urban-to-rural gradient, which was a necessary condition for the research question
posed in this thesis. Second, these counties possessed a large number of wetland mitigation
projects, also a requisite for this research. In the past fifteen years, these counties have
experienced significant growth with increasing urbanization. See Table 1 for population
trends in these counties from 2000-2015.
Table 1. Population estimates and growth rates from 2000 to 2015 in the three-county study
area (United States Census Bureau, 2017).
Table 1
Population Increases by County
Census Year

2000

2015

% growth from
2000 to 2015

Clark

345,238

442,800

28.26

King

1,737,034

2,045,756

17.77

606,024

746,653

23.21

Snohomish

According to the Puget Sound Regional Council (2016), 95% of new housing in 2013 was
centered in cities and urban areas. These findings are consistent with national trends that
indicate robust urban growth while rural populations remain flat (EPA, 2017b). Figure 8
displays population trends in the United States over the past 200 years. In this research, the
greater Seattle and Portland metro areas are these urban centers influencing regional growth.
Predictably, this growth led to a higher rate of wetland mitigation permit requested under the
Clean Water Act (CWA) due to development and growth pressure. For example, of the 286
statewide mitigation sites for Washington State Department of Transportation (WSDOT)

52

transportation projects, 178 were found within the three-county study area. Selecting these
counties enabled an adequate sample size from which to draw.

Figure 8. Urbanization and population trends in the United States (EPA, 2017b).
The study area chosen includes Clark County, the only county not within the Puget
Sound Basin. While not geographically adjacent to the other counties, Clark County has in
fact been undergoing the fastest growth in the state. This growth can largely be attributed to
the expanding Portland, Oregon metro area, which borders Clark County to the south. Clark
County also has unique features that enrich the data set. For example, it is the only county
within the study area with a mitigation bank for impacts to a river system (e.g. Columbia
River). In sum, Clark County exhibited a large number of mitigation projects, an urbanizing
population, and a unique landscape setting to examine wetland mitigation.
Wetland mitigation projects often cross county lines. As long as the mitigation site
was located within the three-county study area, this study analyzed cross-county mitigation
projects, regardless of whether its corresponding impact fell within the study area. For
example, the Columbia River Wetland Mitigation Bank’s service area includes three
53

counties, even though the actual bank is located in Clark County. Excluding sites that crossed
over into different counties would have limited this study’s full ability to assess spatial
relocation within wetland mitigation. Since this research is interested in the geographic
relocation of wetlands, including sites outside the study area was pertinent information and
allowed for a more complete spatial analysis.
Acquiring Data
This analysis required three distinct types of data. The first type of data was
information on wetland mitigation data. In order to complete the spatial analysis, coordinate
information for both impact and mitigation sites was requisite. In addition, the mitigation
approach (mitigation bank, ILF, or PRM) enabled this research to examine the characteristics
of each approach. Socioeconomic data was the second type of data. The United States Census
Bureau and ESRI, the computer mapping and spatial data analytics software company,
maintain detailed spatial socioeconomic data. Once sites were geospatially located, these
sources could provide site data on population, economic, and ethnic characteristics. Map
layers were the third data type that enabled me to properly display the data. These layers
included county boundaries, urban growth areas, and water bodies.
Acquiring wetland mitigation data involved contacting numerous state and county
agencies involved with wetland mitigation. My primary source of data came from the
Washington State Department of Ecology (DOE). Upon request, DOE staff at their
headquarters in Lacey, WA granted access to their mitigation files to compile appropriate
geospatial and site-specific data. King County Mitigation Reserves Program also provided
GIS files with georeferenced locations. WSDOT and the ACOE both provided data, but only
included mitigation site data. I was not able to track down impact site data for either of these

54

sources. Requests were sent to Clark County and Snohomish County for county mitigation
files, but these requests did not yield any site information.
In total, this research analyzed 139 wetland mitigation projects. For each project,
there is one impact site and one mitigation site, referred to as a paired project. Mitigation
banks accounted for 108—or 78%—of the 139 paired projects. A mitigation banking ledger
contains addresses or some reference to physical locations to their impact sites. The DOE had
a partially completed GIS layer for their bank impacts. For the remaining sites, addresses
were georeferenced with Google Maps to obtain coordinate data. A small number of ledgers
contained parcel numbers, which were georeferenced using county websites. If location
could not reasonably be determined, sites were omitted. This often occurred with utility
companies that work in areas without street addresses. As each purchaser of wetland credits
provides the site location, a high diversity of address types were listed.
For socioeconomic data, the Census Bureau and ESRI were the primary sources of
data. The Census Bureau completes a nationwide census every ten years that tracks
demographic data. The last census was completed in 2010. Data from the ten-year census
provide the public with detailed sociodemographic information. In addition, they track
demographic changes using the American Community Survey, which assesses demographic
patterns in between the ten-year census. Using three different area scales—census blocks,
block groups, and census tracts—Census Bureau data can be spatially analyzed using
geographic information systems (GIS). In addition to storing spatial census data, ESRI tracks
socioeconomic data of their own that they make available through their online platform,
ArcGIS Online. While ESRI calculates new layers for each year, these estimates are based
off initial 2010 census data.

55

Importance of Scale
Measuring socioeconomic and population characteristics near impact and mitigation
sites requires determining an appropriate scale. Given the fact that different wetland
ecosystem services benefit populations at different scales, determining scale presented
challenges (Mitsch & Gooselink, 2000). I used ArcGIS to create a ¾-mile buffer around each
site. This methodology closely follows Brass (2009) which utilized a 1-mile buffer to study
mitigation banks in Oregon. This buffer should not be interpreted as an agreed-upon
perimeter in which a population therein achieves maximum benefits of wetland ecosystem
services. Rather, this buffer provides the means to examine indicators near the immediate
surroundings of impact and mitigation sites.
The scale used in this research is smaller than previous studies analyzing
socioeconomic and population density characteristics of wetland mitigation. In previous
studies, census tracts have been used to examine socioeconomic characteristics (BenDor et
al, 2007; BenDor & Stewart, 2011). Census tracts with low population densities however can
have large surface areas. Thus, census tracts may not properly evaluate the population near
the site. The tradeoff with the methodology used in this study, however, was low populations
at some sites. Eight impact sites had no population within their ¾-mile buffer. Since there
was no population to draw from, these sites were excluded from differences in racial equity
analyses.
Given the small acreage of most wetland impacts, these sites were saved as a point
feature in ArcMap. Wetland mitigation banks in this study area, however, are 100-225 acre
sites. To address these larger sites, mitigation banks were saved as polygon features. Saving

56

these banks as a point feature may have artificially lowered the totals, since part of the buffer
would have been within the mitigation bank itself.
Spatial Analysis
This section describes the methodology used after impact and mitigation data and
coordinates were collected. Figure 9 presents a simplified workflow of this analysis. First,
site data were transferred into ArcMap. Since these sites were saved with an exact
coordinate, a ¾-mile buffer was added to each site using the Buffer tool. As was previously
stated, mitigation banks were saved as a polygon feature class, but the same Buffer tool was
used to create a ¾-mile buffer. After the buffering step, GIS layers were transferred into
ArcGIS Online, where data enrichment from Census data were applied.

