-
Title
-
Eng
A Multifunctional Landscape Approach to Reconciling Renewable Energy and Crucial Habitat Needs in Washington State
-
Date
-
2014
-
Creator
-
Eng
Keese, Krystal
-
Subject
-
Eng
Environmental Studies
-
extracted text
-
A MULTIFUNCTIONAL LANDSCAPE APPROACH TO
RECONCILING RENEWABLE ENERGY AND
CRUCIAL HABITAT NEEDS IN WASHINGTON STATE
by
Krystle Keese
A Thesis
Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
June 2014
© 2014 by Krystle Keese. All rights reserved.
This Thesis for the Master of Environmental Studies Degree
by
Krystle Keese
has been approved for
The Evergreen State College
by
________________________
Edward A. Whitesell, Ph.D.
Member of the Faculty
________________________
Date
ABSTRACT
A Multifunctional Landscape Approach to
Reconciling Renewable Energy and
Crucial Habitat Needs in Washington State
Krystle Keese
Habitat conservation and renewable energy development are both
environmentally beneficial initiatives. Habitat conservation aims to protect and restore
biodiversity and important habitats. Renewable energy development is an important
climate mitigation strategy. Although both land uses are important to address the
environmental challenges of today, management of these environmental initiatives has
stayed segregated and sometimes works at cross purposes. One approach to reducing this
conflict is to design multifunctional landscapes, where ecological, cultural, economic,
and energy resource values of the land are optimized. For this to happen, management
and analysis must be approached from a landscape perspective. Within a landscape-level
perspective, this study aims to understand “how do wind and solar energy development
and habitat conservation priorities conflict with one another in Washington State?” This
research question is analyzed using GIS basic spatial analysis and spatial autocorrelation
(Moran’s Local I) within three spatial contexts: existing wind farms, suitable wind and
solar development lands, and Washington habitats. Results show that there is a moderate
to low conflict between habitat conservation priorities and both existing wind farms and
suitable wind or solar energy development lands. Wind energy development could be
restricted to less crucial habitat lands 3-6 and still grow by an estimated 440% of current
wind energy production. Solar energy development could be restricted to the least crucial
habitat levels and still increase existing total state energy production by 50%. Regarding
Washington habitats, there are significant wind and solar resources in grasslands and
shrublands, but also a high risk of conflict with most crucial habitats. However, the
agriculture, pasture, and mixed-environments habitats present the greatest opportunities
to explore multifunctional landscape designs. With this type of assessment, landscape
planners can begin exploring how to approach landscape management from a
multifunctional landscape design, balancing the value of renewable energy potential and
habitat conservation priorities.
Table of Contents
Chapter 1: Introduction ................................................................................1
Chapter 2: Literature Review ......................................................................8
2.1 Introduction ....................................................................................................................... 8
2.2 Conservation of Habitats ................................................................................................ 8
Population Dynamics .................................................................................................. 9
Island Biogeography ................................................................................................. 12
Landscape Ecology ................................................................................................... 15
2.3 Renewable Energy Development ................................................................................ 17
2.4 Environmental Impacts of Renewable Energy Development .............................. 19
Infrastructure: Electrical Transmission Lines and Roads ......................................... 20
Wind Energy Development....................................................................................... 22
Solar Energy Development ....................................................................................... 25
Environmental Impact Assessments ......................................................................... 27
2.5 Multifunctional Landscapes ........................................................................................ 29
Chapter 3: Methods and Analysis ..............................................................37
3.1 Methods ............................................................................................................................ 37
Study Area ................................................................................................................ 37
Research Questions and Hypotheses ........................................................................ 39
Methods..................................................................................................................... 42
iv
Data Sources ............................................................................................................. 42
Wind and Solar Data Resources ........................................................................... 42
Existing Wind Turbines ........................................................................................ 45
Crucial Habitat Assessment .................................................................................. 45
Washington Wildlife Habitats............................................................................... 48
Protected Areas of the United States .................................................................... 49
National Wetland Inventory.................................................................................. 49
National Historic Places ........................................................................................ 50
Washington Cities and Urban Growth Areas........................................................ 51
Washington Agricultural Land Use ...................................................................... 52
Data Limitations........................................................................................................ 52
Data Preparation........................................................................................................ 55
3.2 Data Analysis ................................................................................................................... 58
Existing Washington Wind Farms ............................................................................ 58
Wind and Solar Energy Development Landscapes ................................................... 59
Washington Habitats ................................................................................................. 63
Chapter 4: Results .......................................................................................67
4.1 Existing Washington Wind Farms ............................................................................. 67
4.2 Wind and Solar Energy Development Landscapes ................................................ 74
Wind Energy Assessment ......................................................................................... 74
Solar Energy Assessment .......................................................................................... 79
4.3 Washington Habitats ..................................................................................................... 85
v
Wind Energy Resource Analysis .............................................................................. 88
Solar Energy Resource Analysis ............................................................................... 96
Chapter 5: Discussion ................................................................................104
5.1 Existing Washington Wind Farms ........................................................................... 104
5.2 Wind and Solar Energy Development Landscapes .............................................. 108
Solar Energy Resource Analysis Limitations ......................................................... 110
Opportunity for Future Research ............................................................................ 112
5.3 Washington Habitats ................................................................................................... 113
Wind Energy Resource and Crucial Habitat Analysis ............................................ 114
Significant High-Wind Clustering ...................................................................... 115
Significant Most-Crucial Habitat Clustering ...................................................... 116
Solar Energy Resource and Crucial Habitat Analysis ............................................ 117
Agriculture, Pasture, and Mixed-Environments Habitat..................................... 117
Grassland and Shrubland Habitats ...................................................................... 120
5.4 Policy Implications ....................................................................................................... 122
Washington State Energy Independence Act .......................................................... 122
Washington State Environmental Protection Act (SEPA) ...................................... 124
5.5 Conclusions .................................................................................................................... 127
Appendix .....................................................................................................140
vi
List of Figures
Figure 1. Wild Horse Wind and Solar Facilities (Ellensburg, WA) .................................27
Figure 2. Crucial habitat on all existing and operating Washington wind farms..............68
Figure 3. Crucial habitat on existing Washington wind farms and statewide...................70
Figure 4. Crucial Habitat Assessments of Wind Farm Zones ...........................................72
Figure 5. Crucial habitat on suitable wind energy lands and statewide ............................74
Figure 6. Crucial habitat on suitable wind energy development landscapes ....................76
Figure 7. Estimated annual average energy production (GWh) of suitable wind
energy development landscapes by crucial habitat rank ........................................79
Figure 8. Crucial habitat on suitable solar energy development lands and
statewide ................................................................................................................80
Figure 9. Crucial habitat on suitable solar energy development landscapes ....................82
Figure 10. Estimated annual average energy generation (GWh) of suitable solar
energy development lands by crucial habitat ranking............................................85
Figure 11. General Habitats in Washington State .............................................................86
Figure 12. Habitat distribution in Washington..................................................................87
Figure 13. Spatial interaction of wind energy resources and crucial habitat
statewide ................................................................................................................89
Figure 14. Spatial interaction between wind energy resources and crucial habitat
in the Washington grassland and shrubland habitats .............................................93
Figure 15. Spatial interaction of wind energy resources and crucial habitat in
Washington forest and woodland habitats .............................................................95
vii
Figure 16. Spatial interaction of solar energy resources and crucial habitat
statewide ................................................................................................................97
Figure 17. Spatial interaction of solar energy resources and crucial habitat in
Washington agriculture, pasture, and mixed-environment habitats.....................100
Figure 18. Spatial interaction of solar energy resources and crucial habitat in
Washington grassland and shrubland habitats .....................................................102
Figure 19. Crucial habitat on each existing Washington wind farm ...............................140
Figure 20. Spatial interaction of wind energy resources and crucial habitat in
Washington agriculture, pasture, and mixed-environment habitats.....................142
Figure 21. Spatial interaction of wind energy resources and crucial habitat in
Washington urban and mixed-environment habitats ...........................................142
Figure 22. Spatial interaction of wind energy resources and crucial habitat in
Washington wetland, rivers, lakes, and reservoir habitats ...................................142
Figure 23. Spatial interaction of solar energy resources and crucial habitat in
Washington forest and woodland habitats ...........................................................142
Figure 24. Spatial interaction of solar energy resources and crucial habitat in
Washington urban and mixed-environment habitats ...........................................142
Figure 25. Spatial interaction of solar energy resources and crucial habitat in
Washington wetland, rivers, lakes, and reservoir habitats ...................................142
viii
List of Tables
Table 1. Data Categories and Sources for Assigning Crucial Habitat Values ..................47
Table 2. Consolidation of Individual Habitat Types into General Habitats and
Habitat Exclusions .................................................................................................57
Table 3. Wind energy potential by crucial habitat levels ..................................................79
Table 4. Solar energy potential by crucial habitat level ....................................................85
Table 5. Significant wind resource landscapes in each habitat as a percentage of
total significant landscape areas.............................................................................91
Table 6. Significant wind resource landscapes as a percentage of total habitat area ........91
Table 7. Significant solar resource landscapes in each habitat as a percentage of
total significant landscape areas.............................................................................99
Table 8. Significant solar resource landscapes as a percentage of total habitat area ........99
Table 9. Crucial Habitat Ranking Factors .......................................................................141
ix
Acknowledgements
The process of preparing this thesis was one of the greatest academic challenges I
have embarked upon up to this time in my life. It was a test of my intellect, dedication to
knowledge, as well as my willpower and motivation to persevere. I could not have
succeeded without the help and support of many individuals over the past two years.
First, I would like to convey my appreciation for my reader, Dr. Edward Whitesell, for
partnering with me on this exciting yet challenging project. Dr. Whitesell’s feedback and
collaboration have enabled me to get this far and helped me learn many things about my
capabilities and my current self. Additionally, I would like to thank the remaining MES
faculty for their ever-present support and encouragement throughout my MES career, as
well as their exemplary teachings that have transformed my perspectives and helped me
develop into the student I am today and the future leader I hope to be as time goes
forward. Second, I would like to thank my colleagues in the MES class of 2014 for
continually challenging my perceptions of reality in our coursework and always passing
along valuable information and networks to help me continue exploring my passions for
the environment. Of particular importance I would like to thank my dear friend and
colleague Kyli Rhodes for working closely with me throughout this process. Her support
and encouragement helped me move through the personal challenges of managing such a
large project and enabled me to make steady progress toward my thesis completion.
Third, I would like to thank Alan Hardcastle and Tony Usibelli for supporting my
enthusiasm for renewable energy technologies and always connecting me to the industry
experts in their fields, to help me enhance my understanding of the Washington energy
industry. In addition, Dr. Greg Stewart provided an exemplary education in how to use
GIS as his continued support and interest in this project has enabled me to be successful.
Finally, my family and friends have also been a tremendous support to my goals and
aspirations throughout my graduate career and into the future. Thank you all who have
helped me get to the MES graduating class of 2014.
x
Chapter 1: Introduction
As the human population continues to grow, more and more of the landscape will
be required to support societal needs and wants. However, as more of the landscape is
used to meet these demands, a conflict ensues regarding how to balance the needs of
society and maintain ecological health and integrity across the landscape. The history of
anthropogenic land use has resulted in serious, large, negative impacts to Earth’s
biodiversity and a general decline in ecological health across the globe (Lubowski,
Plantinga, and Stavins 2008; Hanski 2011). This reinforces the importance of finding a
balance between human societies and the rest of the natural world. Two environmentally
beneficial initiatives that work toward this objective are habitat conservation and
renewable energy development.
Conservation biology, a field of science dedicated to protecting and restoring
ecologically important habitats, has never been more important to mitigate the many
negative ecological changes from anthropogenic land use (Trombulak et al. 2004). One
way to accomplish this is to apply active habitat conservation and management practices
to the landscape that will preserve important habitats, ecological services, and maintain
local biodiversity. However, while conservation biology aims to restrict land use and
restore impacted landscapes back to healthy ecological systems, human population
growth and economic pressures encourage continued landscape development and change.
Energy is a critical resource for any society. The production of energy draws from
various natural resources and impacts the landscape in a number of ways through the
process of development, operation, and eventual deconstruction (Burger and Gochfeld
1
2012). Due to growing climate change concerns there has been an increase in demand for
the development of renewable energy resources to meet societal energy needs. Over the
past decade wind and solar energy resources have seen the most growth both globally and
nationally and are expected to exhibit similar levels of growth over the next several years,
making these technologies important topics of study (Demirbas 2009; US Department of
Energy 2011). Despite being a favorable alternative for energy production in many ways,
various scales of land use are required for both wind and solar renewable energies and
negative impacts to the environment are still incurred (Northrup and Wittemyer 2013).
As humans continue to grow and expand the consumption and use of land,
conflicts between habitat conservation initiatives and renewable energy production are
bound to occur. Historically, land use has been approached from a single-function
perspective giving land management priority to a single land use (Harden et al. 2013).
However, new, more holistic approaches to land management including multifunctional
landscapes and energyscapes have been explored.
Under a multifunctional landscape design, the priorities of both the ecological
systems and energy potential of the land are considered from a more expansive
landscape-level perspective with the aim of optimizing the land use according to the
needs of both functions (Reyers et al. 2012; Howard et al. 2013). This will enable an
understanding of the risks and opportunities of future renewable energy development and
provide the information needed to work toward optimizing the landscape interaction
between both initiatives. While a multifunctional landscape approach to reconciling the
conflict between habitat conservation initiatives and renewable energy development
seems promising, there has been no real application of multifunctional landscape designs
2
within this context (Reyers et al. 2012; Howard et al. 2013). This is mainly due to the
large requirements of capital, time, as well as the challenges associated with stakeholder
agreement needed to implement multifunctional landscapes (Waltner-Toews, Kay, and
Lister 2008; Harden et al. 2013). Despite this, the first steps in moving toward a
multifunctional landscape design, is to gain an understanding of the landscape-level
interaction and levels of conflict between renewable energy development and habitat
conservation initiatives.
One of the largest challenges to investigating land use from a multifunctional
landscape perspective within this context has been that existing conservation planning
and management focuses on individual conservation priorities and specific species within
local and regional contexts (Washington Department of Fish and Wildlife 2005). There
has been no standardized indicator available to assess all conservation priorities from a
landscape-level perspective in relation to land use, until quite recently. The recent
publication of the Crucial Habitat Assessment Tool by the Western Governors’
Association now enables an understanding of the landscape according to the priorities of
multiple conservation initiatives (Western Governors’ Wildlife Council 2013a).
Crucial Habitat is a landscape-level environmental indicator that quantifies the
conservation value of the land. It was derived from the aggregation and prioritization of
many conservation programs in the state of Washington according to three primary
conservation themes: habitat for species of concern, habitat for species of economic and
recreational importance, and native and un-fragmented habitats. In general, lands with a
crucial habitat rank of 1 are considered to be most crucial habitats and are lands with the
highest conservation value. At the other end of the spectrum, lands with a crucial habitat
3
rank of 6 are considered to be the least crucial habitats and are lands with the lowest
conservation value. With this new environmental indicator, the interaction between
habitat conservation and renewable energy resources can now be better understood within
a landscape perspective. By understanding spatial distributions, interactions, as well as
risks and opportunities for future renewable energy development, land use planning
concerning renewable energy development and habitat conservation can begin to be
approached from a multifunctional perspective and work toward optimizing the lands use
among both.
To explore how management of renewable energy development and habitat
conservation might be approached from a multifunctional landscape perspective, this
study investigates the research question “How do wind and solar energy development and
habitat conservation priorities conflict with one another in Washington State?” This is
achieved within the contexts of three different spatial perspectives: existing wind farms,
suitable wind and solar energy development locations, and Washington Habitats. These
contexts were chosen to enable an understanding of the levels of conflict from past
development actions, of future development actions, and to better understand specific
impacts to certain habitat types.
From the perspective of existing wind farms, this study informs of how well or
poorly the 20 existing Washington wind farms have been sited according to landscapelevel conservation priorities. Further, from the perspective of suitable wind or solar
energy development lands, this study identifies the risk of landscape conflict for future
energy development and the potential to optimize the land use between these two
initiatives. Finally, from the perspective of Washington habitats, this study explores the
4
risk or opportunity of future energy development within different habitat types. Using
Geographical Information Systems (GIS) as the method of analysis, this study found that
in general there is a moderate to low landscape level conflict between habitat
conservation priorities and renewable energy development in Washington State. This
thesis will argue that the findings of this research produce new and much needed
information for Washington land managers as they attempt to identify and respond to
conflicts between these two beneficial land uses.
There are six chapters in this thesis. The present chapter introduces the topics of
this research and begins to frame the importance of investigating the landscape conflicts
between renewable energy development and habitat conservation. The second chapter
continues to frame the importance of this research by providing historical background
and a review of the existing scientific literature illuminating our current knowledge of the
themes explored in this research. Focus is given to the various fields that have
contributed to defining habitat conservation, identifying trends in renewable energy
growth, investigating environmental impacts associated with wind and solar energy
development, and exploring the potential for creating multifunctional landscapes. This
review establishes the specific need to understand the levels of conflict between
renewable energy development and habitat conservation at a landscape-level perspective.
Chapter 3 identifies the state of Washington as the research study area and
outlines the five specific research questions and hypotheses that were analyzed as part of
this study. These specific research questions guide the investigation of the levels of
landscape conflict within the three spatial contexts identified above. These are the
following:
5
Existing wind farms
1. How do crucial habitat distributions in existing Washington wind
farms compare to the crucial habitat distributions across the entire
state of Washington?
Suitable wind and solar development lands
2. How do crucial habitat distributions in suitable wind and solar
development lands compare to the crucial habitat distributions
across the state of Washington?
3. At what levels of crucial habitat could future wind and solar
energy development be restricted in order to both protect habitat
quality and contribute substantially to future Washington energy
production?
Washington habitats
4. Which habitat types are more suitable for future wind and solar
energy development in Washington State?
5. What is the risk of significant landscape conflicts between crucial
habitat and wind or solar resources within those habitats?
The remainder of this chapter describes the research methodology using geographical
information systems (GIS). The many data sources utilized in this research are defined
and the GIS and statistical analyses conducted in this study are outlined, including basic
spatial analyses and local Moran’s I spatial autocorrelation analyses.