57

58

59

Figure 9. Workflow of spatial analysis after obtaining and organizing wetland mitigation
data.
Data enrichment by ArcGIS Online enabled this research to procure site-specific data
for each ¾-mile buffer area for a range of population and socioeconomic variables. Enriched
data included the following:
1. Urban-rural variables: Population density, percent developed, relocation
distance
2. Economic variables: median household income, number of households in
poverty, median home value
3. Racial/Ethnic variables: Included populations of Whites (non-Hispanic),
Hispanic, Black, Asian, Native American, Pacific Islander, Minority
Population, and Diversity Index
Data that were analyzed independent of the data enrichment services on ArcGIS
Online included whether a site was located within urban growth areas (UGA) and the average
distance between impact and mitigation sites.
Urban-Rural Equity
Three variables were used to assess whether wetland mitigation relocates wetlands
from urban to rural areas: population density, presence within a UGA, and the percent
developed land according to the United States Geological Survey (USGS) National Land
Cover Database (NLCD). The NLCD defines four varying levels of development within their
definition. For the lowest level of development, impervious surfaces average less than 20%.
For high intensity development, on the other hand, impervious surfaces average 80-100%
(USGS, 2017). Since this research was interested in human populations affected by wetland
mitigation, comparing population densities was the primary indicator to assess differences in

60

impact and mitigation sites. Nevertheless, the suite of three indicators provide a more
complete picture of wetland relocation.
Population density and percent developed were calculated using enrichment data.
Presence within a UGA was calculated using the “Select by Location” function on ArcMap,
using the UGA layer to select all sites contained within UGA. These sites were coded with a
1. Sites outside of UGA were given a 0.
The average relocation distance between impact and mitigation site is another spatial
variable. The average relocation context adds context to how and where wetlands are being
relocated through mitigation practices. Calculating distance was achieved by two different
methods. For mitigation banks that exhibited a many-to-one mitigation scheme, the “Point
Distance” tool was used. This calculated distance in feet between the mitigation bank and
impact site. A separate column was created to convert this value to miles, which dividing all
values by 5280, the number of feet in a mile. For PRM sites, the “Measure” function was
used to mileage between sites. King County had already calculated distances between their
ILF sites.
Economic Equity
Three variables were used to calculate the economic status of populations, including median
household income, number of households in poverty, and median home value. These were all
calculated using data enrichment, but their sources came from a variety of sources. Median
household income and median home value came from 2016 ESRI data. Households below
the poverty level came from 2010-2014 American Community Survey.

61

Racial Composition
With 2010 Census data, population estimates from multiple racial groups were used
to calculate racial composition within wetland mitigation. These categories included the
following: White (non-Hispanic), Hispanic, Asian, American Indian Pacific Islander, Black,
and minority. Hispanics are not defined as a race, compared with the others. Rather, the
Hispanic designation is defined as an ethnicity. Minority populations are defined as
containing any of the following categories: Black, American Indian, Asian, Pacific Islander,
Other, and Two or More races. Using enrichment data, total inhabitants from each category
was calculated. In order to adjust for different population totals, totals for each racial
category were divided by the total population within each buffer, which created the
percentage of each racial group.
Measuring Difference in Means and Statistical Significance
Each impact and mitigation site had values for the aforementioned variables within
the ¾-mile buffer surrounding each site. ArcGIS Online used Census data to extrapolate the
values within each buffer. Measuring the difference between impact site values and
mitigation site values provided the requisite data to determine environmental equity within
wetland mitigation. The null hypothesis in this scenario is the difference in means should be
close to zero, with averages across the sites balancing each other.
After enrichment, the data tables were brought into Microsoft Excel. In addition to
the site name and coordinates, two features increased the facility of analyzing each site. First,
each paired project was given a unique site code. Second, a column contained an “I” or “M,”
indicating whether it was an impact or mitigation site. This organization enabled easy
reference during analysis.

62

.
For each variable, the impact site value was subtracted from the mitigation site value.
Thus, positive values indicated higher values at impact sites. Negative values indicated
higher values at mitigation sites. The differences from each impact and mitigation site were
added and then divided by the total number of sites. Finally, using JMP 12.1 statistical
software, differences in means between impact and mitigation sites were analyzed using a
paired t-test. Since this research aims to increase overall understating populations living near
impact and mitigation sits, this research used a two-tailed test to note differences in either
direction.
Limitations
Several limitations need to be acknowledged that may limit generalizing results.
These limitations include a sample heavily weighted by wetland mitigation banking, treating
every mitigation project the same, regardless of impact size, and receiving wetland benefits
far from its geographic location.
Despite efforts to collect a balanced sample between mitigation approaches, over 77%
of paired mitigation projects were from wetland mitigation banks. The Snohomish mitigation
bank alone accounts for over 1/3 of all paired projects. This reliance on mitigation banking
limits the ability to generalze findings for ILF and PRM projects. This fact also needs to be
considered when reviewing the overall findings. While ILF and PRM projects were included
in the overal sample, the overall findings should not be misconstrued as broadly representing
wetland mitigaiton in general. To address this limitation, findings are presented by mitigation
approach as well as individual mitigation bank to note the differences.

63

The extent of wetland benefits also limits this study. While a ¾-mile, buffer was used
to examine populations living near areas of wetland gain and wetland loss, the range of
benefits may expand or shrink depending on the ecosystem service provided. Thus, while this
research informs of population differences in the immediate vicinities surrounding wetland
loss and gain, this methodology simplifies the nuance of wetland ecosystem service benefits
at spatial scales.
The methods employed in this study also treats each mitigation project as equal,
despite different magnitudes of wetland impacts. While most wetland impacts are under 0.1
acres, there is a wide variance of impact and restoration acreage. The relative impacts to
surrounding populations will vary largely due to the size or acreage. Hence, a wetland impact
of .05 acres may have a small impact on the population surrounding the wetland damage. On
the other hand, the ecosystem service benefits from a 200-acre mitigaiton bank would be
substantially higher. While this variance is substantial, this research does not factor in
wetland impact sizes.
One of the primary challenges with wetland mitigation analysis was finding
information that links impact sites and mitigation sites. This challenge was not an isolated
challenge for a thesis project. Rather, this problem continues to be an institutional limitation
within the state and federal regulatory agencies. Some of the institutional challenges that
resulted in a small sample size include the following:


Agency personal do not prioritize keeping an updated database of wetland mitigation
projects. As individuals are in charge of reviewing dozens—sometimes hundreds—of
mitigation sites, staff prioritize single site evaluation. Little time remains for updating
a database or spreadsheet. Multiple members from the DOE expressed interest in

64

maintaining a more reliable and up-to-date database both for internal and external
use.


Coordinate information for wetland impact and mitigation sites may not be accurate.
The DOE maintains a publicly accessible geographic information system (GIS) layer
entitled “Facility/Site” that attempts to display all permitted site locations across the
state, including those for wetland mitigation. Unfortunately, some of these sites are
not accurately georeferenced, or linked to a geographic place from a coordinate
system (Georeference, n.d.). Instead of taking coordinates with a global positioning
system (GPS), the township and range code is often used instead. When inputted into
the GIS layer, the site location will display the center—or in GIS parlance, centroid—
of the township. This prevents accurately mapping site locations. Research conducted
in Oregon also identified this problem (Brass, 2009). In instances where the centroid
was used, a maximum of 0.7 mile error was calculated.



Data sets copy coordinate information for impact and mitigation site, even when
occurring off-site. Through a Freedom of Information Act (FOIA) request, I obtained
a data set containing all mitigation projects overseen by the ACOE over a five-year
period within the three-county study area. This file contained 776 individual rows of
separate wetland impacts. While projects were separated between impacts and
mitigation sties with coordinates for both, upon further examination, the coordinates
listed for impact and mitigation sites were the same, regardless if the mitigation
occurred on- or off-site. The remaining data were remarkably complete, signifying a
unified dataset with compiled indicators such as acres affected, credits used, wetland
classification, project start date, and type of project initiated. With accurate

65

coordinate data for both impact and mitigation sites, this database could be used by
developers to know if wetland credits are available in a mitigation bank service area
that many have advocated for previously (Martin and Brumbaugh, 2011). This act
would also improve the ability to analyze spatial patterns of wetland mitigation.


Linking impact sites and mitigation sites for permittee-responsible mitigation (PRM)
remains vexing. This research required linking impact sites and mitigation sites.
However, the current regulatory framework does not incentivize coupling these two
areas. Rather, an initial assessment by regulatory agencies at the impact site
determines the amount of mitigation required. From there, developers decide an
appropriate course of action for mitigation. As mitigation plans develop, impact site
information is often not included. During personal conversations with both permittees
and regulators, this challenge was consistently acknowledged. These challenges are
not as acute for mitigation banks and in-lieu fee (ILF) programs. These two types of
mitigation benefit from having just one large mitigation site. Owners of mitigation
banks keep a running ledger of impact sites to track how many credits can be released
at a given time.
A final limitation of this data recognizes on-site mitigation. While my research

question addresses off-site mitigation, permittees can also mitigate for wetland impacts onsite. With disparate data, there was no easy method to assess how many impacts are
mitigated on-site and how many are relocated off-site. In mitigation banking for example, the
credit-debit system is utilized only for off-site relocation. A percentage of mitigation may (or
may not) be conducted on-site, but the credit/debit system does not capture these actions.
Recent directives from federal and state authorities recommend off-site mitigation however

66

(Hruby et al., 2009; ACOE & EPA, 2008). Without having this information, no conclusions
can be drawn for on versus off site mitigation. DOE staff acknowledged this difficulty in
compiling complete information that assesses the totality of wetland mitigation from
beginning to end (Kate Thompson personal correspondence, Feb 17, 2017).

67

CHAPTER 5: RESULTS
Urban development stresses the landscape and may compromise environmental quality.
Since some communities are disproportionately impacted by changes in land use and land
cover, understanding the environmental justice implications of changing the landscape is
important. Likewise, the additive effects of degraded landscapes and decreased
environmental quality have human health implications.
Jennings et al. (2012)
Summary
For this analysis, 139 paired impact-mitigation projects within the three-county study
area were analyzed. These sites included wetland banks, In-Lieu Fee (ILF), and off-site
Permittee-Responsible Mitigation (PRM). First, results are listed for the entire 139 paired
projects. Second, results are listed for individual mitigation bank programs, ILF, and PRM
projects. The small sample size in many of the individual programs increases the likelihood
the sample mean deviates from the population mean, decreasing the probability of finding
statistically significant trends at the 0.05 level. However, the variability and spatial context of
these programs warrant their own analysis. This section presents the findings of the analysis.
The proceeding Discussion chapter explores the implications of these results.
Complete Study Area
Results indicate that over the three-county study area, mitigation relocates wetlands
along a pronounced urban-rural gradient, from lower to higher income neighborhoods, and
from sites with a higher percentage of White populations to sites with higher percentages of
minority populations.
For urban-rural indicators, population densities are 926 people higher per square mile
near impact sites than mitigation sites. Impact sites are 8.5% more developed than mitigation
sites. On average, wetland mitigation relocates wetlands and their ecosystem services 11.3

68

miles. These findings confirm previous research that wetland mitigation does in fact relocate
wetlands away from high-density population areas to less-populated locations.
For economic indicators, median income is on average, $9,032 higher at mitigation
sites. The median home value was $43,250 higher at mitigation sites. These findings were
consistent with research by Brass (2009), who found more affluent populations near
mitigation sites. On average, there are 26 more households in poverty near impact sites than
mitigation sites. However, households in poverty were not weighted with population
densities.
Indicators on racial equity indicate that minority populations are 3.7% higher near
mitigation sites. Conversely, White populations are 3.7% higher at impact sites. For
individual racial categories, Black populations are, on average, 3.4% higher at mitigation
sites. Native American populations are, on average, 0.3% higher at mitigation sites. Pacific
Island populations are, on average, 0.2% higher near mitigation sites. No significant
differences are noted in Asian or Hispanic populations between impact and mitigation sites.
These summary statistics can be found in Table 2. Since the mean difference was
calculated by subtracting impact site values from mitigation site values, positive numbers
indicate impact sites have a greater mean average. Negative values indicate mitigation sites
having a greater mean average.

69

Table 2. Summary statistics of study area. Positive values signify greater values at impact
sites. Negative values indicate greater values at mitigation sites.
Three-County Study Area (n=139)
Mean
Difference

Indicator
Population Density
Developed Land
Median Household Income
Median Home Value
Households in Poverty
White population
Minority population
African-American
American Indian
Asian population
Pacific Islander
Hispanic population

926
8.5%
-$9,032
-$43,250
26
3.7%
-3.7%
-3.4%
-0.3%
0.6%
-0.2%
-0.5%

P-value
0.0001
0.0001
0.0027
0.0067
0.0299
0.0053
0.0053
0.0001
0.0188
0.2838
0.0335
0.4182

Urban Growth Areas (UGA) provide a signpost for developed landscapes. For impact
locations, 89 out of 139 sites were located within UGA. With the presence of mitigation
banks and ILF programs that represent many-to-one mitigation, there were fewer mitigation
sites. However, each paired project was considered one impact site and one mitigation sites.
For mitigation sites, 45 out of 139 were located within UGAs.
The average relocation distance varied by approach and location. Over the complete
study area, the average relocation distance was 11.3 miles. Mitigation banks and ILF
programs showed greater relocation distances than PRM projects. However, the uneven
sample between different mitigation approaches prevent any strong conclusions on variation
between them. See Table 3 for a summary of average relocation distances between mitigation
approach and locations.

70

Table 3. Average relocation distance by approach and location.

NAME
All Sites
Col River
EFL
ILF
PRM
Skykomish
Snohomish
Springbrook

Avg. Relocation
Distance (In
Miles)
11.3
6.7
11.7
13.0
1.5
16.2
15.3
3.4

Mitigation
Approach
All
Mitigation Bank
Mitigation Bank
ILF
PRM
Mitigation Bank
Mitigation Bank
Mitigation Bank

Findings by Approach and Location
The following section displays summary findings for the individual mitigation banks,
ILF, and PRM programs within the study area. Since many of the sample sizes and
populations numbers for some racial categories are very small, only White and minority
populations are listed in the summary data. These findings are also displayed geospatially.
See Figures 10-15.

71

Mitigation Bank Summary Results

Figure 10. Map of Snohomish mitigation bank impact and mitigation sites.

72

Table 4. Summary statistics for Snohomish mitigation bank.

Snohomish Mitigation Bank (n=52)

Indicator
Population Density
Developed Land
Median Household
Income
Median Home Value
Households in Poverty
White population
Minority population

Mean
PDifference
value
1,336 0.0001
14.2 0.0001
-$28,097
-$180,806
25.5
-3.6%
3.6%

0.0001
0.0001
0.0001
0.0024
0.0024

73

Figure 11. Map of Springbrook mitigation bank impact and mitigation sites.

74

Table 5. Summary statistics for Springbrook mitigation bank.
Springbrook (n=8)
Indicator
Mean Difference
P-value
Population Density
2,905
0.0170
Developed Land
-9.5
0.2720
Median Household Income
-$34,973
0.0104
Median Home Value
-$53,348
0.0276
Households in Poverty
218.1
0.0921
White population
37.5%
0.7950
Minority population
50.0%
0.7950

75

Figure 12. Map of Skykomish mitigation bank impact and mitigation sites.