6
Chapter 4 presents the results of the GIS and statistical analyses for each of the
five specific research questions. Chapter 5 provides a thorough discussion of the study
results and associated opportunities and implications of utilizing the crucial habitat
variable as a land use planning indicator. This includes exploring ways to reduce land use
conflict, exploring ways to begin optimizing landscape planning in an effort to obtain
multifunctional energyscapes, and identifying specific opportunities and risks of wind
and solar energy development within different habitats. Cautions associated with the
study results are also discussed as well as opportunities for future research and
improvements to the methods of analysis as conducted in this study. The end of this
chapter concludes this study by presenting a concluding summary of this study and final
thoughts on the importance and capability of working toward establishing and managing
landscapes according to a multifunctional model.
7
Chapter 2: Literature Review
2.1 Introduction
This chapter presents the most relevant background, history, and current research
surrounding the study themes of habitat conservation, wind and solar renewable energy
resources and the implications to wildlife and the environment, and the concept of
reducing landscape conflicts by creating multifunctional landscapes. Having an
understanding of the philosophies, history, motivations and trends, current scientific
progress, and future directions of study that influence each of these themes begins to
frame the importance of this research. After a thorough review of the literature, gaps in
knowledge and opportunities for future research surrounding these important topics
emerges. With this information, this study aims to address the identified gaps in
knowledge and contribute to the development of landscape planning and management
philosophies and techniques.
2.2 Conservation of Habitats
Natural habitats have become fragmented across the landscape and even lost
completely as a predominant effect of anthropogenic land use change (Trombulak et al.
2004; Bennett and Saunders 2010). This has had catastrophic impacts on local
biodiversity and is contributing to rapid rates of extinction throughout the world (Pimm
and Jenkins 2010). In an effort to guard against these fatal consequences, the field of
conservation biology was established as a mission-oriented science working toward
halting and reversing the ecological damage caused by humans and their interaction with
8
the environment. Only in the mid-1980s has conservation biology developed as an
independent, interdisciplinary science with primary goals to maintain biodiversity,
ecological integrity, and ecological health (Trombulak et al. 2004; Fetene, Yeshitela, and
Desta 2012; Meine 2010). During this time, philosophies and fundamental concepts from
several other areas of study within the biological sciences were incorporated into the
foundation of conservation biology (Meine 2010). Population dynamics, island
biogeography, and landscape ecology are key fields of study that inform conservation
efforts centered on habitat conservation. These fields of study are particularly important
to understand when considering how renewable energy development may impact the
landscape.
Population Dynamics
The population size and demographics of a species is a critical component in
assessing the ecological integrity of a specific landscape and is often used as a critical
indicator in managing the effects of landscape change, habitat loss, and threat of species
extinction. Ecological integrity refers to the “degree to which an assemblage of
organisms maintains its composition, structure, and function over time” (Trombulak et al.
2004, 1181). There are many components involved in assessing the ecological integrity of
an area. However, population dynamics plays a critical role in understanding the
relationships between anthropogenic landscape change and the habitat requirements for a
species or biological community to persist and ecosystems to function (ed. N. S. Sodhi
and Ehrlich 2010).
9
Population size of a species in an area is generally dependent on the interaction of
four factors: births, deaths, immigration, and emigration. If these factors interact in such a
way that species population levels become too low, the species can become at risk of
extinction and important environmental services can diminish. This can trigger a negative
cascading effect, impacting many other organisms in the ecosystem, including humans
(Pimm and Jenkins 2010). Understanding how the environment impacts species
demographics and changes in population growth is a fundamental goal of population
ecology. Much research has been conducted modeling the complexity of population
dynamics in an effort to predict the effects of environmental and demographic change on
species populations (Schaub and Abadi 2011). As population modeling becomes more
integrated to incorporate numerous environmental variables, more reliable insights into
specific drivers of population growth and decline can be achieved within various
environmental settings. This is essential to effective management of biological
conservation efforts (Schaub and Abadi 2011).
The number one cause of population decline and threat to species extinction on
local and global scales in more recent times is habitat loss from anthropogenic land use
changes (Trombulak et al. 2004; ed. N. S. Sodhi and Ehrlich 2010; Hanski 2011). The
threat of extinction becomes even more concerning since the rates of extinction have
been occurring at faster rates than observed during any previous time in history
(Trombulak et al. 2004; Pimm and Jenkins 2010). Many conservation activities focus on
preserving and increasing available habitat to stabilize declining species populations,
both in local and global regions, to avoid the detrimental impacts associated with species
extinction. If a species becomes extinct on a global scale, these effects become
10
irreversible; biodiversity is altered and often reduced indefinitely (Trombulak et al. 2004;
ed. N. S. Sodhi and Ehrlich 2010). On a larger scale, these conservation efforts combined
contribute to the overarching conservation biology goals to maintain the biodiversity
within a landscape, thereby also maintaining the ecological health of that area.
While the focus on biodiversity and ecological health through population
management is certainly a fundamental goal within conservation biology, there is some
debate within the field surrounding the approach to ecological policy and management. In
a paper written by Robert Lackey (2007), a professor of fisheries science at Oregon State
University and prior member of the U.S. Environmental Protection Agency’s national
research laboratory, perspectives on scientific advocacy and affiliation to policy
preferences are explored. Lackey points out that the opinions of some conservation
biologists is to manage the environment strictly from the perspective that ecosystems
unaltered by anthropogenic influences is inherently good and preferable to those changed
by humans. Further, he describes the following environmental policy preferences as
common: “human-caused extinctions are inherently bad and should be avoided; unaltered
ecosystems are preferable to altered; reducing complexity in ecosystems is undesirable;
natural evolution is good, human intervention is not; more biological diversity is
preferable to less biodiversity; and native or indigenous species are preferable to nonnative species” (Lackey 2007, 14). While these perspectives are often true, there have
been cases where the inverse has proven to be beneficial for particular species’
populations, resulting in thriving ecosystems. Nature is a complex entity that requires
careful study and analysis to determine the most appropriate methods of management
concerning the interaction between species population dynamics and human land use, to
11
ensure the optimal state of environmental health for both. Concerning population
dynamics, this effort is reflected in the continued research around population models to
better understand the specific environmental factors that affect species populations in
specific areas.
Island Biogeography
In addition to the field of population ecology, the field of island biogeography
also contributes substantially to the growing body of knowledge and efforts associated
with conservation biology. Originally presented by MacArthur and Wilson, the theory
and study of island biogeography refers to the relationship of species population and
richness in a relatively confined geographical area (such as an island) and the related
geometric size and level of isolation of those lands (MacArthur and Wilson 1967;
Bennett and Saunders 2010). This field of study has made important contributions to
conservation biology when studying the effects of habitat fragmentation on the
population dynamics and species area relationships for individual species as well as entire
biological communities (Bennett and Saunders 2010; Campos et al. 2012; Campos et al.
2013).
When humans develop natural lands, major landscape changes may occur, leaving
little natural habitat. The habitat that does remain is often fragmented into small pieces
and scattered across the newly developed anthropogenic landscape (Bennett and Saunders
2010). These habitat fragments often exhibit habitat characteristics similar to those
observed in habitats on islands, hence the connection to island biogeography. However,
the population dynamics in these habitat fragments are different from actual islands in
12
that the space separating the habitat fragments are not barren. In this environment, it is
often easier for non-resident species to access the habitat fragments through the adjacent
lands, which can make it harder for resident and specialist species to persist with this
additional competition (MacArthur and Wilson 1967).
Habitat fragmentation is often viewed as a negative environmental impact of
anthropogenic land use since habitat is reduced and the scattered nature of the remaining
fragments can have negative implications to species population dynamics (Bennett and
Saunders 2010; Laurance 2010; Hanski 2011; Campos et al. 2013). Studies have shown
that, as natural habitat is reduced in size, the capacity of the habitat to support species
sustenance and breeding requirements is also reduced and leads to biodiversity loss
(Bennett and Saunders 2010; Lloyd, Campbell, and Neel 2013). However, there is an
ongoing debate within the field of ecology concerning the effects of habitat fragmentation
on species within the landscape (Villard and Metzger 2014).
While there is a clear relationship between habitat loss and biodiversity loss,
habitat fragmentation has proven to have both positive and negative effects on different
species according to specific species’ habitat requirements, as well as the landscape
configuration of habitat (Villard and Metzger 2014). In a study conducted by Rueda et al.
(2013), seven specialist forest bird species were used in a modeling exercise to
understand the impacts of habitat fragmentation on species populations. The results of the
study show that four of the seven bird species were negatively affected by habitat
fragmentation exhibiting population loss and eventual extinction. However, contrary to
these results three of the bird species showed a positive affect from habitat fragmentation,
exhibiting population growth in and among multiple habitat fragments on the landscape
13
(Rueda et al. 2013). These species were observed to have higher dispersal capabilities,
allowing them to more easily move between habitat fragments so that they could thrive
within the edges of the remaining natural habitat, where predators were not as prevalent
(Bennett and Saunders 2010; Rueda et al. 2013). As the main finding of the Rueda et al.
(2013) study, habitat fragmentation impacts can vary depending on species’ sensitivity to
landscape change. This study reinforces the idea that, while many species may be
negatively impacted by habitat fragmentation, this general assumption cannot be applied
to all species and to all habitat configurations across a landscape.
Depending on the distance between habitat fragments across a landscape, some of
the more subtle characteristics of population dynamics help inform conservation biology
on the importance of landscape configuration and habitat connectivity within a
fragmented landscape (Villard and Metzger 2014). Metapopulation dynamics and the
importance of genetic variation help conservation biologists better understand the
ecological risks that can be associated with habitat fragmentation. When habitat
fragments become isolated from other, similar habitats, this may hinder the immigration
and emigration within species’ metapopulations, restricting gene flow and reducing
genetic variation (Gotelli 2001; Trombulak et al. 2004; Hanski 2011; Bennett and
Saunders 2010).
A metapopulation is a group of local subpopulations of a particular species where
each of the subpopulations exhibit a genetic makeup that is different from the others, but
are linked to one another through immigration and emigration (Gotelli 2001; Bennett and
Saunders 2010; Hanski 2011). When genetic variation associated with population
movement is lost, the isolated subpopulation will become less able to adapt to changing
14
environmental conditions and negative effects associated with inbreeding may become
more frequent (Trombulak et al. 2004; Bennett and Saunders 2010; Shirk et al. 2010;
Lloyd, Campbell, and Neel 2013). This weakens the resilience of the subpopulation and
further increases the chance of population loss and extinction within the fragmented
habitat.
Researchers have begun to use gene flow monitoring and dispersal capability as
indicators to assess the impacts of landscape fragmentation and anthropogenic barriers on
threatened species in specific geographic locations. These studies assess the risks and
monitor the effects of anthropogenic landscape change in those areas. Such studies also
inform of the necessity for conservation management to mediate negative landscape
effects (Shirk et al. 2010; Lu et al. 2012; Lloyd, Campbell, and Neel 2013). With this
information, conservation biologists can more effectively understand the value of
maintaining large habitat areas as well as re-establishing habitat corridors to connect
different habitat fragments across the landscape (Bennett and Saunders 2010; Lu et al.
2012). In this way, conservation biology efforts, in concert with landscape planning and
management, can work to reduce the hazards of habitat fragmentation and improve future
landscape designs.
Landscape Ecology
Landscape ecology is the third scientific field of study that informs conservation
biology how to best protect wildlife habitat in an effort to prevent local species extinction
and minimize the loss of biodiversity from the landscape. While the field of landscape
ecology has been interpreted and applied to many different disciplines in many ways
15
(Kirchhoff, Trepl, and Vicenzotti 2013), conservation biology uses the term to represent a
spatial configuration of habitat within a particular geographical area. Multiple habitat
types, ecosystem types, as well as many habitat fragments and connecting corridors are
typically present within a single landscape area (Bennett and Saunders 2010; Villard and
Metzger 2014). It becomes important to understand and manage conservation activities at
the landscape level because species often exist in, use, and move between the multiple
ecosystems and habitat fragments in a particular landscape (ed. N. S. Sodhi and Ehrlich
2010; Mueller et al. 2011).
While many conservation activities tend to focus on the species level, a
landscape-level approach to conservation has proven to be more effective in managing
the overarching goals of conservation biology—biodiversity, ecosystem integrity, and
ecosystem health (Trombulak et al. 2004; ed. N. S. Sodhi and Ehrlich 2010). In landscape
ecology, habitat fragments are assessed and compared across the whole landscape, and
the most appropriate conservation strategies can then be applied for maximum
conservation efficiency (Fetene, Yeshitela, and Desta 2012). However, managing
conservation activities at a landscape level is particularly challenging because measuring
and tracking landscape biodiversity and other environmental variables is very difficult to
achieve. As the scope of management and analysis increases to reach a landscape-level,
the number of interacting environmental variables also increases, emerging as a complex
system with much uncertainty (Waltner-Toews, Kay, and Lister 2008).
Despite the challenges, understanding and managing conservation at the
landscape-level will become increasingly important as human development continues to
occur, climate change increases the rate of environmental change, and many of the
16
species at risk of extinction come to depend upon conservation management for
population viability (Trombulak et al. 2004; ed. N. S. Sodhi and Ehrlich 2010; Bellard et
al. 2012; Goble et al. 2012). However, managing conservation initiatives at the
landscape-level becomes even more challenging when landscape uses have both a
societal and ecological value. This is particularly evident with the growth and
development of renewable energy technologies.
2.3 Renewable Energy Development
Between the years 2000 and 2010, greenhouse gas emissions (GHG) produced
from worldwide energy production have increased by 47% from the burning of fossil
fuels (Intergovernmental Panel on Climate Change 2014). As climate change challenges
continue to grow in importance at a global scale, climate mitigation strategies have
become critical to managing human interaction with the natural environment and stabilize
the global warming trend that has been occurring over the past several decades. If
successful, extreme shifts in Earth’s natural systems can be avoided, lessening the need
for both ecological and social systems to adapt quickly to a rapidly changing environment
(Intergovernmental Panel on Climate Change 2007).
According to the Intergovernmental Panel on Climate Change (IPCC), renewable
energy technologies have become known for their “large potential to displace emissions
of greenhouse gases from the combustion of fossil fuels and thereby mitigate climate
change” (Intergovernmental Panel on Climate Change 2012, 1). In fact, according to the
IPCC climate mitigation scenarios, a tripling to nearly quadrupling of renewable energy
production would be required to achieve low-stabilization atmospheric GHG levels by
17
2100 (Intergovernmental Panel on Climate Change 2014). In addition, the decarbonization of the energy production systems through increased renewable energy
production is also viewed as a key component of the most cost-effective climate
mitigation strategies (Intergovernmental Panel on Climate Change 2014). Because
renewable energy production is environmentally important and is viewed to be costeffective, a shift is now underway toward integrating renewable energy production into
landscapes on a global scale.
Renewable energy is considered to be clean energy, producing little or no carbon
emissions and having inexhaustible primary energy resources (Demirbas 2009). These
energy sources include biomass, solar, wind, geothermal, hydropower, and the oceans.
While renewable energy still only accounts for about 14% of global energy use today,
renewable energy development has been increasing rapidly over the past several years
(Demirbas 2009; US Department of Energy 2011; Intergovernmental Panel on Climate
Change 2012). This is especially true for wind and solar technologies, with an annual
average global growth rate of 40% and 27% respectively over the past decade
(International Energy Agency 2012; Intergovernmental Panel on Climate Change 2014).
Rapid growth in solar and wind renewable energy technologies is not unexpected,
since both of these resources have a tremendous global supply and are available in nearly
every country and region of the world. Additionally, economic cost barriers have been
declining and political support has encouraged growth in these industries (Demirbas
2009; Intergovernmental Panel on Climate Change 2012). For these same reasons, it is
expected that solar and wind energy development will continue to exhibit rapid growth.
Furthermore, various renewable energy growth models have shown that solar and wind
18
energy will likely be large contributors of renewable energy production in the future
(Demirbas 2009; Intergovernmental Panel on Climate Change 2012).
In the United States, renewable energy production has exhibited growth trends
similar to those observed globally. Total renewable energy contribution in the U.S. was
11.7% in 2011 (US Department of Energy 2011) and wind and solar electricity
generating capacity has seen the most growth over the past decade; installed wind energy
capacity increased by a factor of 10 from 2001 to 2011 and installed solar energy
capacity increased by a factor of 12 from 2006 to 2011 (US Department of Energy 2011).
Future growth in these renewable energy resources is also expected to increase for the
following three reasons. 1) There are ample wind and solar resources in the United States
(Lopez et al. 2012). 2) Many states have set specific Renewable Portfolio Standard (RPS)
growth targets for solar and other renewable energy technologies (US Department of
Energy 2010; US Department of Energy 2013). 3) Most states have incentive and rebate
programs to encourage renewable energy development (US Department of Energy 2010;
US Department of Energy 2013). As renewable energy technology development grows,
this means landscape changes will occur and negative environmental impacts are likely to
be incurred.
2.4 Environmental Impacts of Renewable Energy Development
As mentioned in the prior section, it is commonly understood that renewable
energy generation is immensely beneficial to the environment due to the GHG emission
reductions. However, renewable energy technologies are not entirely beneficial. There
are known negative impacts to the environment that coincides with the development of
19
renewable energy technologies. These can occur throughout the entire technological
lifecycle, including resource acquisition, initial manufacturing, energy production
development on the landscape, continued maintenance, and finally deconstruction and
removal from the landscape (Burger and Gochfeld 2012; Intergovernmental Panel on
Climate Change 2012; Athanas and McCormick 2013). While the environmental impacts
at all lifecycle stages are important, those specifically associated with the energy
production landscapes from development, continued maintenance, and deconstruction is
the primary focus of this thesis. To understand these negative environmental impacts, a
comprehensive review of the ecological footprints (amount of space required, type of use,
and conversion factor) and impacts to habitat and biodiversity for each specific renewable
energy technology is required (Burger and Gochfeld 2012).
Infrastructure: Electrical Transmission Lines and Roads
To some extent, all renewable energy technologies will have common
infrastructure components, including electrical transmission lines and roads. Studies have
shown that the addition of this infrastructure to the landscape can cause varying degrees
of negative environmental impacts. However, the extent of environmental impacts
depends on the size of the renewable energy development, site habitat type, and local
ecology (Trombulak and Frissell 2000; Kuvlesky Jr. et al. 2007; McDonald et al. 2009).