76

Table 6. Summary statistics for Skykomish mitigation bank.
Skykomish (n=25)
Indicator
Mean Difference P-value
Population Density
122
0.4181
Developed Land
16.5
0.0001
Median Household Income
-$5,723
0.6227
Median Home Value
$45,540
0.0686
Households in Poverty
-24.6
0.1002
White population
17.8%
0.0001
Minority population
-17.8%
0.0001

77

Figure 13. Map of Columbia River mitigation bank impact and mitigation sites.

78

Table 7. Summary statistics for Columbia River mitigation bank.
Columbia River (n=14)
Indicator
Mean Difference
P-value
Population Density
669
0.0958
Developed Land
-13.1
0.0459
Median Household Income
31600
0.0301
Median Home Value
210838
0.0047
Households in Poverty
42.2
0.0118
White population
13.8%
0.0060
Minority population
19.4%
0.0060

79

Figure 14. Map of East Fort Lewis mitigation bank impact and mitigation sites.
Table 8. Summary statistics for East Fort Lewis mitigation bank.
EFL (n=9)
Indicator
Mean Difference
P-value
Population Density
904
0.2055
Developed Land
14.6
0.0639
Median Household Income
$11,309
0.0290
Median Home Value
5589
0.8397
Households in Poverty
10.9
0.4152
White population
-4.1%
0.0186
Minority population
4.1%
0.0186
80

In-Lieu Fee Summary Results

Figure 15. Map of King County In-Lieu Fee Program’s impact and mitigation sites.

81

Table 9. Summary statistics for King County In-Lieu Fee program.
ILF (n=20)
Indicator
Mean Difference
Population Density
329
Developed Land
7.8
Median Household Income
$14,101
Median Home Value
$460,907
Households in Poverty
47.8
White population
7.2%
Minority population
-7.2%

P-value
0.5405
0.1727
0.0512
0.1888
0.3434
0.2070
0.2070

Permittee-Responsible Mitigation Summary Results
Minimal average distances of 1.5 miles between impact and mitigation sites posed a
challenge to represent PRM programs cartographically. For this reason, the map is omitted.
See Table 10 below for the summary table.
Table 10. Summary statistics for Permittee-Responsible Mitigation.
PRM (n=11)
Indicator
Population Density
Developed Land
Median Household Income
Median Home Value
Households in Poverty
White population
Minority population

Mean
Difference P-value
805
0.0941
0.2
0.0001
-$16,561
0.1360
-$13,252
0.1690
113.3
0.0791
-2.5%
0.2168
2.5%
0.2168

82

CHAPTER 6: DISCUSSION
Planners and legislators will not respond to the impacts of individual losses that they
perceive to be small and insignificant, but they may respond to the collective value, and the
impact of cumulative loss, of many small natural amenity environments in the urban
landscape.
Manuel (2003)
Introduction
Results from this study strengthen research that indicates wetland mitigation
relocates wetlands and their ecosystem services along urban-to-rural gradients. This study
supports previous research that has examined the landscape effects of wetland mitigation
(BenDor, Brozovic, & Pallathucheril, 2007; BenDor & Stewart, 2011; King & Herbert, 1997;
& Ruhl & Salzman, 2006). In relation to socioeconomic equity, average household incomes
were nearly $9,032 higher near mitigation sites. However, these results were not uniform;
two mitigation banks and the King County ILF program all had higher incomes near impact
sites. In examining racial equity, mitigation relocated wetlands to areas with higher
percentages of minority populations.
Urban-Rural Equity
The results in this study confirm previous studies finding that wetland mitigation
favors mitigation site selection in less densely populated areas. For example, Ruhl and
Salzman (2006) found wetland migration along an urban-rural gradient in 19 of the 24
mitigation banks they surveyed, averaging 2,419 more people/sq. mi. near impact sites. In
addition, Bendor et al. (2007) found that on average, 359 more people/sq. mi. lived near
impact sites than mitigation sites when assessing all off-site compensatory mitigation
programs. In a study of wetland banking in four Oregon counties, Brass (2011) found

83

population density to be, on average 1,060 people/sq. mile more near impact sites. See Table
11 for a comparative summary of key findings including the results from this analysis.
Table 11. Comparative summary statistics of differences in population densities.

Study

Year

Ruhl &
Salzman

Bendor et
al.

Brass
BenDor &
Stewart

McKellips

Study
Area

2006 Florida
Four
counties
in
northwest
2007 IL
Four
counties
near
Eugene,
2011 OR
North
2011 Carolina
Three
counties
in western
2017 WA

Precision

Mitigation
Type

Zip codes

wetland
banking

Census
tracts

off-site
compensatory
mitigation

1 mile
buffer using
Census
Block data

wetland
banking

wetland
census tracts mitigation
.75 mile
buffer using off-site
2010
compensatory
Census data mitigation

Population
difference Ppeople/mi² value

2,419

N/A

355

<.01

1,060

0.0038

1,082

<.01

926

<.0001

When assessing previous research, mitigation approaches with “many-to-one”
mitigation (e.g. wetland banking and ILF) tend to increase the difference in population
densities. BenDor and Stewart (2011), for example, found ILF programs to exhibit the
greatest difference, followed by mitigation banking. PRM mitigation, with one-to-one
mitigation showed the lowest differentials in population densities. While sample sizes in this
study for ILF were low (n=20), ILF sites showed lower differences in population densities,

84

averaging 164 people/sq. mi. at each site. Mitigation banking exhibited the greatest
difference in population densities, ranging from 328-720 people/sq. mi.
The area of developed land serves as another indicator to understand mitigation site
location. Overall, mitigation sites were 8.5% less developed than impact sites. Some sites
may have low population densities, but still be located in highly developed industrial
landscapes. Summary statistics for two mitigation banks, Columbia River and Springbrook,
exhibited higher percentages of development but lower population densities at impact sites.
For the Columbia River mitigation bank, this its proximity to the Port of Vancouver may
explain the high-development percentage, but low population findings.
Socioeconomic Equity
Results from this study indicate that on average, wetland mitigation relocates
wetlands away from populations with lower home values and lower median household
incomes. On average, home values were $43,250 greater where wetland mitigation (i.e.
wetland gains) were taking place. These findings however varied considerably among the
different mitigation approaches and between mitigation banks. For example, the King County
ILF program, Columbia River, East Fort Lewis, and Skykomish mitigation banks, which
comprise nearly half of the 139 projects, had higher home values near impact sites (n=68).
The differential in home values was disproportionately influenced by the Snohomish
mitigation bank, where the median home value was $541,667. Values at this site had a large
influence in the overall mitigation data, as replicate data from the mitigation bank were used
to compare with each of the 52 impact sites.
Median household income displayed similar variance. While on average, incomes
were $9,032 lower near impact sites, the King County ILF program, Columbia River and

85

East Fort Lewis mitigation banks had higher incomes at the impact sites (n=43). The
Snohomish mitigation bank also significantly influenced this variable, with an average of
over $28,000 greater income at mitigation sites. The degree of variability between the
mitigation banks points to a more nuanced conclusion of socioeconomic equity than the
summary data indicate.
With the exception of Skykomish mitigation bank, the total number of households in
poverty are greater near impact sites than mitigation. Given there are, on average, 926 more
people per square mile near impact sites than mitigation sites, one expects a corresponding
increase in households in poverty. Across the 139 sites, the average difference of households
in poverty is 26. These households would have the least resources to soften the impacts of a
proposed development project requiring mitigation, whether through loss of wetland
ecosystem services or other consequences of a project, such as increased housing prices.
While this study quantitatively identifies the number of households in poverty near impact
and mitigation sites, determining the aggregate effect of households in poverty in the study
area is beyond the scope of this study. Further research could examine at a finer scale how
proposed mitigation projects affect households in poverty.
Racial Equity
This research supports the claim that a greater percentage of minority populations are
living closer to mitigation sites than impact sites, indicating that wetland mitigation equitably
distributes wetlands and their ecosystem services to minority populations. These findings
were consistent across all minority racial groups, with the exception of the Asian population.
However, the Asian population was also the only racial population to not show a significant
difference between impact and mitigation sites (p=0.2838). In general, the large majority of

86

White populations posed a challenge when looking at the minority racial populations, since
minority population numbers and percentages are low to begin with. While instructive to
examine individual racial groups, combining these groups under minority population—
particularly with the small sample size in this study—created a clearer picture of racial
equity.
With minority populations having higher populations near mitigation sites, it follows
that more White populations live near impact sites. Thus, mitigation inequitably distributes
wetland ecosystem service benefits to White populations. This finding supports research by
BenDor and Stewart (2011) that found higher percentages of White populations near impacts
sites but counters findings in northeast Illinois (BenDor et al., 2007) that observed higher
White populations near more rural mitigation sites. While off-site compensatory wetland
mitigation causes wetland relocation from one population to another, results from across
these studies do not indicate that wetland site selection strongly favors one racial population
over another.
It is worth noting that these findings appear to stand in contrast to the racial income
gap in the United States. This gap shows that among working families, minorities are 24%
more likely to be low-income or poor than non-Hispanic Whites (Povich, Roberts, & Mather,
2015). However, this dataset does not link household income to specific racial groups. Thus,
while minority populations are observed to have higher percentages near mitigation sites with
higher incomes, this fact does not per se indicate that minority populations are the
populations with these higher incomes.

87

Limitations
This research examined populations living near impact and mitigation sites through
CWA wetland mitigation to address equity in the relocation of wetland resources. Using a ¾mile buffer enabled a snapshot of populations living nearby these wetland losses and wetland
gains. However, the fact that wetland ecosystem services benefit human populations at larger
spatial scales than ¾-mile limits this research’s ability to fully capture populations affected
by wetland relocation. For example, the ecosystem service benefits of improved water quality
and aquifer recharge occur at a watershed or regional level. In addition, increased wetlands
and floodplains along rivers provide the ecosystem service of reduced flood risk many miles
from its location. Furthermore, many of the adverse effects of increased urbanization and
degraded wetland resources are addressed at the watershed and regional level. Past hedonic
studies point to populations valuing cultural ecosystem services such as aesthetics and
recreation within a ¾-mile buffer, but wetlands’ variability makes measuring the extent of
their multiple benefits decidedly demanding.
One assumption embedded within this research presumes that human populations
benefit from proximity to wetlands. The same logic assumes populations incur adverse
effects as wetland resources diminish. Undoubtedly, this logic simplifies this relationship.
For example, a proposed development that damages wetlands may represent an economic
investment in a community. To use a clear example, a proposed health clinic may represent a
public good whose benefits to the surrounding population outweighs the impacts to wetland
resources. The wetland mitigation sequence permits these developments to occur while not
losing net wetland acreage or ecosystem services.

88

The anthropogenic lens of this research should also be acknowledged. This study
scarcely acknowledges wildlife and land conservation goals. While human populations
benefit from wetlands, urban wetlands present more potential dangers (e.g. car traffic,
eutrophication) to wildlife. When done in tandem with broader goals, wetland mitigation
could increase habitat connectivity and wildlife corridors that benefit non-human
populations. Of course, human populations value wildlife and its corresponding habitat as
well. Many existing federal policies such as the Endangered Species Act require prioritizing
existing habitat and habitat connectivity. Thus, rural wetland site selection may have benefits
outside the parameters discussed in this research.
Future Research
The results from this study analyzed aggregate wetland mitigation projects in a spatial
context. While individual mitigation projects follow site selection criteria, regulatory
agencies rarely spatially analyze aggregate wetland mitigation in a region. Rather, regulatory
agencies provide total acreage lost and gained through mitigation to track no-net-loss
objectives but cannot fully evaluate these projects in a spatial context (Boyd & Wainger,
2002). Over time, annual wetland relocation could affect aquatic resources at the landscape
level. In turn, this relocation may affect populations losing and gaining aquatic resources and
their ecosystem service benefits. Maintaining accurate data on how wetlands are moving
across the landscape can help land use planners understand changes in wetland resources
over time. Accurate, up-to-date data also enables analyses of environmental equity among
different populations. The current permitting structure does not emphasize linking impact and
mitigation sites or geospatially maintaining data. Integrating these two initiatives would
greatly increase the facility to conduct future research.

89

Future research should seek to further target spatial dynamics and how it relates to
wetland mitigation. There are two principle areas where research can be refined. Integrating
quantitative and qualitative wetland data into GIS will support spatial analyses. While the
ACOE data set did not contain linked impact and mitigation sites, the ORM2 database from
which the data set came contained organized site characteristics. For example, each site
indicated if the mitigation was in-kind (same type of wetlands) or out-of-kind (different type
of wetlands) and wetland class, which classifies a wetland’s functionality. These
characteristics should be examined spatially. Acreage should be included as part of the
analysis. Linking wetland acreage with mitigation would enable future research to assess the
relative influence of each project. Linking sites could solve some of the challenges dealing
with impacts that mitigates impacts at multiple sites.
In addition, socioeconomic characteristics surrounding wetland mitigation projects
should increase. Due to the small impacts of individual projects, wetland mitigation projects
do not often initiate strong opposition, even though they are valued by local residents
(Manuel, 2003). Precisely because these impacts are small and often under the radar of local
citizens, regulatory agencies have even more responsibility to ensure the aggregate impacts
are not adversely affecting local populations.
Conclusion
The results from this study contribute to the growing body of research examining the
effects that wetland mitigation has on wetland relocation along an urban-rural gradient and
on local populations. Determining the full effects on local populations remains elusive, given
that wetlands provide multiple ecosystem services, these ecosystem service benefits accrue to
the public at large and wetland area is not always an accurate indication of wetland value and

90

wetland functions are variable, which often times are influenced by human perturbations
(Mitsch & Gooselink, 2000).
Despite the difficulty—if not the impossibility—of pinning down specific wetland
values, the cumulative importance of wetlands in this country has been recognized under the
no-net-loss policy of the first Bush Presidency and supported by every proceeding
administration. While slow to coalesce, government agencies such as the EPA, ACOE and
DOA have united to administer and promote wetland protection, enhancement, and
restoration. To maintain healthy wetland resources, wetland mitigation under §404(1)(b) of
the CWA represents the most important piece in the puzzle. This law sets up the framework
to avoid, minimize, and compensate for wetland impacts. Through compensatory wetland
mitigation, hundreds of thousands of wetland acres have been restored, enhanced, created, or
preserved. Wetland mitigation also represents the largest ecosystem service market in the
country (Salzman & Ruhl, 2001).
Through the decades, regulators and policy makers have sought to amend mitigation
practices to improve upon the many challenges of ecosystem service trading. Salzman and
Ruhl (2001) recognized that in order to trade an ecosystem service, a common currency must
be traded equally over space, time, and type. While scales of type and time have substantial
challenges, this research addressed the challenge of trading wetlands over space. To address
the challenges of spatial relocation of wetlands, EPA directives have moved from preferences
of on-site mitigation and little relocation distances to favoring greater relocation distances
through the development of “many-to-one” mitigation projects such as wetland banks and
ILF mitigation. In addition to increasing the likelihood of site success (i.e., replacing wetland