Electrical transmission lines present a smaller environmental risk than roads, but
have been known to cause bird mortality from risks of collision and electrocution
(Kuvlesky Jr. et al. 2007). Since this infrastructure is critical to move energy from
production to consumption locations, all renewable energy developments could exhibit
20
some level of associated environmental impacts. However, the extent of negative
environmental impacts from electrical transmission line construction will likely depend
on the proximity to specific bird populations and migration paths, as well as the quantity
of transmission lines required for the renewable energy development (Kuvlesky Jr. et al.
2007; McDonald et al. 2009). For example, greater environmental impacts would be
expected for renewable energy developments located in or near raptor migration routes
and habitat, since raptor populations cannot absorb mortality as easily as other bird
species (Kuvlesky Jr. et al. 2007; Intergovernmental Panel on Climate Change 2012).
Further, wind farms would be expected to incur greater environmental impacts from
transmission lines than solar because of the greater landscape requirements for wind farm
facilities. Each wind turbine must be spaced approximately 100meters to 250 meters
apart (2—6 blade widths) necessitating more transmission lines, whereas solar panels can
be placed directly next to one another (Manwell, McGowan, and Rogers 2009; McDonald
et al. 2009; Nelson 2009).
Road construction presents a more complex environmental risk than transmission
lines due to the many potential ecological effects imposed by roads on the landscape.
Trombulak and Frissell (2000) completed an extensive review of scientific literature
regarding the ecological effects of roads and found several general concerns—mortality
from construction, mortality from vehicle collision, animal behavior modification,
physical environment alteration, increased spread of exotic species, and increased land
use by people. All of these effects have some level of impact on the habitat, biodiversity,
and species presence and movement patterns in the area where roads are constructed.
21
As the number of roads increases in an area, certain species become threatened
with mortality from the physical act of clearing and construction. Additionally, natural
habitat becomes fragmented, creating a barrier to natural dispersal for some species and
altering movement behaviors for others (Trombulak and Frissell 2000; Kuvlesky Jr. et al.
2007; Intergovernmental Panel on Climate Change 2012). Increasing the network of
roads will also increase the human traffic in an area, amplifying the risk of vehicle
collisions, as well as changing the local biodiversity, including a greater risk for the
spread of exotic and invasive species in native habitats (Gelbard and Belnap 2003;
Trombulak and Frissell 2000). Any combination of the above effects related to the
increase in road networks has been shown to have some level of impact on habitat quality
and quantity, biodiversity loss, genetic isolation of local populations, and population
decline (Trombulak and Frissell 2000; Kuvlesky Jr. et al. 2007). Again, due to the
necessity of transmission lines and road construction, all of the above-mentioned
environmental concerns should be considered during the siting and planning phases of
any renewable energy development. This can ensure a thorough understanding of the
specific risks to the local habitat and ecology.
Wind Energy Development
Wind energy facilities, often known as wind farms, have unique environmental
impact concerns according to the ecological footprint of the facilities and nature of the
technology employed. Wind energy technologies have been found to mainly impact the
ground surface of a landscape as well as the “airshed,” (i.e., the space from the ground to
the space above the turbine blades of a particular site) (Burger and Gochfeld 2012).
22
While the infrastructure footprint of the wind turbines is only about 3—5% of the total
wind farm site, the total land required for wind farm facilities can be quite expansive
because of the space required between turbines for maximum turbine efficiency
(McDonald et al. 2009; Intergovernmental Panel on Climate Change 2012). This tends to
cause environmental impacts such as habitat fragmentation and related biodiversity loss,
species behavior modification, and direct mortality from the turbine blades (McDonald et
al. 2009; Intergovernmental Panel on Climate Change 2012; Northrup and Wittemyer
2013).
Habitat fragmentation, biodiversity loss, and behavior modification of species
related to wind farm development are largely connected to the expansive networks of
roads required to construct and access each individual wind turbine in the wind farm
(Kuvlesky Jr. et al. 2007; Intergovernmental Panel on Climate Change 2012). As
discussed in the section above, roads can have many impacts on the local ecology of an
area and are often observed as a negative effect of wind farm development. In addition to
road construction, species behavior modification has also been attributed to the acoustic
noise and vibration from turbine operation (Northrup and Wittemyer 2013). In general,
the severity of these environmental impacts will vary substantially depending on the type
and sensitivity of local habitat as well as the local biodiversity of each individual wind
farm location.
Direct mortality of birds and bats from collision with turbine blades is an
environmental impact unique to wind farm facilities (Kuvlesky Jr. et al. 2007; Arnett et
al. 2008; Intergovernmental Panel on Climate Change 2012; Northrup and Wittemyer
2013). Many studies have been conducted over the past 20 years to better understand the
23
impacts of wind farms on bat and bird populations. Findings have shown there is a
significant increase in mortality for both species in relation to wind farm developments
(Barclay, Baerwald, and Gruver 2007; Arnett et al. 2008). However, the severity of
environmental impacts again relates to the specific ecology of each independent wind
farm location as well as the specifications of turbine construction, operation, and
arrangement.
The more pressing concerns related to bird and bat mortality are often anchored
around species-specific impacts and proximity to migration paths (Kuvlesky Jr. et al.
2007; Arnett et al. 2008). Mortality of raptors and some less resilient endemic bird
species is generally of greater concern because their population levels and global
presence are less able to absorb mortality than other, more prolific species (Kuvlesky Jr.
et al. 2007; Intergovernmental Panel on Climate Change 2012). Likewise, proximity to
bird and especially bat migration paths is of great concern because of the increased rate
of mortality as opposed to sites away from migration routes (Barclay, Baerwald, and
Gruver 2007; Arnett et al. 2008).
While there are several known environmental impacts associated with wind farm
developments, as discussed above, there are also many unknown environmental impacts.
Additionally, the overall biological significance of each environmental impact is unique
to each specific ecological landscape and remains largely unclear (Intergovernmental
Panel on Climate Change 2012; Park, Turner, and Minderman 2013). Much of the
literature expresses a clear need for more extensive studies that aim to understand the
impacts of wind farm development related to specific habitats and ecological impacts
24
(Intergovernmental Panel on Climate Change 2012; Northrup and Wittemyer 2013; Park,
Turner, and Minderman 2013).
Solar Energy Development
Like wind farms, solar energy facilities also have unique environmental impact
concerns according to the ecological footprint of the facilities and nature of the
technology employed. Solar farms generally impact the surface and subsurface
components of the landscape according to solar panel placement and connection to
groundwater systems for facility cooling (Burger and Gochfeld 2012; Intergovernmental
Panel on Climate Change 2012). As a general benefit of solar energy development, land
requirements for construction are often lower compared to the requirements of other
renewable energy facilities. However, development often requires utilization of the entire
landscape sited for the facility, impacting nearly 100% of the physical area (McDonald et
al. 2009). While there is a clear understanding of the ecological footprint associated with
solar farms, a crucial limitation to understanding the associated environmental impacts of
solar farms is the serious lack of peer reviewed literature investigating the topic
(Intergovernmental Panel on Climate Change 2012; Northrup and Wittemyer 2013).
However, there have been speculations into the types of environmental concerns that are
likely to occur from solar farm development including habitat loss, fragmentation, and
potentially microclimate alteration around the solar arrays (Northrup and Wittemyer
2013).
Construction of solar farms typically requires the clearing of all land within the
site since solar panels are typically placed close to one another (McDonald et al. 2009).
25
This can be seen in Figure 1 below, an image of the solar and wind facilities at the Wild
Horse Wind Facility in Ellensburg, Washington. Depending on the size of the solar
facility and location of the site, this could have important effects on local ecological
systems. These could include the loss of food production areas and biodiversity reduction
from habitat reduction (Burger and Gochfeld 2012). Also, depending on the solar facility
distribution, habitat fragmentation can occur from the increase in required road networks
and general placement of the solar arrays (Intergovernmental Panel on Climate Change
2012). Finally, concerns about microclimate alteration resulting from solar panel heat
loss have been theorized. However, scientific studies are needed to better understand the
significance of this concern and the technicalities of disturbance (Intergovernmental
Panel on Climate Change 2012; Northrup and Wittemyer 2013). The severity of these
environmental concerns will again depend on the unique ecological and habitat
characteristics of the solar development sites. However, since attractive solar farm siting
locations often occur in ecologically sensitive desert habitats, environmental impacts of
solar farms may be greater than we have anticipated (Burger and Gochfeld 2012; IPCC
2012).
26
Figure 1. Wild Horse Wind and Solar Facilities (Ellensburg, WA)
Image taken by Puget Sound Energy:
https://www.flickr.com/photos/pugetsoundenergy/4494987410/
Environmental Impact Assessments
Many countries, including the United States, require the completion of an
environmental impact assessment (EIA) as part of the permitting process for energy
development (Jay 2010; Intergovernmental Panel on Climate Change 2012). The EIA
process functions to assess potential impacts to the environment from development along
with mitigation strategies, when possible (Jay 2010). While being a regulatory control in
place that encourages the protection of the environment from the effects of significant
development projects, the environment can still be negatively impacted and development
does not always occur within least impactful alternative.
Several studies have found problems with the EIA process, including poor-quality
EIAs, failure to conduct an EIA in some cases, and EIAs that fail to identify mitigation
plans (Jay 2010; Athanas and McCormick 2013; Vandergast et al. 2013). Athanas and
27
McCormick completed a review of 195 active renewable energy development project
EIAs and Strategic Environmental Assessments (an SEA is a more rigorous form of EIA)
in the World Bank Renewable Energy Database; a collection of renewable energy
projects in developing countries (The World Bank 2014). Results of the review found that
14% of projects sited were in areas highly likely to have “significant adverse
environmental impacts that are sensitive, diverse, or unprecedented” (2013, 26).
Additionally, 18% of the projects were sited in areas that “have potential adverse
environmental impacts on human populations or environmentally important areas”
(Athanas and McCormick 2013, 26). While these EIAs did identify the environmental
impacts, none of the projects had environmental safeguards included in their development
plans. This means that 32% of the renewable energy projects had been sited in
environmentally sensitive and/or important areas and were actively being developed or in
the pipeline for investment and future development despite the environmental risks.
In some cases, renewable energy projects have been developed on landscapes that
should have been protected and preserved. In another study conducted by Vandergast et
al. (2013), evolutionary hotspots were analyzed in the Mojave Desert, California, in
relation to existing and proposed renewable energy development projects. Evolutionary
hotspots are important environmental regions that contain an overlap of species with high
genetic divergence and diversity. This collection of species is vital to preserving genetic
variation within a species’ population and will contribute to improving species resilience
(ed. N. S. Sodhi and Ehrlich 2010; Vandergast et al. 2013). Landscapes with evolutionary
hotspots should be preserved to protect the species’ population size and biodiversity in
the region. However, according to Vandergast et al. (2013), renewable energy
28
development projects had the potential to impact 6 of the 10 identified hotspot regions,
given that 10–17% of these hotspot regions overlapped with renewable energy
development sites. The results of this study indicate that, in some cases, there is a conflict
of land use between priority habitat conservation lands and renewable energy
development. Although both forms of land use are important to meet the environmental
challenges of today, management of these environmental challenges has stayed
segregated and sometimes the two are at cross purposes. One approach to reducing this
conflict is to move toward designing multifunctional landscapes.
2.5 Multifunctional Landscapes
Historically, land use has often been centered on a single objective or function,
which was determined by the landowner and supported by private property laws. This
system of land use and resource management gave preference to private property owners
where resources of the commons, such as watersheds and biodiversity, were not managed
according to the greatest public interest but left to the management of the individual
landowners (Vejre et al. 2012). In more recent times, human population growth has
transformed the landscape into a patchwork of both private and public lands with many
different functions across the landscape. This growth and continued land development
has spurred the regulation and collective management of some aspects of common
resources such as water quality and protection of important habitats such as wetlands.
However, these regulations are targeted to specific environmental concerns, and
landscape management is still often approached from a single-function perspective (Vejre
et al. 2012). With the pressures of continued development, in concert with land scarcity,
29
some scientists and sustainability advocates have begun focusing on constructing
multifunctional land use strategies. These strategies serve many functions or objectives
and work to consider ecological, cultural, and economic values on the land (Lovell and
Johnston 2009; Groot, Jellema, and Rossing 2010; Vejre et al. 2012).
Multifunctional land use strategies have typically been associated with
agriculture, where the extent of functional integration tends to be a patchwork of various
land use systems without an emphasis on full integration across an entire landscape
(Harden et al. 2013). This occurs due to the challenges associated with a system where
land functions compete with one another and includes many different stakeholders,
landowners, land use drivers (i.e., environmental, social, or economic), and trade-offs
(Groot, Jellema, and Rossing 2010; Vejre et al. 2012; Harden et al. 2013). The conflicts
among these various challenges create a complex situation that requires a lot of research,
assessment, and communication to effectively plan, design, and manage multifunctional
landscapes (Reyers et al. 2012; Harden et al. 2013; Howard et al. 2013). The extreme
commitment and investment of time, money, and resources required to work through
these conflicts often prevents multifunctional landscape designs from transcending
property boundaries and including large portions of the landscape (Harden et al. 2013).
Many land managers and scientists in the field of landscape ecology have
explored various frameworks to guide the transition from segregated functional land use
to a multifunctional model. These frameworks often emphasize completing an extensive
environmental assessment to understand the ecological state of the landscape and risks of
land changes, identifying land use objectives including trade-offs and potential synergies,
as well as continually addressing stakeholder and land owner resistance (Waltner-Toews,
30
Kay, and Lister 2008; Reyers et al. 2012; Vejre et al. 2012; Harden et al. 2013).
Additionally, an interdisciplinary approach should be applied so, at least, ecologists and
land managers might work together for a successful transition to a multifunctional
landscape (Lovell and Johnston 2009; Groot, Jellema, and Rossing 2010).
One of these frameworks moving toward a more integrated form of landscape
assessment and management is what Waltner-Toews, Kay, and Lister have called “the
ecosystem approach” (2008). Under this framework, landscape systems are researched
using an interdisciplinary approach and diagramed to gain an understanding of all the
interactions and feedbacks between the landscape variables of interest. Using this
information, a general sense of landscape health can be obtained, negative landscape
interactions can be identified, and mitigation techniques can be explored to better
optimize the interactions within the landscape system.
The University of Guelph conducted a case study of the ecosystem approach
framework, investigating agroecosystem health in the Great Lakes Basin (WaltnerToews, Kay, and Lister 2008). This study divided landscape variables into two general
categories—ecological and socio-economic—aiding in the development of a landscape
system model. This process engaged all relevant stakeholders to collect information about
the agricultural systems, ecological systems, relevant biodiversity, and history of land use
in the area. The outcome of the landscape system model and stakeholder engagement
identified negative impacts of agriculture on stream health in the Great Lakes Basin
(Waltner-Toews, Kay, and Lister 2008).
With the landscape system model as a starting point, more rigorous analyses were
conducted to understand the specific landscape variables that influenced this negative
31
outcome. The final results of the analysis showed that cattle grazing in stream habitats
were detrimental to fish communities, and that wealthier farms had better stream health
due to a greater ability to protect stream habitats from livestock grazing (Waltner-Toews,
Kay, and Lister 2008). All in all, the ecosystem approach helped researchers and
stakeholders to organize and understand many landscape variables within an
agroecosystem, and to identify the different influences on agroecosystem wellbeing.
However, this approach to landscape management is largely dependent on stakeholder
engagement to take management to the next steps of intervention, mitigation, and
improved policy action, which is viewed as the largest challenge and potentially greatest
weakness of this framework (Waltner-Toews, Kay, and Lister 2008). While applied in a
few case studies, the stakeholder engagement challenge, as well as the time and cost
required for this framework, have prevented successful application of the ecosystem
approach to landscape management in many cases.
Regarding habitat conservation and renewable energy development, there is a
clear potential for land use conflict between these two initiatives. Conservation biology
aims to protect habitats from destruction, fragmentation, and loss of ecosystem services,
while renewable energy developments impose new features on the landscape that in some
cases degrade and fragment natural lands. However, due to the importance of both of
these initiatives, scientists have begun working to identify how landscapes could be
designed and managed to best optimize the land use for both. Howard et al. (2013)
encourages a shift in mindset to think of renewable energy development in terms of what
he calls an “energyscape.” Under this model, planning and developing renewable energy
facilities should aim to find the balance between maximizing energy system requirements
32
and minimizing disruption to ecosystem services and landscape ecology (Howard et al.
2013). Here, a landscape-level management approach is considered, where environmental
impacts are assessed not only for the renewable energy development site but in the
connecting landscape as well.
In a different multifunctional landscape approach, Reyers et al. (2012) include
energy generation as a key component of the landscape function. Reyers et al. argue that
landscapes should be designed to “provide multiple environmental, social, and economic
functions and are able to achieve multiple societal needs including energy and food
production, management of waste, conservation of biodiversity, and the management of
water quantity and quality across the landscape” (Reyers et al. 2012, 1122). Both the
energyscape and the multifunction landscape approach to landscape planning and
management express innovative, interdisciplinary, and systems approaches to tackling the
land use conflicts that have been surfacing between habitat conservation and renewable
energy development. While some multifunctional frameworks, such as the ecosystem
approach explored above, have been applied to landscape management, approaches
related to energy production landscapes are only theoretical at this point and no real
application has been studied.
2.6 Conclusion
Conservation biology is a mission-oriented science that aims to reverse
anthropogenic ecological damage and protect biodiversity, ecological integrity, and
ecological health. Population dynamics as a function of biodiversity management, island
biogeography related to habitat fragmentation and connectivity, and landscape ecology
33
inform the study of conservation biology and are fundamental components of many
wildlife conservation and management strategies. Although these goals are applicable on
a global scale, conservation initiatives tend to focus on individual species and their
related habitats from a local perspective. However, philosophies of landscape ecology
suggest that conservation programs will be more effective in managing these goals when
approached from a landscape-level perspective. Current research looks to expand existing
conservation programs to include a larger landscape perspective when assessing and
managing conservation initiatives.
With regards to renewable energy technologies, there has been substantial growth
in these industries over the past decade, both globally and nationally, in response to
increasing concerns for climate change. Wind and solar energy technologies have seen
the most growth, and many energy researchers forecast continued rapid growth in these
industries. While these technologies do mitigate anthropogenic carbon emissions, there
are still negative environmental impacts associated with development and operation.
Electrical transmission lines threaten wildlife with potential electrocution, road
construction fragments native habitats, and habitat loss occurs from development.
Increased human traffic brings threats of invasive species and wildlife vehicle collisions,
wildlife avoidance has been observed, wind turbines kill birds and bats, and solar panels
could potentially have microclimate impacts. All of these impacts directly conflict with
habitat conservation initiatives.