91

functions), this system also improves oversight from the ACOE and DOE charged with
ensuring all off-site compensatory mitigation meet their performance standards (NRC, 2001).
While increasing efficacy of wetland management, these directives may also
exacerbate the migration of wetlands from densely populated environments. Urban
environments that lack wetlands are absent the many ecosystem services wetlands provide. In
the Puget Sound, for example, die-offs of spawning Coho salmon (Oncorhynchus kisutch)
increasingly point to stormwater pollution carrying toxic pollutants as the driver. Scholz et al.
(2011) conducted stream surveys and found Coho die-offs linked to rainfall events. These
die-offs, however, only occurred in urban areas, with O. kisutch in non-urban creeks being
unaffected. Roads—a common source of wetland mitigation projects—in particular have
shown to be a source of contaminants that disrupts aquatic biota (Trombulak & Frissell,
2000). Thus, wetland mitigation may impair natural resiliency in urban areas two ways, by
relocating ecosystem services to less-populated areas and worsening non-point source
pollutants. In order to improve upon the overarching goal of the CWA— to “restore and
maintain the biological, chemical, and physical integrity” of U.S. waters,” regulatory
agencies should recognize the potential impacts to urban and urbanizing environments. In
order to minimize these impacts, this research offers the following recommendations.
Recommendations
This research proposes three recommendations to increase understanding of spatial
dynamics and improve environmental equity in wetland mitigation. First, where feasible,
regulatory agencies should promote wetland mitigation banks in urban areas. Second,
regulatory agencies should improve access to mitigation data to better understand aggregate

92

effects of wetland relocation. Lastly, incentives to increase mitigation site selection in urban
areas should be considered.
First, wetland mitigation should prioritize wetland bank site selection to benefit urban
areas. To reduce the amount of urban-rural wetland relocation, wetland mitigation bank site
selection should be evaluated to sustain wetland ecosystem services in urban areas. Wetland
mitigation banks represent a system where one mitigation site accounts for multiple impacts
across the landscape. With only one mitigation site, the site selection of mitigation banks
provides a critical opportunity to sustain wetland ecosystem services within urban and
urbanizing locations. On the contrary, if wetland bank site selection favors rural locations
displaced from urban areas, the effect will exacerbate the urban-rural migration of wetlands.
Second, improving data management would improve overall understanding of
landscape scale impacts of wetland relocation. Over the past few decades, hundreds of
thousands of wetland acres have been relocated through compensatory wetland mitigation.
While the ACOE tracks the total acres of permitted impacts and mitigation at national scales,
narrowing wetland mitigation impacts to smaller scales proves difficult given the dearth of
publicly available information. Requiring all wetland mitigation project to clearly link
wetland impacts and wetland mitigation sites could vastly improve geospatial analysis. Given
the growth of publicly available geospatial data from government and non-government
agencies alike, coupling wetland mitigation data could increase collective knowledge of
wetland mitigation patterns and trends. In particular, understanding population characteristics
near impact and mitigation sites remain poorly understood. In one of the first analyses of its
kind to document wetland relocation, Ruhl and Salzman (2006) lamented the significant data
vacuum that exists within mitigation banks. These researchers identified include land values

93

of sites and demographic data associated with banks as primary data gaps. Addressing the
difficulty if wetland relocation along an urban-rural gradient has any significance, the authors
wrote that “it is difficult to approach this question intelligently, since no actor in the banking
process takes steps that would allow us to test the policy implications of the phenomenon—
i.e., tracks the redistribution of wetlands, estimates the effects thereof on ecosystem service
values, notifies the affected public, and provides opportunity for public input,” (p. 10). This
vacuum still exists today. Given the large amount of documentation required in wetland
mitigation projects, changing guidelines to ensure regulatory agencies link wetland impacts
and mitigation would facilitate increased understanding by regulatory agencies and the public
at large of aquatic resources through §404(1)(b) permits.
Third, wetland mitigation guideline could change incentive structures to level playing
field for urban-rural site selection. The current market-based structure does not take into
account human populations. As over 2/3 of the United States’ population now live in urban
areas, urban areas and their growing populations could benefit greatly from functioning
wetland ecosystem services. Since these considerations are not included in mitigation site
selection criteria, any arguments that “efficient allocation” of resources under principles of
market-based economies are rendered null, (Ruhl & Salzman, 2006). Rather, land prices have
been identified as the prime criteria under which mitigation sites are selected (Kaplowitz &
Bailey, 2008).
These recommendations maintain the core essence of the CWA. During the 40-year
history of wetland mitigation, guidelines have changed to improve upon mitigation practices
that prevents further degradation of our country’s wetland resources. Taking into account
surrounding human populations is an extension of this iterative process of improving wetland

94

mitigation practices. As nationwide trends indicate increasing urban populations in the next
century, I recommend that wetland mitigation address this new literal and metaphorical
landscape to improve the equitable distribution of wetland ecosystem services to urban and
urbanizing populations.

95

Works Cited
ACOE & EPA (2008). Compensatory Mitigation for Losses of Aquatic Resources; Final
Rule. 33 CFR Parts 325 and 332 and 40 CFR Part 230. Published in Federal Register
April 10, 2008.
ACOE (1984). EPA’s role in determining compliance with sec 404(b)(1) Guidelines.
Regulatory Guidance Letter 84-06, issued 26 March, 1984, Washington DC.
Alberti, M., Booth, D., Hill, K., Coburn, B., Avolio, C., Coe, S., & Spirandelli, D. (2007).
The impact of urban patterns on aquatic ecosystems: An empirical analysis in Puget
lowland sub-basins. Landscape and Urban Planning, 80, 345-361.
Azous, A., & Horner, R. R. (Eds.). (2000). Wetlands and urbanization: Implications for the
future. New York, NY: Lewis Publishing.
BenDor, T.K., & Bruzovic, N. (2007). Determinants of spatial and temporal patterns in
compensatory wetland mitigation. Environmental Management, 40, 349-364.
BenDor, T.K., Bruzovic, N., & Pallathucheril, V.G. (2007). Assessing the socioeconomic
impacts of wetland mitigation in the Chicago Region. Journal of the American
Planning Association, 73(4), 263-282.
BenDor, T.K., Brozovic, N., & Pallathucheril, V.G. (2008). The social impacts of wetland
mitigation policies in the United States. Journal of Planning Literature, 22(4), 341357.
BenDor, T.K., & Stewart, A. (2011). Land use planning and social equity in North Carolina’s
compensatory wetland and stream mitigation programs. Environmental Management,
(47)2, 239-253.
Bockstael, N. E., Freeman, A. M., Kopp, R. J., Portney, P. R., & Smith, V. K. (2000). On
96

measuring economic values for nature. Environmental Science & Technology, 34(8),
1384-1389.
Boyd & Wainger (2002). Landscape indicators of ecosystem service benefits. American
Journal of Agricultural Economics, 84(5), 1371-1378.
Boyer, T., & Polasky, S. (2004). Valuing urban wetlands: A review of non-market valuation
studies. Wetlands, 24(4), 744-755.
Brander, L. M., Florax, R. J., & Vermaat, J. E. (2006). The empirics of wetland valuation: A
comprehensive summary and a meta-analysis of the literature. Environmental and
Resource Economics, 33(2), 223-250.
Brass, T.W. (2009). Who is affected by wetland mitigation banking? A social and geographic
evaluation of wetland mitigation banking in Benton, Lane, Linn and Polk Counties,
Oregon (Unpublished Masters Thesis). University of Oregon, Eugene, OR.
Bush, G.H.W., (1989, June 8). Remarks to Members of Ducks Unlimited. Retrieved from The
American Presidency Project, http://www.presidency.ucsb.edu/ws/?pid=17125
Chichilnisky, G., & Heal, G. (1998). Economic returns from the biosphere. Nature, 391, 629630.
Clare, S., Krogman, N., Foote, L., & Lemphers, N. (2011). Where is the avoidance in the
implementation of wetland law and policy? Wetlands Ecological Management, 19,
165-182.
Costanza, R., d'Arge, R., De Groot, R., Faber, S., Grasso, M., Hannon, B., ... & Raskin, R. G.
(1997). The value of the world's ecosystem services and natural capital. Nature, 387,
253-260.
Costanza, R., Farber, S. C., & Maxwell, J. (1989). Valuation and management of wetland

97

ecosystems. Ecological economics, 1(4), 335-361.
Costanza, R., Stern, D., Fisher, B., He, L., & Ma, C. (2004). Influential publications in
ecological economics: A citation analysis. Ecological Economics, 50(3), 261-292.
Cowardin, L. M., Carter, V., Golet, F. C. & LaRoe, E.T. (1979). Classification of wetlands
and deepwater habitats of the United States. US Fish & Wildlife Service, FWS/OBS79/31.
Craig, R. K. (2008). Justice Kennedy and ecosystem services: A functional approach to
Clean Water Act jurisdiction after Rapanos. Environmental Law, 38, 635-666.
Dahl, T.E. (1990). Wetlands: Losses in the United States 1780’s to 1980’s. U.S. Department
of the Interior, Washington, D.C. 13 p.
Daily, G. C., Söderqvist, T., Aniyar, S., Arrow, K., Dasgupta, P., Ehrlich, P. R., ... & Levin,
S. (2000). The value of nature and the nature of value. Science, 289(5478), 395-396.
Department of Ecology (2013). Non-Discrimination, Executive Policy 1-4. Retrieved from
http://www.ecy.wa.gov/jobs/DiversityPolicy.pdf
Dorman, P. (2004). Ecological versus environmental economics: What’s the difference?”
(Unpublished article). Olympia, WA.
Downing, D. M., Winer, C., & Wood, L. D. (2003). Navigating through Clean Water Act
jurisdiction: A legal review. Wetlands, 23(3), 475-493.
Dunn, A.D., (2010). Siting green infrastructure: Legal and policy solutions to alleviate urban
poverty and promote healthy communities. Environmental Affairs, 37(41), 41-66.
Environmental Law Institute [ELI] (2006). The status and character of in-lieu fee mitigation
in the United States. Washington, D.C.
EPA (2017a). Memorandum of agreement between the Department of the Army and the

98

Environmental Protection Agency. Retrieved from https://www.epa.gov/cwa404/memorandum-agreement
EPA (2017b). Urbanization and population change. Report on the Environment. Retrieved
from https://cfpub.epa.gov/roe/indicator.cfm?i=52
Exec. Order No. 12898, 3 C.F.R. 5 (1994).
Frayer, W.E., Monahan, T.J., Bowden, D.C., & Graybill, F.A. (1983). Status and trends of
wetlands and deepwater habitats in the conterminous United States, 1950’s to 1970’s.
US Fish and Wildlife Service: Washington, D.C.
FWPCA [Federal Water Pollution Control Act Amendments of 1972], 33 U.S.C. §§
1251-1387 (1972).
Gardner, R.C. (2005). Mitigation. In Connolly, K.D., Johnson, S.M., & Williams, D.R.
(Eds.), Wetlands Law and Policy. Chicago, IL: American Bar Association.
Georeference. (n.d.). In ESRI Online GIS Dictionary. Retrieved from
http://support.esri.com/other-resources/gis-dictionary/term/georeferencing.
GAO [US General Accounting Office] (1988) Wetlands: the Corps of Engineers’
administration of the Section 404 program. Report GAO/RCED-88-110. GAO,
Washington DC.
Godschalk, D. R. (2004). Land use planning challenges. Journal of the American Planning
Association, 70(1), 5–13.
Hansen, L., (2006). Wetland status and trends. Wlebe, K., & Gollehon, N., eds. Agricultural
Resources and Environmental Indicators, 2006 Edition. USDA.
Heal, G. (2000). Nature and the marketplace: Capturing the value of ecosystem services.
Washington, DC: Island Press.
Hough, P., & Robertson, M. (2009). Mitigation under Section 404 of the Clean Water Act:
99

where it comes from, what it means. Wetlands Ecology and Management, 17(1), 1533.
Hruby, T., Harper, K., & Stanley, S. (2009). Selecting wetland mitigation sites using a
watershed approach. Washington State Department of Ecology Publication #09-06032.
Jennings, V., Gaither, C.J., & Gragg, R. S. (2012). Promoting environmental justice through
urban green space access: A synopsis. Environmental Justice, 5(1), 1-7.
Kaplowitz, M.D., & Bailey, D. (2008). Wetland mitigation banking: The banker’s
perspective. Journal of Soil and Water Conservation, 63(3), 162-172.
Kaza, N., & BenDor, T. K. (2013). The land value impacts of wetland restoration. Journal of
Environmental Management, 127, 289-299.
Kentula, M. E., Gwin, S. E., & Pierson, S. M. (2004). Tracking changes in wetlands with
urbanization: Sixteen years of experience in Portland, Oregon, USA. Wetlands, 24(4),
734-743.
King, D. M., & Herbert, L. W. (1997). The fungibility of wetlands. National Wetlands
Newsletter, 19(5), 10-13.
King, D. (1998). The dollar value of wetlands: Trap set, bait taken, don’t swallow. National
Wetlands Newsletter, 20, 7-11.
Lele, S., Springate-Baginski, O., Lakerveld, R., Deb, D., & Dash, P. (2013). Ecosystem
services: Origins, contributions, pitfalls, and alternatives. Conservation and
Society, 11(4), 343.
Manuel, P.M. (2003). Cultural perceptions of small urban wetlands: Cases from the Halifax
regional municipality, Nova Scotia, Canada. Wetlands, 23(4) 921-940.

100

Mahan, B.L., Polasky, S., & Adams, R. (2000). Valuing urban wetlands. A property price
approach. Land Economics, 76, 100-113.
Martin, S., & Brumbaugh, R. (2011). Entering a new era: What will RIBITS tell us about
mitigation banking. National Wetlands Newsletter, 33(3), 16-18.
Millennium Ecosystem Assessment (2005a). Ecosystems and human well-being: Health
synthesis. Washington, DC: Island Press.
Millennium Ecosystem Assessment (2005b). Ecosystems and human well-being:
synthesis. Washington, DC: Island Press.
Millennium Ecosystem Assessment (2005c). Ecosystems and human well-being: water and
wetlands synthesis. World Resources Institute, Washington, DC: Island Press.
Merino-Saum, A., & Roman, P. (2012). What can we learn from ecological
economics? Economics, 1, 1-7.
Mitsch, W. J., & Gosselink, J. G. (2000). The value of wetlands: Importance of scale and
landscape setting. Ecological economics, 35, 25-33.
Mitsch, W.J., & Gosselink, J.G (2015). Wetlands (5th ed.). Hoboken, New Jersey: Wiley.
National Research Council (1995). Wetlands: Characteristics and Boundaries. Washington,
DC: National Academy Press.
National Research Council (2001). Compensating for wetland losses under the Clean Water
Act. Washington, DC: National Academy Press.
National Wildlife Federation v. John O. Marsh Jr. 665 F2d 390. US Court of Appeals for the
District of Columbia (1981).
Perrings, C., & Hannon, B. (2001). An introduction to spatial discounting. Journal of
Regional Science, 41(1), 23-38.