To manage the potential environmental impacts of renewable energy
technologies, environmental impact assessments are required and appropriate mitigation
strategies should be identified. However, as seen in several studies, this is not always
34
accomplished and serious environmental impacts resulting in grave consequences to
conservation goals have occurred. This is often because land use management has
historically been conducted from the perspective of a single objective or function.
However, more sustainable approaches have been suggested to plan landscapes with a
multifunctional perspective. Multifunctional landscapes would approach land use
management with consideration of the ecological, cultural, economic, and energy
resource values of the land with the goal of improving and optimizing land function.
However, the few attempts at implementing this type of landscape design related to other
industries have been fraught with challenges to achieve stakeholder agreement and costly
environmental assessments to understand how to best design and manage the lands.
Despite these challenges, the concept of the multifunctional landscape approach is still an
important land use aspiration given the increasing land scarcity of current times.
To begin working toward a multifunctional landscape design, land management
first needs to be viewed and analyzed from a landscape-level perspective rather than from
individual land ownerships or independent functional designations. This is something that
has not been readily researched or practiced up to this point (Howard et al. 2013). As
previously mentioned, conservation activities as well as energy development projects
have not been investigated or managed from a landscape spatial scale. A primary reason
for this has been the lack of measureable data to describe the conditions of the land at a
landscape level (Waltner-Toews, Kay, and Lister 2008). Given the importance of both
habitat conservation and renewable energy development, further research needs to be
conducted from this more expansive perspective, to better understand the extent of the
land use conflict that occurs between these two initiatives. With this knowledge,
35
landscape planners and managers will be better able to understand how landscape
management might be approached from a multifunctional perspective. Further, attempts
to balance the landscape requirements and priorities for both environmentally beneficial
initiatives can be explored.
Fortunately, newly released crucial habitat assessment data from the Western
Governors’ Association, now allows a landscape-level assessment of habitat conservation
priorities (Western Governors’ Association 2013). By utilizing landscape-scaled data to
investigate the conflicts between renewable energy development and habitat
conservation, this study investigates land use and design from a new perspective and will
address a gap in the literature. With an understanding of the land use conflicts between
these two initiatives, opportunities and risks associated with conservation and renewable
energy development activities can be identified. Landscape planning can then look into
the possibility of working toward designing multifunctional landscapes where land use is
more optimally managed according to both the ecological and energy resources of the
landscape. The following chapter identifies the specific research design and methodology
that were used in this study, to investigate the opportunities and implications of looking
toward a multifunctional landscape design across the state of Washington.
36
Chapter 3: Methods and Analysis
3.1 Methods
The main objective of this research is to better understand the interaction between
renewable energy development and habitat conservation priorities within a landscapelevel perspective. More specifically, how do wind and solar energy development and
habitat conservation priorities conflict with one another in Washington State? To address
this research question several Geographical Information Systems (GIS) analyses were
performed using wind and solar resource data and the newly released crucial habitat
indicator, among others. As previously mentioned, crucial habitat is a new, nonregulatory environmental indicator that quantifies the conservation value of the land with
the purpose to provide consistent and comprehensive wildlife information to decision
makers assessing landscapes for planning and development. Using this new
environmental indicator, this research provides new knowledge about the opportunities
and implications of looking toward an improved, multifunctional landscape design, where
land use could be more optimally managed according to both the ecological qualities and
energy resources of the landscape. To begin this research, the study area of Washington
State is more specifically described and five specific research questions are defined and
analyzed.
Study Area
Washington State was chosen as the study area for this research because
increasing renewable energy development and improving habitat conservation are both
37
important state goals. Renewable energy production in the United States has been on the
rise over the past decade and as of 2011, Washington State produced the most electricity
from renewable resources than any other state in the United States (US Department of
Energy 2011). While much of this energy came from the state’s large hydroelectric
facilities, the Washington renewable energy standard, passed in 2006, mandates the
continued growth of renewable energy resources across the state. Under this renewable
energy standard, 15% of the state’s electricity consumption is to come from new, nonhydroelectric renewable energy resources by the year 2020 (DESIRE 2013). In order to
meet this renewable energy standard target, renewable energy development is expected to
increase across the state of Washington over the next several years.
While state renewable energy standards are important state policies working
toward climate mitigation, the state of Washington also has an extensive wildlife and
habitat conservation program that is important for protecting, preserving, and restoring
state wildlife and important habitats. Managed by the Washington Department of Fish
and Wildlife (WDFW), the state’s comprehensive wildlife conservation plan represents a
unique approach to wildlife management. This plan not only approaches conservation
with species-specific priorities, but it is also one of the first state conservation plans to
include landscape-level biodiversity considerations as part of its conservation strategy.
As this approach has evolved, the state has increasingly identified and prioritized its
habitat conservation activities to include a diverse array of conservation initiatives
(Washington Department of Fish and Wildlife 2005). This is particularly apparent when
reviewing the various conservation initiatives used to define the crucial habitat indicator
38
for the state of Washington and is discussed in more detail in the “Data Sources” section
of this chapter beginning on page 45.
However, despite both being important and well supported state goals, it seems
reasonable that in some instances these initiatives are likely to conflict with one another
on the landscape due to the contrasting land use requirements. To better understand the
conflict between these two important statewide initiatives, the entire state of Washington
was chosen as the scope of analysis. This is primarily because the data for such an
assessment was available and a statewide scope would include a full representation of
statewide conservation activities as well as lands that could contribute to statewide
energy production.
Research Questions and Hypotheses
To address the research objective of this study within the state of Washington,
five main research questions were defined. First, how do crucial habitat distributions in
existing Washington wind farms compare to the crucial habitat distributions across the
entire state of Washington? This research question illuminates how well or poorly
existing wind farms have been sited from a landscape-level habitat conservation
perspective. As a general hypothesis, it was expected that existing wind farms will have
been developed on less crucial habitat lands. For any energy development proposal, an
environmental impact assessment (EIA) is required to obtain the necessary permits.
While the EIA process is a local assessment unique to each development site, federal and
state environmental policies governing in the EIA process were presumed to have
regulated and protected priority conservation initiatives. This process leads to the
39
expectation that existing wind energy development on the most crucial habitat lands has
been minimized.
An analysis of crucial habitat in existing Washington solar farms was not included
as part of this study. According to the Northwest Power and Conservation Council, there
is only one existing commercial solar farm in the state of Washington—part of the Wild
Horse renewable energy facility in Ellensburg, Washington (Northwest Power and
Conservation Council 2013). Given the large effort that would be required to complete
this analysis for a single solar farm and lack of existing data, it was excluded as part of
this study.
Second, how do crucial habitat distributions in suitable wind and solar
development lands compare to the crucial habitat distributions across the state of
Washington? And, third, at what levels of crucial habitat could future wind and solar
development be restricted in order to both protect habitat quality and contribute
substantially to future energy production? The second research question will inform of
the risk of landscape conflict for future wind and solar energy development. Then, the
third research question estimates potential annual energy contributions according to the
inclusion or exclusion of specific crucial habitat levels. This will inform of the
opportunity or challenge of optimizing the land use between future wind and solar
development and habitat conservation. As a general hypothesis it was expected that the
distribution of crucial habitat levels across suitable wind and solar energy development
landscapes is similar to the distribution of crucial habitat statewide. It was also
hypothesized that at least lands with the highest conservation value (most crucial
habitats—level 1) could be placed off limits to future wind and solar development while
40
still allowing substantial future renewable energy production. Only 15% of the
Washington landscape is rated crucial habitat level 1, leaving 85% of the state to be
considered for future wind or solar energy development.
Fourth, which habitat types are more suitable for future wind and solar energy
development in Washington State? And, fifth, what is the risk of significant landscape
conflicts between crucial habitat and wind or solar resources within those habitats? These
research questions categorize all lands in the entire state of Washington into five general
habitat types and then identify those with high wind and solar resource potential, as well
as high or low landscape conflicts between wind and solar energy resources and habitat
conservation priorities. This will inform of the risk or opportunity of future energy
development within specific habitat types. Since statistical methods are used to address
this research question, there is a null hypothesis stating that there is no spatial clustering
or interaction between wind and solar energy resources and crucial habitat within specific
habitat types. The alternative hypothesis is that there will be some significant spatial
clustering and interactions between the variables within specific habitat types. It is
expected that there will be significant spatial clustering and interactions since the crucial
habitat and energy resource variables are highly influenced by the contiguous lands as
they change in value over the landscape.
To investigate these research questions and test these hypotheses, data were
analyzed using basic spatial analyses as well as spatial autocorrelation analyses, using the
Moran’s I statistic. The next section describes the methods used to conduct this research.
41
Methods
For this study, spatial analyses were completed using the ArcGIS 10.2 for
Desktop computer program (Esri Inc. 2013). All data sources incorporated into the
analyses were projected into the USA Contiguous Albers Equal Area Conic USGS
version coordinate system and transformations were performed when necessary to use the
NAD 1983 datum. The Albers Equal Area Conic coordinate system was chosen in order
to reduce the projection distortion for geometric area across the state of Washington
(Bolstad 2012). Proper geometric area is an important spatial element to maintain, since
this study takes into account landscape area suitable for wind and solar resources for
potential future development. The NAD 1983 datum was chosen as it has become the
standard datum used in most current GIS analyses (Bolstad 2012).
Data Sources
To address the identified research questions, this study has incorporated many
different data sources to complete the GIS spatial analyses. The following section
describes these data sources in detail:
Wind and Solar Data Resources
The wind and solar GIS data used in this study are publicly available from the
National Renewable Energy Laboratory (NREL) of the U. S. Department of Energy
(National Renewable Energy Laboratory 2013a; National Renewable Energy Laboratory
2014). The NREL has focused on renewable energy and energy efficiency research and
42
development over the past 35 years. Its history of many major accomplishments has
established the NREL as a trusted, leading expert in this field of study.
Wind Power Density Dataset
The NREL wind power density data were derived from over 3,200 wind resource
assessment stations set up across the United States, beginning in 1979, to measure daily
wind speeds within heights of 10 to 60 meters above ground surfaces. Wind resource
measurements were then summarized and extrapolated into 0.25° latitude by 0.33°
longitude grid cells across the United States. Each grid cell was assigned a wind power
class level from 1 to 7 (7 being the strongest) according to observed wind measurements
as well as various topographic and meteorological indicators including elevation changes,
eolian landforms, vegetation conditions, and coastal conditions. Several wind power
assessments have been made with this information and summarized into twelve regional
wind energy atlases that identify annual and seasonal average wind resources across the
United States (Elliott et al. 1986).
For the purposes of this study, the annual wind energy datasets for the Pacific
Northwest Region, at a 50-meter wind resource resolution, were used in the analyses. The
associated annual wind power class value is the primary attribute used to analyze wind
energy resources within the study area of Washington State. This data set is updated
regularly with current wind resource information and was last updated in June of 2012
(National Renewable Energy Laboratory 2014).
43
Solar Irradiance Dataset
The NREL solar irradiance data were derived using the State University of New
York (Albany) Satellite-To-Irradiance model developed by Dr. Richard Perez and his
collaborators (Perez et al. 2002; National Renewable Energy Laboratory 2013b). This
model uses geostationary weather satellites to monitor hourly solar irradiances, cloud
cover, daily snow cover, atmospheric water vapor, trace gases, and aerosols (Perez et al.
2002; National Renewable Energy Laboratory 2013c). All satellite-collected information
is then used in combination with terrain elevation, local ground albedo variations, and
sun-satellite angle adjustments to produce a grid of 0.1° longitude by 0.1° latitude annual
solar irradiance estimates across the United States. To gauge the model’s accuracy, the
solar irradiance output was tested against 10 ground weather stations in different climatic
environments across the United States. The results showed improved accuracy and
reduced bias from previously used satellite irradiance models, and it is believed that each
grid cell is accurate to approximately 15% of true, measured solar irradiance values
(Perez et al. 2002; National Renewable Energy Laboratory 2013c).
For the purposes of this study, the latitude equals tilt irradiance dataset estimating
solar resources appropriate for photovoltaic technologies at a 10-kilometer resolution was
used in the analyses. The annual solar resource estimate value is the primary attribute
used to measure the solar resources across the state of Washington. This dataset was
derived from satellite and meteorological information collected between 1998 and 2009,
with the most current dataset update made in September of 2012 (National Renewable
Energy Laboratory 2013a).
44
Existing Wind Turbines
The existing wind turbine GIS data used in this study is publically available from
the United States Geological Survey (USGS). The USGS provides impartial and reliable
scientific information about the environment and natural resources across the nation (US
Geological Survey 2013). One of their more recently published GIS datasets is an
inventory of onshore commercial wind turbine locations for the United States through
July 2013. This dataset is a synthesis of wind turbine location and technical specifications
from the Federal Aviation Administration Digital Obstacle File, the U.S. Energy
Information Administration, the Wind Energy Data and Information dataset from the Oak
Ridge National Laboratory, and various industry reports, environmental assessments, and
planning documents. All turbine data were verified with visual interpretation using highresolution aerial imagery in ArcGIS (Diffendorfer et al. 2014). For the purpose of this
study, USGS wind turbine data for the state of Washington were used in analyses.
Crucial Habitat Assessment
The crucial habitat assessment GIS data used in this study are publically available
from the Western Governors’ Association Crucial Habitat Assessment Tool (CHAT),
recently released in December, 2013. The CHAT application is a first-ever landscapelevel approach to assessing and prioritizing wildlife habitat and connection corridors
according to state conservation objectives (Western Governors’ Wildlife Council 2013a).
According to the Western Governors’ Wildlife Council, crucial habitat has been defined
as:
45
Places containing the resources, including food, water, cover, shelter and
“important wildlife corridors,” that are necessary for the survival and
reproduction of aquatic and terrestrial wildlife and to prevent unacceptable
declines, or facilitate future recovery of wildlife populations, or are important
ecological systems with high biological diversity value (Western Governors’
Wildlife Council 2013a, 6).
To classify crucial habitat across the landscape, a six-level relative ranking scheme was
employed according to state conservation objectives and existing wildlife data for the
western United States. A CHAT ranking model aggregated and prioritized state and
regional wildlife information according to three high-level themes—habitat for species of
concern, native unfragmented habitat, and species of economic and recreational
importance. (For a more extensive list of the wildlife information used to define crucial
habitat values see table 1.) The outcome of this model produced a one-square-mile
hexagonal grid spanning the western United States, with a single crucial habitat value for
each grid (Hamerlinck and Terner 2013; Western Governors’ Wildlife Council 2013a). It
is generally understood that the higher the crucial habitat ranking (rank 1 being the
highest), the higher the relative wildlife and overall conservation value for that area.
Additionally, the higher the ranking the more likely there will be wildlife resources that
may require mitigation or avoidance if development were to occur (Western Governors’
Wildlife Council 2013a).
For the purposes of this study, the crucial habitat ranking data for the state of
Washington were used in the analyses. While these data were not meant to replace
specific environmental and habitat assessments at the local scale, nor is it meant to be
applied as a regulatory tool at this time, the crucial habitat value is a good landscape-level
indicator of habitat importance across the state of Washington.
46
Table 1. Data Categories and Sources for Assigning Crucial Habitat Values
(Western Governors’ Wildlife Council 2013a, 8–10)
Data Category
Habitat for Species
of Concern
Native and
Unfragmented
Habitat
Riparian and
Wetland Habitat
Connectivity or
Linkage Assessment
Quality Habitat for
Species of
Importance
Terrestrial or
Aquatic Native
Species Richness
Data Definition
Species of greatest conservation need within State Wildlife Action Plans or
similar assessments:
Locations of federally- or state-listed threatened or endangered species
Key or priority habitat boundary delineations from State Wildlife Action
Plans or Comprehensive Wildlife Conservation Strategy
Plant and animal species with special protective-rankings
High priority areas for management of core conservation populations
Areas that are contiguous, possess a high degree of intact core areas or diversity
of natural habitat, or supply ecological functions to meet wildlife objectives.
These areas are unfragmented, or relatively unfragmented, by transportation
routes, human habitation, industrial infrastructure, or other human-caused
disturbances:
Natural Vegetation Classification habitats maps
Ecological systems of concern
Plant communities of concern (Heritage Rankings)
Priority habitat areas identified in updated State Wildlife Action Plans
(SWAPs)
Areas that represent unique environments and function to support animal and
plant diversity with respect to wildlife objectives and connectivity:
Spring/Seep/Cienega Locations
National Wetlands Inventory
National Hydrologic Database
Wetland components from State Comprehensive Outdoor Recreation Plans
Priority wetland areas and priority riparian habitats identified in updated
SWAPs
Areas described explicitly for aquatic or terrestrial wildlife habitat connectivity:
Major animal movement corridors or pathways
Landscape connectivity zones
This category provides for species consideration if not otherwise included as
“Habitat for Species of Concern”:
Sport Fish Quality Habitat: Areas recognized as important to meeting
biological requirements and objectives of fish species whose harvest is
regulated (i.e., blue ribbon streams)
Game Animal Quality Habitat: areas recognized as important to meeting
biological requirements and objectives of game species regulated by harvest,
such as winter concentration areas or important breeding areas (i.e., crucial
big game ranges, grouse lek locations or core grouse habitats if designated)
Areas where species composition represents a native, intact community and
where habitats are associated with a relatively high and distinctively described
species assemblage:
Aquatic species distribution maps
Ecoregional Assessments – Biodiversity Areas
Audubon Important Bird Areas
Gap-ReGap species composite maps
Christmas bird count and breeding bird survey data
47
Valued Lands
Important
Restoration Habitat
Lands that are protected or designated for their wildlife or aquatic values:
Protected Areas Database (PAD)
Priority areas identified from ecoregional analyses
Dedicated conservation land locations
Outdoor recreation priority/favored areas
Lands that are proximate to other important habitats and have the potential to
restore function or resiliency to target populations of fish and wildlife:
Spawning or rearing habitat for fishes that are isolated from current
populations
Habitat that was historically in one of the crucial habitat categories (2 or 3)
and could provide fish or wildlife benefits with restoration
Washington Wildlife Habitats
The Washington wildlife habitat GIS data (Johnson and O’Neil 2001) used in this
study are publicly available from the Northwest Habitat Institute (NHI). The NHI is a
non-profit scientific and educational organization in the Pacific Northwest that develops
data-rich and verifiable information to facilitate and promote state conservation efforts
(Northwest Habitat Institute 2011). The Washington wildlife habitat GIS dataset is a
synthesis of existing habitat information, field surveys, and Landsat TM imagery
interpretation (Kiilsgaard 1999). The existing habitat information sources contributing to
this project included the National Wetlands Inventory; the Washington Department of
Fish and Wildlife (WDFW) Blue Mountains habitat/vegetation mapping project; the
WDWF shrub-steppe vegetation mapping project; the Washington Department of Natural
Resources (WDNR) Heritage Program mapping project; US National Park Service; and
US Biological Service GAP Analysis Project.
Washington wildlife habitat was primarily classified according to vegetation
patterns across the Washington landscape. Wildlife habitat was first identified digitally
according to the existing wildlife habitat information, followed by field surveys in the
defined habitat areas for verification. Areas where vegetation information and Landsat
imagery were ambiguous were also followed up with field surveys for verification. The
48
final product identified 32 unique terrestrial or oceanic wildlife habitats across the state
of Washington (Kiilsgaard 1999). These data were last updated in 1999.
Protected Areas of the United States
The protected areas database of the United States (PADUS) GIS data (US
Geological Survey, Gap Analysis Program 2012) are also publically available from the
USGS science organization. Protected areas “are lands that have been dedicated to the
preservation of biological diversity and to other natural, recreation and cultural uses, and
managed for these purposes through legal or other effective means” (Protected Areas
Database-US Partnership 2009, 1). The PADUS GIS inventory of all US protected areas
is aggregated from several sources, including The Nature Conservancy (TNC), the US
Endowment for Forestry and Communities, as well as local, state and federal agency data
stewards (Protected Areas Database-US Partnership 2009; US Geological Survey, Gap
Analysis Program 2013). As of 2009, it is expected that approximately 90% of all
protected areas are present in the PADUS inventory. For the purposes of this study, the
PADUS GIS data are used to identify landscape exclusions in analyses for suitable
renewable energy development locations. Version 1.2 of this data was used; this was last
updated in May of 2011.
National Wetland Inventory
The National Wetland Inventory GIS data are publically available from the
United States Fish and Wildlife Service, a federal agency within the Department of the
Interior. The U.S. Fish and Wildlife Service works to guide conservation, development,
49
management, and education regarding fish, wildlife, plants, and their habitats for the
benefit of the American people (US Fish & Wildlife Service 2013). Wetlands are
important landscape features that contribute to the hydrologic and nutrient cycles across
the landscape and are included within the scope of the U.S. Fish and Wildlife Service’s
mission. The National Wetland Inventory is a collection of wetland and deep-water
landscape features across the entire United States and associated territories. These data
were produced using U.S. Geological Survey topographic maps, analysis of high-altitude
imagery, and a wetland classification system based on vegetation, visible hydrology, and
geography (Cowardin et al. 1979; US Fish and Wildlife Service, National Standards and
Support Team 2012; US Fish & Wildlife Service 2014). For the purposes of this study,
the National Wetland Inventory data are used to identify landscape exclusions in analyses
for suitable wind and solar energy development locations. This data was originally
published in 1979 and was last updated in 2012.
National Historic Places
The National Historic Places GIS data are publically available from the National
Park Service, a federal agency within the U.S. Department of Interior. The Register of
National Historic Places Program coordinates efforts to identify, evaluate, and protect
America’s historic, cultural, and archeological resources as authorized by the National
Preservation Act of 1966 (National Park Service 2011). The official list of National
Historic Places has been tracked and managed in a federal database called FOCUS since
1966, when the National Preservation Act was first enacted. More recently, this
information has been digitized according to the NPS cultural resource spatial data
50
transfers standards, to be useable within ArcGIS as spatial data (National Park Service
2012). For the purposes of this study, the National Register of Historic Places data is used
to identify landscape exclusions in analyses for suitable wind and solar energy
development locations. These data were last updated in 2012.
Washington Cities and Urban Growth Areas
The Washington cities and urban growth areas (UGA) GIS data are publicly
available from the Washington Department of Ecology, a state agency that collects and
records data concerning Washington’s air, water, and land (Washington Department of
Ecology 2014). The cities and UGA data include GIS polygons identifying city
boundaries and UGAs, as defined and managed by the Growth Management Act, across
the state of Washington (WA Department of Ecology 2011). By definition, cities and
UGAs are areas where urban growth and higher population densities are expected. They
make “intensive use of land for the location of buildings, structures, and impermeable
surfaces to such a degree as to be incompatible with the primary use of land for the
production of food, other agricultural products, or fiber, or the extraction of mineral
resources, rural uses, rural development, and natural resource lands” (Hunt et al. 2012,
13).
For the purposes of this study, the cities and UGA GIS data are used to identify
landscape exclusions in analyses for suitable renewable energy development locations.
Additionally, the cities and UGA GIS data are used to adjust the Washington Wildlife
Habitats “Urban and Mixed Environments” habitat type to reflect more current land use
patterns. These data were recently released in January of 2014.
51
Washington Agricultural Land Use
The Washington Agricultural Land Use GIS data are publically available from the
Washington State Department of Agriculture (WSDA), a state agency that supports the
producers, distributors, and consumers of Washington’s food and agriculture products
(Washington State Department of Agriculture 2014). This GIS dataset identifies
agricultural land use across the state of Washington with detail down to the specific type
of crops grown and types of irrigation used for each agricultural operation. The United
States Department of Agriculture (USDA) and National Agricultural Statistics Service
(NASS) manages the collection and annual publication of this dataset. Satellite imagery
and other geo-referenced inputs from the Landsat 8 OLI/TIRS, Disaster Monitoring
Constellation DEIMOS-1 and UK2 sensors, USGS National Elevation Dataset, and
USGS land cover data are used to identify agricultural land use over an annual growing
season down to a 30-meter ground resolution (US Department of Agriculture et al. 2014).
The dataset is then verified by GIS technicians trained in crop identification by the
USDA Farm Service Agency Common Land Unit Program. For the purposes of this
study, the Washington Agricultural land use 2013 dataset is used to adjust the
Washington Wildlife Habitats “Agriculture, Pasture, and Mixed-Environments” habitat
type to reflect more current land use patterns.
Data Limitations
There are three main limitations to the data used in this study. The first is the age
of the Washington Wildlife Habitats dataset. This dataset was published in 1999 and has
52
not been updated since its original publication. This is not surprising, since the original
data collection was field-survey verified and updating it is likely to be a costly and timeconsuming process (Kiilsgaard 1999). However, the remainder of the data sources used in
this study have been published or updated within the past two years (June 2012—January
2014), ensuring the utilization of the most current data available.
The Washington Wildlife Habitats dataset is an important component in
addressing the fourth and fifth research questions in this study. However, there is likely to
be some error in the results because of the age of the data. It is almost certain that
landscape changes have occurred within the past 13 years that would alter the boundaries
of the habitat types identified in the 1999 dataset. This would commonly be manifested in
increasingly anthropogenic landscapes, though not in all cases. To account for this, the
Washington Wildlife Habitat data should be adjusted to reflect the anthropogenic changes
that have occurred, where possible. After adjustment, this dataset can still offer valuable
information regarding the associations between crucial habitat levels and wind and solar
energy potential within individual Washington habitat types.
A second limitation to the data used in this study is that many of the data sources
were prepared for regional spatial investigations rather than at the local scale. These data
sources maintain clear use statements indicating that they are not intended for regulatory
purposes and that they were not prepared at the level of precision to replace small-scale,
local analyses (Western Governors’ Wildlife Council 2013b; US Fish & Wildlife Service
2014; Kiilsgaard 1999). As identified earlier, the scope of this study is to understand
variable interactions at a landscape level for the entire state of Washington. While the
scope of this study does not violate this data limitation, it should be acknowledged and
53
echoed that any outcomes from this study should be taken at the landscape level and that
local investigations for specific sites should use more precise methods in addressing these
research questions. This data source limitation applies to the following data sources:
Crucial Habitat Assessment Tool (CHAT)
Washington Wildlife Habitats
National Wetland Inventory
The final limitation to the data used in this study specifically involves the
National Register of Historic Places dataset. This dataset is a collection of spatial
polygons and point features identifying the historic places across the United States. While
it is easy to incorporate the polygon features into the analyses of this study, the point
features are more difficult to address. The point features can represent a variety of
historic places, including buildings, archeological artifacts, or other cultural places like
historic viewpoints (National Park Service 2012). However, there is no sense of size
associated with a point feature, and no guidelines were identified by the National Park
Service as to how to accommodate this dilemma.
For fear of grossly misrepresenting the spatial footprints of the historic places
represented by these point features, these data were excluded from the analyses in this
study. This amounts to about half of the National Historic Places identified in the dataset.
This means that, in the spatial analysis of suitable land for wind and solar energy
development, some areas identified may not truly be suitable due to the presence of
historic places. However, since many of the point features are historic buildings, it is
probable that they will be located in urban areas. These features are likely to be
accounted for in the exclusion of the city and urban growth areas dataset as an additional
part of this analysis.
54
Data Preparation
Data preparation was required in most cases prior to use in analysis and can be
grouped into three main data preparation tasks. First, many of the datasets contained
information that exceeded the study area of Washington State and required redefinition
according to the study area boundaries. For these datasets the intersect, clip, or select
attribute by location GIS tools were utilized to create new GIS shapefiles with spatial
data for Washington State only. This ensured all analyses were conducted within the
study area and not influenced by data in neighboring states.
The second data preparation task addressed the Washington Wildlife Habitats data
age limitation and out-of-scope habitat types. To address the data age limitation the
habitat types of “Urban and Mixed-Environments” and “Agriculture, Pasture, and MixedEnvironments” were redefined to reflect more current spatial boundaries. To redefine
these habitat types, a two-step process was employed using the 2014 Washington
Department of Ecology Cities and Urban Growth Areas dataset, as well as the 2013
USDA Agricultural Land Use dataset. First, the merge and dissolve GIS tools were used
to expand the urban and agriculture habitat type polygons to include the areas identified
in the more current datasets. In the case of a conflict between the urban and agriculture
habitat types, the conflicting areas were included in the “Urban and Mixed
Environments” habitat since, by definition, this habitat type is often bordered by
agriculture, and landscape modifications occur frequently (Chappell et al. 2001). Second,
the erase GIS tool was used to reduce the boundaries in the adjoining habitat type
polygons where the urban and agriculture habitats had expanded, as indicated in the more
55
recent data sets. All but three of the 22 habitat types present in Washington State were
affected by the growth of agriculture and UGAs across the state, so this data adjustment
was important to improve the accuracy of analyses.
In addition to the data limitation adjustment, the number of habitat types
identified in the Washington Wildlife Habitats was also adjusted to reflect the proper
scope of analysis. The Washington Wildlife Habitat dataset includes both terrestrial and
oceanic habitats. Since this study focuses exclusively on terrestrial landscapes and onshore wind and solar energy development, this dataset was redefined to only include
terrestrial habitat types. This was accomplished by using an attribute definition query to
exclude all oceanic and related near-shore habitat types. In addition, the sheer number of
habitat types identified was too granular and dispersed across the landscape to use in the
analysis. To adjust for this data challenge, the habitat types were grouped by similar
habitat characteristics into five general habitats to be used in the analysis (Table 2).
56
Table 2. Consolidation of Individual Habitat Types into General Habitats and Habitat
Exclusions
General Habitats
Agriculture, Pasture, and
Mixed-Environments
Forest and Woodlands
Grasslands and Shrublands
Urban and MixedEnvironments
Wetlands, Rivers, Lakes, and
Reservoirs
Habitats Excluded from
Analysis
Specific Habitat Types
Agriculture, pasture, and mixed-environments
Westside lowland conifer-hardwood forest
Westside oak and dry Douglas-fir forest and woodlands
Montane mixed conifer forest
Eastside (interior) mixed conifer forest
Lodgepole pine forest and woodlands
Ponderosa pine and eastside white oak forest and woodlands
Upland aspen forest
Subalpine parklands
Alpine grasslands and shrublands
Westside grasslands
Eastside (interior) canyon shrublands
Eastside (interior) grasslands
Shrub-steppe
Urban and mixed-environments
Lakes, rivers, ponds, and reservoirs
Herbaceous wetlands
Westside riparian wetlands
Montane coniferous wetlands
Eastside (interior) riparian wetlands
Costal dunes and beaches
Coastal headlands and islets
Bays and estuaries
Inland marine deeper waters
Marine nearshore
Marine shelf
Oceanic
Finally, the third data preparation task involved the USGS existing wind turbine
dataset to generate wind farm polygons for use in analysis. The USGS existing wind
turbine dataset is a collection of point features indicating turbine geographic location,
technical specifications, and wind farm site identification. However, to address the first
research question a polygon representing individual wind farm areas was required for the
analysis. To accomplish this, a buffer was first established around each wind turbine with
a radius of four turbine rotor diameters, as suggested in wind farm guidelines for turbine
spacing within rows (Manwell, McGowan, and Rogers 2009; Nelson 2009). These
57
buffers ensure that the wind farm polygon boundaries will be drawn to include
appropriate turbine spacing requirements for the outermost turbines in each wind farm.
Next, the minimum bounding geometry GIS tool with a convex hull geometry type
setting (considered the natural bounding area for a set of points) and grouping level
according to the wind farm site location was applied to the turbine buffers. The outcome
of this GIS tool generated a convex polygon for each wind farm. While the wind farm
boundaries do not exactly replicate landscape ownership boundaries, each wind farm
polygon encompasses all wind turbines associated with the wind farm and is a good
geometric estimate of the landscape that could be impacted by each wind farm.
3.2 Data Analysis
Existing Washington Wind Farms
To understand the levels of crucial habitat that occur within existing Washington
wind farms, a basic spatial analysis was performed. First, wind farm sites that were listed
in the original dataset but had unknown site affiliations and specification or were not
fully operational were excluded from this analysis. This ensured that all existing
Washington wind farms in scope for analysis were known and fully operational. To
complete the spatial analysis, the intersect GIS tool was used to select only the crucial
habitat data that occurred within the known, operating wind farm polygon boundaries.
The outcome identified wind farm sites, crucial habitat rankings, and the geometric area
of the crucial habitat landscape coverage within the existing wind farms. These data were
then exported to Excel, where pivot table graphs were generated, presenting simple
58
descriptive statistics of the crucial habitat levels within individual or groups of existing
Washington wind farms across the landscape.
Wind and Solar Energy Development Landscapes
For the assessment of wind and solar energy development landscapes across the
state of Washington, a second spatial analysis was performed. This was then followed by
a simple estimation of potential renewable energy that could be produced, as indicated by
the number of acres suitable for development according to the inclusion or exclusion of
the various crucial habitat levels. This portion of the analysis describes the risk of
landscape conflict and opportunity or challenge of optimizing the land use between
crucial habitat and future wind and solar energy development. To begin the spatial
analysis, landscapes “suitable” for commercial wind or solar energy development were
first defined. Smaller, domestic energy production was not considered as part of this
analysis because land use would not be as impacted as the lands required for commercial
developments. In addition, the technologies available for domestic energy production
have much more variable technical specifications and siting requirements than those
available for commercial energy production. This would require more detailed
calculations of potential energy production on suitable energy development lands that
what was used in this analysis.
The American Wind Energy Association (AWEA) wind energy siting handbook
was used as the primary reference in determining landscapes that were not suitable for
commercial renewable energy development (American Wind Energy Association, Tetra
Tech EC Inc., and Nixon Peabody LLP 2008). Since there are few published guidelines
59
for the development of solar energy, a similar landscape suitability model used to identify
wind energy development landscapes was applied to the analysis for solar energy
development landscapes. According to the AWEA, landscapes that are not suitable for
commercial renewable energy development are those that are protected areas, wetlands,
or historic or cultural resources, as identified by the National Register of Historic Places.
These lands are either protected through legislation or, in the case of wetlands, require
such high mitigation costs that developments in these areas usually prove unprofitable
(American Wind Energy Association, Tetra Tech EC Inc., and Nixon Peabody LLP
2008). For the purposes of this study, these lands were considered unsuitable for
commercial renewable energy development and excluded from the analysis.
In addition, cities and urban growth areas were also excluded, since these areas
are designated for alternative uses. Regarding suitable wind farm development locations
only, existing wind farms were also excluded from the analysis since they are already
developed and unavailable for further energy production. However, these lands were
included in the assessment of suitable solar energy development locations, since wind
turbines have large spacing requirements and there is a potential to develop solar
technologies in between the wind turbines, if conditions are appropriate. The AWEA also
identifies lands with known endangered species habitation as unsuitable for energy
development. However, the AWEA also identifies mitigation and accommodation
techniques as well as exceptions to work within and around such areas. Since this siting
concern requires local investigation to determine the extent of unsuitability, lands with
known endangered species were not excluded from the analysis. Instead, they will be
60
reflected in the crucial habitat rating across the landscape as a high-level habitat
conservation priority landscape.
To conduct the spatial analysis, the Erase GIS tool was used to remove lands that
were unsuitable for wind and solar energy development (i.e., protected areas, wetlands,
National Historic Places, cities and urban growth areas, and existing Washington wind
farms—for wind energy assessment only). Next, the Select by Attribute GIS tool was
used to remove lands that did not exhibit an annual wind or solar energy level that was
suitable for commercial energy production. For wind, lands with a wind power class of
one or two are considered too low for commercial wind energy production and were
removed from the analysis (American Wind Energy Association, Tetra Tech EC Inc., and
Nixon Peabody LLP 2008; National Renewable Energy Laboratory 2014).
Unlike wind energy, solar energy does not have a strict solar irradiance level
defining suitable solar energy resources. In this case, the solar prospectus online GIS
solar energy data mapping tool, developed by the NREL, was utilized to identify the
lowest solar irradiance level of existing U.S. commercial Photovoltaic (PV) solar
facilities. This solar irradiance level became the cutoff point between suitable and
unsuitable solar energy levels for the purposes of this study. The South Burlington Solar
Farm in Vermont was identified as the PV solar facility operating with the lowest average
annual PV solar resource (latitude equals tilt irradiance) level of about 4.26 kWh/m2/day
(National Renewable Energy Laboratory 2009). Using this figure, any annual average
solar tilt equals lateral irradiance levels below 4.25 kWh/m2/day was considered
unsuitable for solar energy production and was removed from the analysis.