101

Povich, D., Roberts, B., & Mather, M., (2015). Low-income working families: The
racial/ethnic divide. the working poor families project. Retrieved from
http://www.workingpoorfamilies.org/wp-content/uploads/2015/03/WPFP-2015Report_Racial-Ethnic-Divide.pdf
Puget Sound Regional Council (2016). Growth in the Puget Sound Region. Retrieved from
http://kingcounty.gov/~/media/depts/executive/performance-strategy-budget/regionalplanning/GrowthManagement/GMPCMeeting033016/Growth_presentation_KC_GM
PC_3-30-2016.ashx?la=en
Rapanos v. United States. 547 US 715. Supreme Court of the United States (2006).
Reynolds, J.E. & Regalado, A., (2002). The effects of wetlands and other factors on rural
land values. Appraisal Journal, 72, 182-190.
Robertson, M.M. (2004). The neoliberalization of ecosystem services: Wetland mitigation
banking and problems in environmental governance. Geoforum, 35, 361-373.
Robertson, M., & Hayden, N. (2008). Evaluation of a market in wetland credits:
Entrepreneurial wetland banking in Chicago. Conservation Biology, 22(3), 636-646.
RHA [River and Harbors Appropriation Act of 1899], 33 USC § 401 et. seq.
Ruhl, J.B. & Salzman, J. (2006). The effects of wetland mitigation banking on people.
National Wetlands Newsletter, 28(2), 7 p.
Salzman, J., & Ruhl, J.B. (2001). Currencies and the commodification of environmental law.
Stanford Law Review, 53, 607-694.
Scholz, N. L., Myers, M. S., McCarthy, S. G., Labenia, J. S., McIntyre, J. K., Ylitalo, G. M.,
... & McMillan, B. (2011). Recurrent die-offs of adult coho salmon returning to
spawn in Puget Sound lowland urban streams. PloS one, 6(12), e28013.

102

Seattle Public Utilities (2013). Water system plan: plan summary. Seattle, WA. Retrieved
from www.seattle.gov/util/WaterSystemPlan
Solid Waste Agency of Northern Cook County v. Army Corps of Engineers. 531 US 159.
Supreme Court of the United States (2001).
Tacoma Public Utilities (2008). Green River water management plan, Vol. II. Tacoma, WA.
Tiner Jr., R.W. (1984). Wetlands of the United States: Current status and recent trends.
USFWS, Department of the Interior. Newton Corner, MA, 71 p.
Trombulak, S.C., & Frissell, C.A. (2000). Review of ecological effects of roads on terrestrial
and aquatic communities. Conservation Biology, 14(1), 18-30.
Turner, R.E., Redmond, A.M., & Zedler, J.B. (2001). Count it by acre or function—
Mitigation adds up to net loss of wetlands. National Wetlands Newsletter, 23(6), 5 p.
United States Census Bureau (2017). Population and housing unit estimates. Retrieved form
https://www.census.gov/programs-surveys/popest.html
United States Geological Survey [USGS] (2017). 2011 National Land Cover Database
product legend. Retrieved from https://www.mrlc.gov/nlcd11_leg.php
Vileisis, A. (1997). Discovering the Unknown Landscape. Washington, DC: Island Press.
Walker, S., Brower, A. L., Stephens, R. T., & Lee, W. G. (2009). Why bartering biodiversity
fails. Conservation Letters, 2(4), 149-157.
Woodward, R. T., & Wui, Y. S. (2001). The economic value of wetland services: A metaanalysis. Ecological economics, 37(2), 257-270.
Zedler, J.B. (1996). Ecological issues in wetland mitigation: An introduction to the forum.
Ecological Applications, 6(1), 33-37.
Zedler, J. B., & Callaway, J. C. (1999). Tracking wetland restoration: Do mitigation sites

103

follow desired trajectories? Restoration Ecology, 7(1), 69-73.

104

Appendices
APPENDIX A: Ecosystem Services by category, adapted from Millennium Ecosystem
Assessment (2005c).
Ecosystem services
Provisioning
Food
Fresh water
Fiber and fuel
Biochemical
Regulating
Climate regulation
Water regulation (hydrological flows)
Water purification and waste
treatment
Erosion regulation
Natural hazard regulation
Pollination
Cultural
Spiritual and inspirational
Recreational
Aesthetic
Educational
Supporting
Soil formation
Nutrient cycling

Examples
Production of fish, wild game, fruits, grains
Storage and retention of water for domestic,
industrial, and agricultural use
Production of logs, fuelwood, peat, fodder
Extraction of medicines and other materials from
biota
Source and sink for greenhouse gases, influences
local and regional temperatures, precipitation
Groundwater recharge/discharge
Retention, recovery, and removal of excess
nutrients and other pollutants
Retention of soils and sediments
Flood control, storm protection
Habitat for pollinators
Source of inspiration, religious and spiritual
values
Opportunities for recreation
Beauty and aesthetic values
Formal and informal education and training
Sediment retention and accumulation of organic
matter
Storage, recycling and processing of nutrients

105

APPENDIX B: Economic valuation methods used to estimate wetland values, verbatim from
Brander, Florax, and Vermaat (2006).

106

APPENDIX C. Definitions of Compensatory Mitigation Methods, verbatim from IWR
(2015).
1. RESTORATION
The manipulation of the physical, chemical, or biological characteristics of a site with the
goal of returning natural/historic functions to a former or degraded aquatic resource.
For the purpose of tracking net gains in aquatic resource area, restoration is divided into
two categories: reestablishment and rehabilitation.
1.1.RE-ESTABLISHMENT
The manipulation of the physical, chemical, or biological characteristics of a site
with the goal of returning natural/historic functions to a former aquatic resource.
Re-establishment results in rebuilding a former aquatic resource and results in a
gain in aquatic resource area and functions.
1.2 REHABILITATION
The manipulation of the physical, chemical, or biological characteristics of a site
with the goal of repairing natural/historic functions to a degraded aquatic
resource. Rehabilitation results in a gain in aquatic resource function, but does
not result in a gain in aquatic resource area.
2. ENHANCEMENT
The manipulation of the physical, chemical, or biological characteristics of an aquatic
resource to heighten, intensify, or improve a specific aquatic resource function(s).
Enhancement results in the gain of selected aquatic resource function(s), but may also
lead to a decline in other aquatic resource function(s). Enhancement does not result in
a gain in aquatic resource area.
3. ESTABLISHMENT (CREATION)
The manipulation of the physical, chemical, or biological characteristics present to
develop an aquatic resource that did not previously exist at an upland site.
Establishment results in a gain in aquatic resource area and functions.
4. PRESERVATION
The removal of a threat to, or preventing the decline of, aquatic resources by an action
in or near those aquatic resources. This term includes activities commonly associated
with the protection and maintenance of aquatic resources through the implementation of
appropriate legal and physical mechanisms. Preservation does not result in a gain of
aquatic resource area or functions.

107

Item sets

The Evergreen State College Masters Theses: Environmental Studies

Media

McKellips_TMMESthesis2017.pdf