61
Once the landscapes suitable for wind and solar energy development were
defined, the Intersect GIS tool was used to identify the crucial habitat levels within these
landscapes. These data were then exported to Excel, where pivot table graphs were
generated presenting simple descriptive statistics of the crucial habitat levels across the
landscapes suitable for wind or solar energy development. Finally, simple estimates were
made of how much average annual energy could be produced if the landscape was
developed within the various crucial habitat levels.
For wind energy, 2011 Washington wind energy production data were used to
calculate the estimated number of acres required for commercial wind energy
technologies to produce one gigawatt hour of energy per year (acre/GWh/year) (US
Department of Energy 2011; Northwest Power and Conservation Council 2013). This
was then applied to the total landscape area (acres) for each crucial habitat level within
suitable wind energy development locations. This estimate is expected to capture the
wind power variation across the landscape because the wind power class level
distributions in existing Washington wind farms is similar to the wind power class level
distributions for the remaining suitable wind energy development landscapes. In both
cases, around 70-75% of the wind power across the landscapes is categorized as class 3
and below, between 20-25% as class 4, and the remaining 5% as class 5 and above.
Additionally, the crucial habitat distributions within each wind power class level of
suitable wind development lands are consistent with having a majority of the lands
ranked as crucial habitat levels 3 and 2. By using existing Washington wind farm
production data as the basis for this calculation, generating estimates for future wind
energy production that are realistic to the conditions in the state of Washington was
62
attempted. The outcomes of these estimates were then used to assess the impact of crucial
habitat on potential future wind energy production in Washington State.
For solar energy, existing commercial solar energy production for the state of
Washington is so small it is not explicitly tracked, so a different approach was applied.
Ong et al., researchers for the NREL, recently published a study that investigated solar
energy land use needs in the United States (Ong et al. 2013). This study collected site
specifications for 150 PV solar facilities in the U.S. (operating or under construction) and
calculated an energy generation weighted average total area requirement for different
specific solar PV technologies (acres/GWh/year). The average of these figures represents
the average number of acres required for solar PV technologies to produce one gigawatt
hour of energy in one year. This was then applied to the total landscape area (acres) for
each crucial habitat level of suitable solar energy development locations. The outcome
was then used to assess the impact of crucial habitat on potential future solar energy
production in Washington State.
Washington Habitats
To assess the landscape-level interactions between the most crucial habitats and
high wind or solar energy potential within Washington habitats, an analysis of spatial
autocorrelation using the Anselin Local Moran’s I spatial statistic was performed. Then,
the results were assessed according to the interaction between the crucial habitats and
wind or solar results within the five general Washington habitat types. The Anselin Local
Moran’s I statistic identifies significant spatial clustering of both high and low values that
occur across the landscape for a single numeric feature attribute. This statistic uses matrix
63
algebra and a spatial weighting mechanism to assess the similarity or difference between
an individual feature and its spatial neighbors (Anselin 1992; Fotheringham, Brunsdon,
and Charlton 2000; Blyth et al. 2007; Esri Inc. 2013). The equation for the Anselin Local
Moran’s I statistic is:
̅
I* =
where
̅
∑
is the feature attribute being assessed, ̅ is the mean of the corresponding and
neighbor feature attributes,
is the spatial weight between the feature attribute
the neighbor feature attribute
, and
=
and
is:
∑
̅
̅
The results from this analysis calculate a z-score and related p-value for each feature in
the study area. Features that have a significant positive z-score indicate significant spatial
clustering of features with high values and a significant negative z-score indicates
significant spatial clustering of features with low values (Fotheringham, Brunsdon, and
Charlton 2000; Esri Inc. 2013).
To conduct a local spatial autocorrelation analysis, certain assumptions must be
met for the analysis to be valid. First, results are only reliable if more than 30 features are
being assessed and each feature has at least one neighbor. Second, no feature should have
all other features in the analysis as neighbors; ideally, each feature should have about 8
other features as neighbors in the analysis. Finally, the attribute of interest must be a
numeric value and have some variation between feature values (i.e., more values than just
0 and 1) (Esri Inc. 2013). The local spatial autocorrelation analysis performed in this
study met all of these assumptions.
64
To address this research question, three local spatial autocorrelation analyses were
conducted across the entire state of Washington according to crucial habitat ranking,
wind power class ranking, and average annual photovoltaic solar irradiation. The most
important input in conducting these local spatial autocorrelation analyses is the
conceptualization of the spatial relationships among features. There are many ways to
define this conceptualization according to the feature types used in the analysis and how
the features interact with one another across the landscape. Subtle differences in analysis
configuration can produce drastic differences in the statistical outcomes (Esri Inc. 2013;
Fotheringham, Brunsdon, and Charlton 2000). This means results should only be
interpreted within the scope of the specific analysis. For this study, a spatial weights
matrix was generated for each local autocorrelation analysis to define the
conceptualization of spatial relationships for the features in each of the three datasets.
The Generate Spatial Weights Matrix tool in ArcGIS was used to accomplish this.
Since the features of interest in this analysis are all polygon features that are, in
most cases, contiguously connected across the landscape, a K nearest neighbor
conceptualization of spatial relationships using the 8 nearest neighbors was defined. This
ensured the assumptions of the analysis were met and the distance threshold of the
features included for each feature analysis was kept at a minimum. This is particularly
important because crucial habitats, wind energy resources, and solar energy resources in
one area of the landscape are more likely to be influenced by the landscape conditions of
areas that are direct neighbors than by those that are farther away. Row standardization
was also defined to standardize the spatial weights mechanism used for the neighbors
included in the individual feature analysis. In this case, the area of each neighboring
65
polygon was used to adjust the weighted influence of each neighboring feature. This
ensured that the influence of each neighbor was a factor of both distance from the
centroid of the individual feature being analyzed and size of neighboring polygons.
Once the local spatial autocorrelation analysis was completed using the Cluster
and Outlier Analysis (Anselin Local Moran’s I) spatial statistic tool, the results were
assessed between crucial habitats and wind energy resources as well as between crucial
habitats and solar energy resources. Since the spatial autocorrelation analyses were
conducted using the same study region with the same scope and the same analysis
configurations, the results of the individual autocorrelation analyses are comparable with
one another. These comparisons identified important patterns of spatial clustering
between crucial habitat and wind or solar energy resources. These include:
Areas of significant high wind or solar energy resource clustering
Areas of significant low wind or solar energy resource clustering (not important
for this analysis)
Areas of significant most-crucial habitat clustering
Areas of significant least-crucial habitat clustering
These patterns were then be used to identify:
Significant areas of high conflict
o Areas that are both significant high wind or solar energy resources
clustering and significant most-crucial habitat clustering
Significant areas of low conflict
o Areas that are both significant high wind or solar energy resources
clustering and significant least-crucial habitat clustering
With these spatial patterns identified, a final assessment was conducted to understand the
distribution of these significant spatial patterns within the five general habitats across
Washington State.
66
Chapter 4: Results
4.1 Existing Washington Wind Farms
The crucial habitat assessment of existing wind farms in the state of Washington
shows that existing wind farms have been sited moderately well according to landscapelevel habitat conservation priorities. With a total area of 170,104 acres, just over half
(56.6%) of the lands developed for wind energy have a crucial habitat ranking of 3–6
signifying development on lesser crucial habitat lands. However, 36.7% of the lands
developed for wind energy have a more crucial habitat ranking of 2, and 6.8% with the
most crucial habitat rank of 1 (see Figure 2 and 3).
67
68
Each zone is explored in further detail in figures 4a and 4b.
Figure 2. Crucial habitat on all existing and operating Washington wind farms.
As noted in the crucial habitat assessment data sources section of chapter 3 (page
45), the crucial habitat ranking is an aggregation of statewide habitat conservation
priorities, where the higher the crucial habitat ranking, the higher the relative wildlife and
conservation value of the land (Western Governors’ Wildlife Council 2013a). The most
crucial habitat lands—with a rank of 1—will have documented threatened or endangered
aquatic species spawning areas, documented threatened or endangered terrestrial species,
level-1 priority habitat ecological systems of concern and confirmed heritage vegetation,
or high-integrity estuaries present on the landscape. Crucial habitat lands with a rank of 2
will have documented or presumed endangered and threatened aquatic species, confirmed
federal and state candidate and sensitive terrestrial species, level-2 priority habitat
ecological systems of concern, moderate integrity estuaries, or are spawning areas for
aquatic species of economic and recreational importance. Levels 3–5 will have lesser
degrees of the ranking factors considered for ranks 1 and 2, as well as consideration and
prioritization of various levels of freshwater integrity, large natural areas, terrestrial
species of economic and recreational importance, landscape connectivity, and wildlife
corridor factors (see Table 9 in the Appendix for a summary of the crucial habitat ranking
factors associated with the six crucial habitat levels). Higher crucial habitat levels will
often require habitat mitigation or habitat avoidance practices when development is
considered, according to federal and state wildlife protection policies (Washington State
1971; Western Governors’ Association 2013; Western Governors’ Wildlife Council
2013a). Alternatively, less crucial habitat levels signify lands with wildlife and habitat
areas that are not considered to have as high a conservation value as those with the more
crucial habitat levels 1 and 2.
69
The resulting distribution of crucial habitat levels for existing wind farms is
somewhat similar to the distribution of crucial habitat levels statewide; a majority of the
landscape (92%) has a crucial habitat ranking of 1, 2, or 3 in both cases (Figure 3).
However, a clear difference between the two distributions is observed when comparing
the portions of land that have been categorized as crucial habitat rank 1. Existing wind
farms have been developed on 66% less of the most crucial habitat lands (rank 1) than
what are observed in the crucial habitat distribution statewide (6.8% vs. 15.3%
respectively, Figure 3). This is consistent with the general hypothesis that the EIA
process would promote wind energy development in areas having lower priority habitat
conservation concerns.
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
6.8%
Crucial
Habitat
Rank
15.3%
36.7%
1
39.2%
2
3
48.4%
4
37.3%
1.5%
6.2%
5
6.7%
2.0%
6
Existing Wind Farms
Statewide
Figure 3. Crucial habitat on existing Washington wind farms and statewide
Despite the overall reduction in existing wind farm development on the most
crucial habitat lands, the assessment of crucial habitat at an individual or zoned wind
farm level is much more varied. For ease of visual comparison, the 20 Washington wind
farms were placed into eight wind farm zones (Zones A – H) across the landscape to
70
assess the crucial habitat distribution in different geographic locations (Figure 4a and 4b).
Many of the zones show crucial habitat distributions that are consistent with the overall
crucial habitat distribution of all 20 existing wind farms. However, some wind farm zones
show notable differences. For example, zone G contains the Palouse wind farm, spanning
11,330 acres of land, which has been sited entirely on less crucial habitat lands with a
rank of 5. This is the best-sited wind farm in the entire state with respect to landscape
conservation priorities, and it is the only wind farm located in an area where crucial
habitat is at a level 5. In contrast, zone F, including the Wild Horse and Vantage wind
farms, spanning 14,988 acres, were sited in areas containing the most crucial habitat, with
39.2% of the lands having a crucial habitat rank of 1 and 52.5% with a rank of 2. These
wind farms are the most poorly sited wind farms out of all the existing wind farms
concerning landscape level habitat conservation priorities.
71
Figure 4. Crucial Habitat Assessments of Wind Farm Zones
Figure 4a shows the
crucial habitat
assessment for the
existing and operating
wind farms in zones A–
C. Most of the wind
farms in these zones
have been developed in
landscapes with crucial
habitat rankings of 1–3.
72
Figure 4. Crucial Habitat Assessments of Wind Farm Zones
Figure 4b shows the
crucial habitat
assessment for the
existing and operating
wind farms in zones D–
G. The Palouse wind
farm in zone G was
developed on lands
with a crucial habitat
rank of 5 for the entire
wind farm operation,
making it the best-sited
wind farm with respect
to landscape level
habitat conservation
priorities. The Vantage
and Wild Horse wind
farms in zone F is the
worst sited wind farm,
with most development
occurring on the most
crucial habitat lands
with a rank of 1.
NOTE: Crucial habitat distributions within individual wind farms appear in appendix,
Figure 19.
73
4.2 Wind and Solar Energy Development Landscapes
Wind Energy Assessment
The crucial habitat assessment of suitable wind energy development landscapes
shows that a majority of these lands is of moderate concern for landscape-level habitat
conservation. Of the lands suitable for wind energy development, 8% have a crucial
habitat ranking of 1, 30% have a crucial habitat ranking of 2, and 50% have a crucial
habitat ranking of 3 (Figure 5). Combined, this occupies just over 1 million acres of the
total 1.2 million acres of land suitable for wind energy development as defined by this
study. Also, it is interesting to note that there are no least-crucial habitat lands with a rank
of 6 in the lands suitable for wind energy development.
100%
90%
8%
15%
80%
70%
60%
Crucial
Habitat
Rank
30%
1
39%
2
50%
3
40%
30%
4
50%
37%
5
20%
10%
0%
6
6%
2%
9%
2%
Suitable Wind Energy
Lands
Figure 5. Crucial habitat on suitable wind energy lands and statewide
Statewide
This distribution of crucial habitat was mostly expected as identified by the
general hypothesis, since it follows similar patterns to the crucial habitat distribution
74
across the entire state of Washington. Both have roughly 80% of the landscape with a
crucial habitat ranking of 2 or 3 and less than 1% with a crucial habitat rank of 6.
However, suitable wind energy development landscapes have slightly lower portions of
most-crucial habitat lands with a rank of 1 compared to those across the entire state of
Washington (8% vs. 15%, as shown in Figure 5). This is 46% less most-crucial habitat
lands proportionally, and shows that wind energy development poses a slightly lower risk
of conflict with the most crucial habitat conservation areas than might be supposed.
Closer investigation of the distribution of crucial habitat across the landscape
shows that a few locations in southeastern and central Washington State have large
contiguous areas suitable for wind energy development (Figure 6). Within these areas
there are about four locations along the southern border of Washington that have
groupings of lands with lower crucial habitat rankings. These are identified by the shades
of green in Figure 6. If landscape-level habitat conservation efforts were a high concern
in relation to wind energy development, these areas of land with lower crucial habitat
rankings should be investigated first to determine the feasibility of wind energy
development at a local scale.
75
76
Figure 6. Crucial habitat on suitable wind energy development landscapes
To investigate the opportunity or challenge of optimizing the land use between
habitat conservation and wind energy development, an assessment of the levels of crucial
habitat that future wind energy development could be restricted in order to both protect
habitat quality and contribute to future energy generation was conducted. To achieve this,
an estimate of the average energy production (GWh) per year according to the total
landscape area (acres) of suitable wind energy lands was calculated according to various
crucial habitat levels. As discovered when investigating the first research question in this
study, existing Washington wind farms span approximately 170,104 acres of land. These
wind farms generated 5,830 GWh of energy in 2011, which made up 1.8% of all energy
produced in the state of Washington (US Energy Information Administration 2012;
Northwest Power and Conservation Council 2013). With this information it is estimated
that, for every 1 GWh of wind energy produced over the course of one year,
approximately 29.2 acres of land would be required. When applied to the total area of the
landscapes ranked at each crucial habitat level, we may examine the impact on potential
future wind energy production that would result from the exclusion of areas with specific
crucial habitat levels.
According to the analysis, lands with the most crucial habitat ranking of 1 or 2
could be excluded from future wind energy development and still allow the generation of
an estimated 25,640 GWh annually from wind energy production on less crucial habitat
lands (Table 3). This represents a 440% increase in existing wind energy generation in
Washington State. This means that there is enough land suitable for wind energy
development that also has a lesser crucial habitat ranking of 3 and above to quadruple the
existing wind energy production across the state if all of these lands were developed.
77
Obviously, it is very unlikely that the total area would be developed; however, it is
important to know there is a healthy growth potential within these landscapes.
If wind energy development were to be restricted further, to crucial habitat lands
of ranks 4–6 only, however, the potential for wind energy growth would be severely
limited. Under this criterion for suitable wind energy development, the potential growth
of wind energy would only be an about 4,636 GWh annually. This estimate is only an
80% increase in existing Washington wind energy production.
As can be seen in Figure 7, the majority of lands suitable for wind energy
development have a crucial habitat rank of 3. Without these lands available for
development consideration other siting issues, such as finding large, contiguous
landscapes for development and working through social siting challenges, could further
inhibit the already limited wind energy growth potential. From this assessment, it seems
reasonable that if future policy or best practices in wind energy siting and development
were to include landscape-level conservation priorities, absolute exclusions could only
include crucial habitat levels 1 and 2. This does not mean the remaining lands rated as
crucial habitat 3–6 are always fit for development, or that local environmental
assessments identifying site-specific environmental concerns can be disregarded in these
lands. What is clear from these findings, however, is that the most crucial habitat lands of
ranks 1 and 2 could be preserved and protected, supporting high-level state conservation
priorities, while still leaving ample room for future wind energy growth across the state
of Washington.
78
Estimated Annual Energy
Production (GWh)
45,000
40,000
Crucial
Habitat
Rank
3,536
35,000
30,000
12,742
25,000
20,000
15,000
21,004
10,000
5,000
0
3,832
803
1
2
3
4
5
6
Table 3. Wind energy
potential by crucial
habitat levels
Crucial Habitat
Levels
0
1–6
2–6
3–6
4–6
5–6
6
Est. Annual
Average Energy
Production (GWh)
41,918
38,382
25,640
4,636
803
0
Figure 7. Estimated annual average
energy production (GWh) of
suitable wind energy
development landscapes by
crucial habitat rank
Solar Energy Assessment
The crucial habitat assessment of suitable solar energy development landscapes
shows a majority of these landscapes are of moderate to high concern for landscape level
habitat conservation. Of the lands suitable for solar energy development, 50% have a
crucial habitat ranking of 1 or 2 and another 37% have a crucial habitat ranking of 3
(Figure 8). Combined, this occupies approximately 11.8 million acres of the total 14
million acres of land suitable for solar energy development as defined by this study. As
observed in the wind energy analysis, there are no least-crucial habitat lands with a rank
of 6 in the lands suitable for solar energy development.
79
100%
90%
10%
15%
80%
70%
60%
40%
39%
50%
40%
30%
37%
37%
20%
10%
0%
6%
Statewide
2%
12%
6%
Suitable Solar Energy
Lands
Crucial
Habitat
Rank
1
2
3
4
5
6
Figure 8. Crucial habitat on suitable solar energy development lands and
statewide
This distribution of crucial habitat was mostly expected, as identified by the
general hypothesis, since it follows a pattern similar to the crucial habitat distribution
statewide; having roughly 90% of the landscape with a crucial habitat ranking of 1, 2, or
3 and less than 1% with a crucial habitat rank of 6. However, there are two notable
differences between the crucial habitat distributions of these two landscapes. Like the
suitable wind energy development landscapes, the suitable solar energy development
landscapes have slightly lower most-crucial habitat lands with a rank of 1, as compared to
those statewide (10% vs. 15%, as shown in Figure 8). This is 33% less most-crucial
habitat land area within the suitable solar energy development landscapes proportionally.
In addition, suitable solar development lands have 125% more crucial habitat lands of
lower conservation value (ranks 4–5) than those observed across the entire state of
Washington (18% vs. 8%, Figure 8). While the general distribution of crucial habitat is
similar within the two landscapes, suitable solar energy development landscapes do have
80
a slightly lower impact on the most-crucial habitat conservation priorities. Moreover,
solar energy development also presents a greater opportunity to develop in areas of lower
conservation value.
When looking at the distribution of crucial habitat across the landscape, nearly the
entire eastern portion of the state of Washington is suitable for solar energy development
(Figure 9). Within this area there are several locations that have large segments of the
landscape with less-crucial habitat rankings. These are identified by the shades of green
in Figure 9 and are generally located in the east-central and southeastern parts of the
State. If landscape-level habitat conservation efforts were taken to be a high concern in
relation to solar energy development, these areas of land with less-crucial habitat
rankings should be investigated first to determine the feasibility of solar energy
development at a local scale.
81
82
Figure 9. Crucial habitat on suitable solar energy development landscapes
To investigate the opportunity or challenge of optimizing the land use between
habitat conservation and solar energy development, an assessment of the levels of crucial
habitat that future solar energy development could be restricted in order to both protect
habitat quality and contribute to future energy generation was conducted. To achieve this,
an estimate of the average energy production (GWh) per year according to the total
landscape area (acres) of suitable solar energy lands was calculated according to various
crucial habitat levels. Since commercial solar facilities have a minimal presence in
Washington State, the average number of acres required for solar PV technologies to
produce one gigawatt hour (GWh) of energy in one year was taken from the Ong et al.
study (2013), which showed that, for every 1 GWh of solar energy produced over the
course of one year, approximately 3.8 acres of land is required. When applied to the total
area of the landscapes ranked at each crucial habitat level in Washington State, we may
examine the impact on potential future solar energy production that would result from the
exclusion of areas with specific crucial habitat levels.
According to the analysis, future solar development could be restricted to crucial
habitat lands with a rank of 5 only, and still allow for the generation of an estimated
209,540 GWh annually (Table 4). This is 65% of the total 2011 energy production
(GWh) for the entire state of Washington. This means that there is enough land suitable
for solar energy development on the less-crucial habitat lands (with a rank of 5) to
increase Washington’s annual energy production over 50%. This is a huge amount of
energy potential but not entirely unexpected, since the total land use requirements per
GWh of energy production in one year is only 3.8 acres. Again, it is extremely unlikely
that the entire crucial habitat area of level 5 will be developed for solar energy
83
generation. There are many economic, social, and technical aspects that further determine
the suitability and feasibility for development at particular locations that must also be
considered. These will be discussed in detail within the “Solar Energy Resource Analysis
Limitations” section of the next chapter, beginning on page 110. However, this initial
analysis does show that there is much solar energy generating potential in the leastcrucial habitat areas wherever economic and other practical challenges permit.
As can be seen in Figure 10, there is a large potential for solar energy generation
across the various crucial habitat levels. However, from this assessment it seems
reasonable that if future policy or best practices in solar energy siting and development
were to include landscape-level conservation priorities, future solar growth could initially
be targeted in areas with a crucial habitat ranking of 5 alone. Not only are these
landscapes sufficiently sized for ample solar energy growth, but they happen to also be
located in the same general area of the southeastern corner of Washington State (Figure 9
above). This would be conducive to establishing larger solar facilities, further expanding
renewable energy production in Washington, while also working toward protecting highlevel state conservation priorities.
84
Estimated Annual Energy
Production (1,000 GWh)
4,000
3,500
368
3,000
2,500
1,419
Crucial
Habitat
Rank
2,000
1,500
1,324
1,000
500
424
210
0
0
Figure 10. Estimated annual
average energy generation
(GWh) of suitable solar
energy development lands
by crucial habitat ranking
1
2
3
4
5
6
Table 4. Solar energy
potential by crucial
habitat level
Crucial Habitat
Levels
1–6
2–6
3–6
4–6
5–6
6
Est. Annual
Average Energy
Production (GWh)
3,745,146
3,377,146
1,957,879
633,498
209,540
0
4.3 Washington Habitats
In the assessment of the interaction between crucial habitat and wind or solar
energy resources, it is important to understand if there are positive or negative
interactions between the variables within the different habitat types of Washington State.
Gaining an understanding of these interactions will help habitat conservation efforts be
more aware of the risks and opportunities of wind or solar development within certain
Washington habitats. As mentioned in the data preparation section of this study, the 20
specific onshore habitat types present in the state of Washington have been combined to
form five general habitat types according to habitat similarity. Figure 11 shows the
distribution of these general habitat types across the Washington landscape.
85
86
Figure 11. General Habitats in Washington State
The forest and woodland habitats are located in the western and northern parts of
the state and most of the agriculture and pasture habitats are in the southeastern corner of
the state. These two habitat types cover the most land, representing 78% of the landscape
and about 33 million acres (Figure 12). The grassland and shrubland habitats are the third
largest habitat type, covering about 16% of the landscape (approximately 7 million acres)
and tend to border both the agriculture/pasture and forest/woodlands habitats. The
wetland and other hydrologic habitats as well as the urban habitat are dispersed
throughout the entire state and are the smallest habitat types, covering only 2% and 4% of
the landscape, respectively.
Figure 12. Habitat distribution in Washington
87
To understand the interaction between crucial habitat and wind or solar energy
resources within the five general Washington habitats, a local autocorrelation analysis
was completed for each of the three variables. This analysis identified significant
clustering of high values and low values across the landscape for each of the variables.
For the purposes of this study, there are four significant clustering outcomes or
interactions that are important for understanding the impacts within the general
Washington habitat. First, significant clustering of high wind or solar energy resources
identifies areas with high energy generating potential. Second, significant clustering of
most-crucial habitat lands represents areas with the highest conservation value. Third,
significant high conflict areas are locations where the significant most-crucial habitat
areas are also significant high wind or solar energy areas. Lastly, significant areas of low
conflict are areas where significant high wind or solar energy are also areas of significant
least-crucial habitat and have a low conservation value.
Wind Energy Resource Analysis
The outcome of this analysis with regards to wind energy resources shows greatly
dispersed areas of significant clustering across the entire state of Washington (Figure 13).
This is not surprising, since the wind power varies greatly as it moves across the
landscape, resulting in many small areas of significant clustering. Since the areas of
significant high wind energy resources are so small, it is also no surprise that only a very
few areas of high or low conflict were detected across the landscape. These are indicated
by the green and red colors in Figure 13.
88
89
Figure 13. Spatial interaction of wind energy resources and crucial habitat statewide
When assessing these results according to general Washington habitats, a better
picture of how these significant clusters are located across the landscape emerges, and
related risks and opportunities within each habitat can be studied. According to this
assessment, the grassland and shrubland habitat contains the highest portion of all
significant high wind resources (61.7%) and high conflict (64.74%) landscape clusters
(Table 5). This means that, while there are more preferred development locations for
wind energy in the grassland and shrubland habitat, there is also a high risk of
encountering high conflict areas in the landscape as well. However, when looking at
these opportunities for development and risks of high landscape conflict, the footprints of
these significant landscape clusters are quite small. The term footprint in this context is
defined as the portion of total habitat area that exhibits a particular significant landscape
cluster.
90
Table 5. Significant wind resource landscapes in each habitat as a percentage
of total significant landscape areas
General Habitat Type
Total Significant Area (acres)
Wetland, River, Lake, and
Reservoir Habitats
Urban and Mixed-Environments
Agriculture, Pasture, and
Mixed-Environments
Forest and Woodland Habitats
Grassland and Shrubland
Habitats
High
Conflict
Low
Conflict
High Wind
47,619
3,262
532,439
Most
Crucial
Habitat
4,545,264
0.48%
0.25%
0.26%
4.85%
0.00%
0.00%
0.52%
2.17%
0.63%
0.00%
2.19%
17.70%
34.15%
99.75%
35.32%
57.16%
64.74%
0.00%
61.70%
18.05%
Table 5 shows which general habitat types have a higher or lower risk of encountering
each significant landscape cluster type.
Table 6. Significant wind resource landscapes as a percentage of total habitat
area
General Habitat Type
All of Washington State
Wetland, River, Lake, and
Reservoir Habitats
Urban and MixedEnvironments
Agriculture, Pasture, and
Mixed-Environments
Forest and Woodland
Habitats
Grassland and Shrubland
Habitats
High
Conflict
Low
Conflict
High
Wind
43,197,857
0.11%
0.01%
1.23%
Most
Crucial
Habitat
10.52%
1,010,784
0.02%
0.00%
0.14%
21.83%
1,788,156
0.00%
0.00%
0.16%
5.52%
14,182,252
0.00%
0.00%
0.08%
5.67%
19,353,822
0.08%
0.02%
0.97%
13.42%
6,862,843
0.45%
0.00%
4.79%
11.96%
Total Habitat
Area (acres)
Table 6 shows the footprint of significant clustering within each general habitat type.
91
Only 4.79% of the total grassland and shrubland habitat lands are significant high
wind areas and less than 1% of the habitat lands are significant high conflict areas (Table
6). In addition, the grassland and shrubland habitat also has a fairly high presence of most
crucial habitat clustering, with 18.05% of this landscape cluster type occurring in this
habitat, and a footprint of 11.96% in the total grassland and shrubland habitat area. This
means that there is some risk of encountering significant most-crucial habitat areas when
interacting within this habitat type. While these opportunities and risks do marginally
exist within the grassland and shrubland habitat landscape, the only meaningful impact in
relation to wind energy development are the areas of significant high wind energy
resource clusters. For a visual representation of the grassland and shrubland general
habitat type and the significant landscape clustering within this habitat, see Figure 14.
92
93
Figure 14. Spatial interaction between wind energy resources and crucial habitat in the Washington
grassland and shrubland habitats
The forest and woodland general habitat has the highest portion of significant
most-crucial habitat clusters (57.16%) and has a habitat footprint of 13.42% (Table 5 and
6). Although significant clusters of most-crucial habitat alone have no implication for
wind energy development, knowing there are areas of most-crucial habitat could inform
other industries working within the forest and woodland habitat of the environmental
risks of some areas. Regarding low conflict areas, the forest and woodland general habitat
also contains the highest portion of this significant landscape cluster type (Table 5). At
first glance having 99.75% of all significant low conflict landscape clustering in a single
habitat type seems like it would be an important discovery. This means the greatest
opportunity for identifying areas of high wind energy potential in concert with lesser
crucial habitat landscapes are within a single habitat type and would represent the most
favorable wind energy development locations. However, when considering the total
footprint of this significant landscape cluster type it becomes a moot point since this only
represents 0.02% of the total forest and woodland habitat (Table 6). For a visual
representation of the grassland and shrubland general habitat type and the significant
landscape clustering within this habitat, see Figure 15.
94
95
Figure 15. Spatial interaction of wind energy resources and crucial habitat in Washington
forest and woodland habitats
The wetland, river, lake, and reservoir habitat type is also impacted by significant
most-crucial habitat clusters. Despite this habitat only containing a small portion of all
significant most-crucial habitat clusters, a large habitat footprint of 21.83% is observed
(Table 6). This is likely due to the fact that this habitat type spans the smallest area across
the landscape and, for Washington, represents a sensitive ecosystem with many
threatened or endangered species present (Cowardin et al. 1979; Washington Department
of Fish and Wildlife 2005). Despite crucial habitat clusters having a higher impact in this
habitat, it is not unexpected and there are already many policies in place to monitor and
regulate wetland and other riparian ecosystems across the state of Washington. For a
visual representation of the additional habitat types and the significant landscape
clustering not presented in this chapter, see Figure 20–22 in the appendix.
Overall, the interaction between crucial habitat and wind energy resources does
not have a very large impact on most of the five general habitat types in Washington
State. The footprints of significant landscape cluster types are in most cases very small or
non-existent. However, the grassland and shrubland habitat is impacted most, out of all
other habitat types, with the most significant high wind resources.
Solar Energy Resource Analysis
The outcome of this assessment with regard to solar energy resources is quite
different from that of the wind energy resources analysis. Solar energy resources are
concentrated, rather than dispersed, forming one large cluster of significant high solar
energy resources in the southeastern corner of Washington State (Figure 16). Within this
area, there are large significant clusters of both low conflict and high conflict landscapes,
96
as indicated by the green and red areas in Figure 16. These large, significant clusters
Figure 16. Spatial interaction of solar energy resources and crucial habitat statewide
could have important impacts on the five general habitat types in Washington.
97
Looking at these significant landscape clusters in relation to general habitat types
in Washington State, there are certain habitat types that are impacted more than others.
Most notable is the agriculture, pasture, and mixed-environment habitat. This habitat
contains the highest portions of the high solar and low conflict landscape cluster types
being 58.02% and 97.31% respectively (Table 7). The footprints for these landscape
cluster types are also notable, with 35.51% of the agriculture and pasture habitat area
being significant high solar resource areas, and 13.16% being areas of low conflict (Table
8). This means that the agriculture and pasture habitat type is most suited to development
of solar energy, given that 48.67% of this habitat area has high solar energy resources and
in some of those areas there are significant clusters of less-crucial habitat with a low
conservation value. The areas of low conflict would be areas to investigate the feasibility
of solar development first, followed by the remaining areas of significant high solar
density. For a visual representation of the agriculture, pasture, and mixed-environment
general habitat type and the significant landscape clustering within this habitat, see
Figure 17.
98
Table 7. Significant solar resource landscapes in each habitat as a percentage
of total significant landscape areas
General Habitat Type
Total Significant Area (acres)
Wetland, River, Lake, and
Reservoir Habitats
Urban and Mixed-Environments
Agriculture, Pasture, and
Mixed-Environments
Forest and Woodland Habitats
Grassland and Shrubland
Habitats
High
Conflict
Low
Conflict
High Solar
963,154
1,917,413
8,680,259
Most
Crucial
Habitat
3,810,422
3.72%
0.10%
1.66%
4.77%
2.54%
1.57%
2.54%
1.86%
33.72%
97.31%
58.02%
12.31%
15.78%
0.30%
11.15%
63.94%
44.13%
0.72%
26.37%
11.10%
Table 7 shows which habitat types have a higher or lower risk of encountering each
significant landscape cluster type.
Table 8. Significant solar resource landscapes as a percentage of total habitat
area
General Habitat Type
All of Washington State
Wetland, River, Lake, and
Reservoir Habitats
Urban and MixedEnvironments
Agriculture, Pasture, and
Mixed-Environments
Forest and Woodland
Habitats
Grassland and Shrubland
Habitats
Total Habitat
Area (acres)
High
Conflict
Low
Conflict
High
Solar
43,197,857
2.23%
4.44%
20.09%
Most
Crucial
Habitat
8.82%
1,010,784
3.55%
0.19%
14.27%
17.98%
1,788,156
1.37%
1.68%
12.34%
3.96%
14,182,252
2.29%
13.16%
35.51%
3.31%
19,353,822
0.79%
0.03%
5.00%
12.59%
6,862,843
6.19%
0.20%
33.36%
6.16%
Table 8 shows the footprint of significant clustering within each general habitat type.
99
100
Figure 17. Spatial interaction of solar energy resources and crucial habitat in Washington
agriculture, pasture, and mixed-environment habitats
Within the agriculture and pasture habitat there is also the smallest footprint of
significant most-crucial habitat clusters, with only 3.31% of the total agriculture and
pasture area (Table 7 and 8). This means that among the favorable amount of high solar
energy resources there is also the smallest risk of encountering large significant areas of
most-crucial habitat. In contrast, the wetland, river, lake, and reservoir habitat, again, has
the highest risk of encountering significant most-crucial habitat, with a footprint of
17.98% of the total habitat area being significant most-crucial habitat areas. As explained
above, this is not unexpected.
Finally, the highest portion of high conflict landscape clusters falls within the
grassland and shrubland habitat (44.13%). This habitat also has the highest footprint of
significant high conflict lands, being 6.19% of the habitat area (Table 8). This is
interesting because the grassland and shrubland habitat type has the second largest
portion of significant high solar energy resources (26.37%) and a total footprint that is
33.36% of the total habitat area. This footprint is almost as large as the footprint for the
agriculture and pasture habitat type. However, these results suggest that, despite having
an ample amount of significant high solar energy resources across the habitat, there is
also the highest risk of also encountering most-crucial habitat clusters within those high
solar areas. For a visual representation of the grassland and shrubland habitat and the
significant landscape clustering within this habitat, see Figure 18. For a visual
representation of the additional habitat types and the significant landscape clustering not
presented in this chapter, see Figure 23–25 in the appendix.
101
102
Figure 18. Spatial interaction of solar energy resources and crucial habitat in Washington grassland
and shrubland habitats
Overall, the interaction between crucial habitat and solar energy resources has a
moderate impact on two of the five Washington habitats. The agriculture, pasture, and
mixed-environment habitat type has the highest potential for solar energy development as
well as the lowest risk of infringing upon landscapes with a high conservation value.
These areas would be preferred development locations to investigate for future
development. In contrast, the grassland and shrubland habitat type also has a high
potential for solar energy development, but there is also a high risk of encountering lands
that have high conservation value. In this habitat, development proposals should be
scrutinized more thoroughly to balance the needs of renewable energy development and
habitat conservation priorities.
Knowing where the significant landscape clusters are and how much of a
particular habitat is impacted by the significant landscape clusters will help land use
managers balance renewable energy development and habitat conservation goals. It is
important to recognize that the grassland and shrubland habitat is impacted by significant
landscape clustering for both wind and solar energy resources. While the impact may be
small-to-moderate, proposals for wind or solar energy development in this habitat should
be reviewed closely to ensure the best balance between development and conservation
efforts.
103
Chapter 5: Discussion
Understanding the interactions between wind and solar energy development
potential and crucial habitat across the landscape provides a first glimpse into how these
two environmentally important initiatives can synergize and minimize landscape-level
conflicts. This study’s five research questions addressed how these two initiatives interact
and potentially conflict across Washington State, within three different contexts—
existing wind farm developments, suitable landscapes for future energy development, and
areas of opportunity or conflict within Washington habitats. This provides many insights
into how a landscape-level conservation indicator could be utilized to better optimize the
priorities of both renewable energy development activities and statewide wildlife
conservation priorities. The context of each research question addressed in this study will
be discussed at length in the following sections, highlighting the opportunities for
improved land use management and exploring how the concept of multifunctional
landscapes could begin to be applied across the state of Washington.
5.1 Existing Washington Wind Farms
Knowledge of how existing Washington wind farms were developed in relation to
crucial habitat sheds light on how well environmental impact assessments (EIA) at the
local scale captured landscape-level habitat conservation priorities. While, overall, there
is a smaller portion of rank 1, most-crucial habitat lands in existing wind farms than what
was observed statewide, the results of the assessment also showed that the distribution of
crucial habitat is varied from one wind facility to another. This indicates that some wind
104
farms were sited in locations where renewable energy development activities and
landscape impacts had minimal engagement with lands of high conservation value.
Others, however, impose a direct conflict across the landscape. Of particular interest were
the Vantage Wind and the Wild Horse Wind Power project, since they were the most
poorly sited wind farms according to crucial habitat distribution. With the highest
landscape percentage of most-crucial habitat rank 1 within these wind farm sites,
questions arise regarding the capability of the EIA process to capture and manage
landscape-level habitat conservation priorities.
Upon investigation of the EIA documentation for both of these wind farms, it is
apparent that different levels of rigor were used in pre-development planning and
environmental assessment. For the Wild Horse Wind Power project, a very thorough draft
EIA was completed and released in August of 2004. In this assessment, Wind Ridge
Power Partners, LLC, highlighted many development impacts across the proposed
landscape and identified tangible and specific mitigation and remediation activities with
an aim to minimize the negative effects of the project (Jones & Stokes et al. 2004).
Nearly all the known negative impacts of wind energy development as discussed in the
second chapter of this study were identified and addressed throughout the EIA. Impacts
of greater concern included habitat disturbance and loss of shrub-steppe grasslands and
sensitive lithosol habitat; sage grouse and big game animal avoidance of the area;
presence of a few rivers and wetland areas; threats of exotic and invasive species
dispersal; and the hedgehog cactus (Pediocactus nigrispinus—also known as snowball
cactus)—a Washington State review listed species—was identified within the project area
(Jones & Stokes et al. 2004).
105
Given the long list of potential environmental impacts, the facility planners in
charge of the Wild Horse Wind Power project identified many mitigation strategies in
acknowledgement of the dangers posed by development, to keep biodiversity and
ecological health as intact as possible. Actions that were explicitly identified included
fencing and avoiding wetlands, rivers, and known locations of the hedgehog cactus;
washing construction vehicles before entering the premises to reduce the risk of exotic
and invasive species dispersal; minimizing new road construction to reduce landscape
fragmentation; replanting disturbed landscapes with native species; using underground
transmission lines to reduce risks to avian species; and on-going monitoring of animal
behavior within the project site. In addition, 600 of the 8,600 acres within the project area
was partitioned to be fenced and left as native shrub-steppe and grassland habitat, serving
as a large corridor to connect with adjoining state lands (Jones & Stokes et al. 2004). This
land serves as a protected reserve with no risk of future development, is inaccessible to
livestock grazing and other ranching activities, and serves to protect biodiversity and
natural habitats in the area. Overall, the Wild Horse Wind Power project EIA shows a
great concern for environmental impacts, incorporating significant measures to minimize
the potential impacts of wind energy development.
In contrast, the Vantage Wind project—built after the Wild Horse Wind Power
project in 2008 and only 6 miles away—did not even complete a full EIA prior to
development. Rather than an EIA, a determination of non-significance (signifying the
project location is not likely to have significant adverse environmental impacts and
including a less rigorous environmental checklist) was completed and submitted as part
of the permitting process. The environmental checklist did state that several wildlife and
106
ecological surveys were completed, but the mitigation proposals identified within the
plan only addressed the most common concerns of wind energy development (White et
al. 2003; Kittias County Community Development Services 2008). Given that the crucial
habitat rating for a majority of the Vantage Wind project site is of the most-crucial
habitat rank 1, a determination of non-significance was probably not appropriate for this
site. A more thorough environmental investigation should have been completed to
understand what environmental factors caused the landscape to be rated as crucial habitat
rank 1, and specific avoidance and mitigation tactics should have been employed.
Comparing the environmental assessments of these two wind power projects, it is
clear that, despite being closely located and in similar ecological regions, the local EIA
process does not always capture landscape-level conservation concerns. On the one hand,
the Vantage Wind project seems to have been completed with minimal effort related to
environmental permitting, and has evidently established the primary purpose of the
landscape to be energy production, regardless of high priority conservation values in the
area. On the other hand, the Wild Horse Wind Power project did complete a rigorous
environmental study of the project landscape and carefully crafted mitigation strategies to
minimize negative environmental impacts. This project demonstrates how the EIA
process can facilitate the optimization of renewable energy development and habitat
conservation efforts, to work toward establishing a multifunctional landscape as explored
by Howard et al. (2013) and Reyers et al. (2012). Here, land use planning seems to have
been balanced between energy, ecological, economic, and social priorities. This plan
promotes energy production along with the protection and preservation of biodiversity
107
and ecological health, enabling ranching activities in some areas, and allowing hunting
and other recreational activities within the same landscape.
For future renewable energy developments, crucial habitat assessments could
enhance environmental impact assessments across different landscapes, informing of
potential landscape conflicts. Even if future renewable energy development is proposed
in more crucial habitat lands, the Wild Horse Wind Power project is an example of how
careful planning and landscape design can attempt to optimize such landscapes.
However, knowing the level of landscape conservation value can also function as guide
to initial project planning and siting. This indicator could inform the rigor of
environmental assessment that should be undertaken across different landscapes as well
as additional costs that may be associated with environmental mitigation activities on
more crucial habitat lands (Western Governors’ Wildlife Council 2013a). This new
measure of landscape-level conservation priorities will be essential to improving the
existing renewable energy development practices, to move toward the successful design
and management of multifunctional landscapes in the future.
5.2 Wind and Solar Energy Development Landscapes
Having an understanding of the interaction between landscape-level habitat
conservation priorities and suitable renewable energy development locations provides
insight into the levels of landscape conflict between these two initiatives. As shown in the
outcome of the analysis for Washington State, suitable wind energy development
locations display a crucial habitat distribution that is similar to what is observed across
the entire state. Despite having a lower impact on high conservation value lands than
108
what is observed statewide, there is still a moderate conflict between these two initiatives
on the landscape. This is observed when considering the portions of land rated as crucial
habitat values 1–3, given that nearly 90% of suitable wind energy landscapes fall within
these crucial habitat values. These would be lands with observed or presumed threatened
or endangered species; federal and state candidate and sensitive species; priority
ecological systems of concern; confirmed heritage vegetation; important estuary and
wetland habitats; large, high quality natural areas; or most-important habitats for species
of recreational and economic importance.
Applying estimated annual energy generation per acre to suitable wind energy
development lands according to various crucial habitat levels further reveals that future
wind energy development could exclude the most crucial habitat lands of ranks 1 and 2,
and still expand existing levels of annual wind energy production by 440%. This means it
is likely that a moderate level of environmental mitigation will still be required for future
wind energy development, but that wind energy growth could be better balanced with
habitat conservation priorities. Under this landscape management strategy, habitats with
observed threatened and endangered species, some federal and state candidate and
sensitive species, priority ecological systems of concern, confirmed heritage vegetation
communities, important estuaries, and spawning ground for aquatic species of
recreational importance, will be protected from disturbance. The remaining landscape
conflicts in areas with lower conservation value (crucial habitat rank 3–6) can then be
identified, assessed, and managed at the local level. Overall, this landscape management
strategy will work toward optimizing the requirements for both the habitat conservation
and wind energy development initiatives.
109
For solar energy resources, a crucial habitat distribution signifying a moderate to
high conflict is observed across the landscape, with 50% of the suitable solar
development landscape having a crucial habitat rank of 1 or 2. However, given that
suitable solar energy development locations span a large geographic area and the land
required for energy generation is small, there is a great opportunity to target less-crucial
habitat lands with low conservation value for future solar energy development. According
to this assessment, suitable solar energy development locations in the least-crucial habitat
lands of rank 5 can be exclusively targeted and still have the potential to generate an
additional 65% of the total annual Washington State energy production. This energy
generation potential is huge, and the conflict between habitat conservation priorities and
solar energy development can be reduced to the absolute minimum. In addition, the
spatial locations of these landscapes are clustered in the southeastern portion of
Washington State, providing large contiguous areas to investigate the feasibility of future
solar energy development in a local context. This is an example of how landscape-level
planning can be utilized to truly optimize the land use requirements of multiple initiatives
to achieve the most desired outcome.
Solar Energy Resource Analysis Limitations
The initial assessment of suitable solar energy development locations in relation
to crucial habitat alludes to major solar resources in eastern Washington State. There are,
however, some limitations to this estimate that may have caused the results of this
analysis to overestimate the potential energy generation per acre across the landscape.
Unlike wind energy, commercial solar energy production has not largely been established
110
in the state of Washington. Due to the lack of state-specific energy information, this study
used a national average landscape footprint per annual gigawatt hour (GWh) of energy
production figure to estimate the potential energy generation of suitable solar resources in
Washington State. The national average figure was calculated from 166 existing or under
construction U.S. solar facilities in locations with a large range of solar irradiance
resource levels (Ong et al. 2013). This means solar facilities in areas with the highest
solar irradiance levels would require less land to generate solar electricity, while areas
with the lowest solar irradiance levels would require more land than what is represented
by the national average.
Washington is located at a northern longitude and has some of the lowest levels of
solar irradiance suitable for commercial solar energy production—Washington annual
average solar irradiance levels range from 3.11–5.3 kWh/m2/day, with a mean of 4.44
kWh/m2/day. In contrast, the annual average solar irradiance levels across the United
States range from 3.11–7.03 kWh/m2/day, with a mean of 5.22 kWh/m2/day (National
Renewable Energy Laboratory 2013a). This difference in solar resource levels is not
precisely reflected in the landscape footprint requirements for Washington solar energy
production. As indicated above, it is likely that a larger landscape footprint per GWh of
annual energy production would be required in Washington State. This would result in a
lower annual solar energy production estimate. Future assessments of how crucial habitat
relates to suitable solar energy resources should attempt to account for the lower solar
irradiance levels in Washington State.
111
Opportunity for Future Research
Suitable wind and solar development locations as defined in this study only
account for lands available for development and appropriate wind and solar energy
resource levels required for commercial energy production. In reality, there are many
more factors that influence suitability for development. To take this assessment further, a
feasibility study should be conducted to more precisely identify preferred renewable wind
and solar energy development locations in Washington State. This research opportunity
would build upon the existing analysis conducted in this study and further restrict suitable
development locations to reflect additional economic, social, and additional technological
conditions that impact renewable energy development at local scales.
To reflect economic conditions associated with renewable energy development
locations, proximity to existing electrical transmission lines and substations should be
included. The associated cost to connect to or create new electrical transmission lines
would begin to more realistically pinpoint the most preferred renewable energy
development locations. Further, the social perspectives of support or opposition for
specific renewable energy technologies could be modeled to reflect the economic
implications of local acceptance or resistance to development. This could be done at a
county level with the general assumption that areas of high opposition would either
prevent development or require more time and investment in the siting and permitting
phases to negotiate an acceptable development location and plan with the local public.
Finally, the topography of a landscape is a major consideration in determining feasibility
and least-cost options for renewable energy development. There should at least be
modeling of slope gradients according to the requirements of the specific technologies,
112
using a weighting system that associates cost with the level of landscape preparation
required for development.
This more advanced model of suitable energy development lands would present
the opportunity to conduct a more detailed analysis of how landscape-level conservation
priorities relate to the most preferred renewable energy development locations across
Washington State. This analysis would provide a more realistic indication of the
landscape-level conflicts between renewable energy development and crucial habitat than
what is presented in this study. With this information, the appropriate energy and wildlife
stakeholders could identify more precise opportunities to approach land use management
from a multifunctional perspective, to optimize land uses for both purposes.
5.3 Washington Habitats
As explored in the first three research questions, the interaction between crucial
habitat and existing as well as suitable renewable energy development locations reveals
the general landscape-level conflict between the two initiatives. These analyses are
important to frame the high-level landscape interactions, but do not identify specific risks
or opportunities that could aid in optimizing the land use between crucial habitat and
renewable energy development. By investigating how significant spatial clustering of
crucial habitat interacts with significant spatial clustering of wind and solar energy
resources within Washington habitats, risks and opportunities of future development
begin to appear.
Of particular interest are the areas of significant high-conflict and low-conflict
clusters and how they are placed on the landscape. These areas specifically define the
113
positive and negative interactions between significant clustering of high renewable
energy resources and significant clustering of both high- and low-crucial habitat areas.
Looking closer at all areas of significant clustering and interactions within the context of
general Washington habitats further define the risks and opportunities for future wind and
solar energy development. This information helps explore how renewable energy
development will impact the landscape according to the ecology and conservation
priorities unique to the various habitat types.
It is important to remember that the 5 general habitat types defined in this study
are a combination of several, more specific, habitat types identified by researchers at the
Northwest Habitat Institute (NWHI). The Washington Department of Fish and Wildlife
(WDFW) has taken these same specific habitats and prioritized them according to
conservation importance as part of the Washington Comprehensive Wildlife
Conservation Strategy (CWCS). Within the Washington CWCS, each specific habitat
type is identified as priority 1, priority 2, or “other,” according to the number of species
of greatest conservation need (SGCN) that occur within each habitat. This information, in
conjunction with specific, known land use threats to the habitat types, enables a thorough
assessment of how renewable energy development is likely to impact the general habitats
as defined in this study. This is discussed in the following sections.
Wind Energy Resource and Crucial Habitat Analysis
In the analysis of significant landscape clustering and interactions between wind
energy resources and crucial habitat, results show that the grassland and shrubland
habitats, forest and woodland habitats, as well as the wetland, river, lake, and reservoir
114
habitats will be impacted the most. However, there were not meaningful impacts related
to significant high- or low-conflict areas on the landscape. Nevertheless, there are still
important risks and opportunities that can be perceived from the landscape clustering of
high wind resource areas and most-crucial habitat areas independently.
Significant High-Wind Clustering
The grassland and shrubland habitats contained the highest portion of significant
high wind clustering and had a landscape footprint of 4.79% within the habitat. This
means future wind energy development is most likely to occur in the grassland and
shrubland habitats because this is where spatial clustering of high wind energy resources
occur the most. However, there are a number of challenges that will need to be managed,
due to the history and ecology of this habitat type.
Since 1889, the grassland and shrubland habitats have suffered the highest
landscape conversion rates of all habitat types. Approximately 50% of historic shrubsteppe habitats and 70% of historic grassland habitats have been converted to agricultural
landscapes (Washington Department of Fish and Wildlife 2005). This history of land use
conversion has resulted in much native habitat loss, habitat fragmentation, and the
introduction of invasive species, which all continue to be the largest threats to the
remaining natural lands. These habitat threats, in addition to a large number of SGCN
present in these landscapes, has resulted in much of this habitat being categorized as
priority 1 conservation habitat.
In relation to wind energy development, the land use threats most impacting the
grassland and shrubland habitats will be encountered as a consequence of the wind
115
facility development process to some extent. A large network of roads is required to
access and maintain the wind turbines, effectively fragmenting the landscape.
Additionally, an increased presence of construction equipment and maintenance vehicles
increases the potential for invasive species to establish in new areas (Washington
Department of Fish and Wildlife 2005). Finally, during the construction phase of wind
farm development, large areas of the landscape are often modified as staging locations for
turbine parts and equipment (McDonald et al. 2009). Although these areas are remediated
after construction, habitat loss does occur and natural lands are altered. Given the
importance of this habitat type to the Washington CWCS and the many potential negative
impacts from wind energy development, careful environmental planning and mitigation
should be conducted for all wind energy development projects occurring within grassland
and shrubland habitats.
Significant Most-Crucial Habitat Clustering
Regarding significant clustering of most-crucial habitat, wetland, river, lake, and
reservoir habitats, forest and woodland habitats, as well as grassland and shrubland
habitats all had notable landscape footprints. Within wetland, river, lake, and reservoir
habitats 23.83% of the habitat area contained significant most-crucial habitat clusters. For
forest and woodland habitat, 13.42% of the landscape contained significant most-crucial
habitat clustering. Finally, within the grassland and shrubland habitats, 11.96% of the
habitat area contained significant most-crucial habitat clustering. This means that, when
interacting within these habitat types, there is a fair chance that large areas of mostcrucial habitat will be encountered. While these significant most-crucial habitat clusters
116
do not have any relation to wind energy development, they could be important to other
natural resource related industries interacting within these habitat types, such as forestry
or mining. Such industries should complete a more thorough investigation of how,
exactly, industry practices would be impacted by the significant areas of most-crucial
habitat. However, this is out of scope of this study.
Solar Energy Resource and Crucial Habitat Analysis
In the analysis of significant landscape clustering and interactions between solar
energy resources and crucial habitat, the results are quite different than the outcome of
the wind energy resource analysis. Regarding solar energy resources, a single, large
significant, high solar energy cluster was identified in the southeastern part of
Washington State. Within this large, significant cluster, several significant clusters of
high-conflict and low-conflict areas were also identified. The habitat types most impacted
by these results are the agriculture, pasture, and mixed environment habitats as well as
the grassland and shrubland habitats, as discussed in detail below. The wetland, river,
lake, and reservoir habitat, as well as the forest and woodland habitat, were again
impacted by significant most-crucial habitat clusters. However, since the impact is the
same as was described in the wind energy resource and crucial habitat analysis section
above, it will not be discussed again in this section.
Agriculture, Pasture, and Mixed-Environments Habitat
The agriculture, pasture, and mixed-environments habitat contained the highest
portions of significant high solar energy resource areas and significant low landscape
117
conflict areas out of all other habitat types. In addition, these significant landscape
clusters had large landscape footprints of 35.52% and 13.6% of the total agricultural,
pasture, and mixed-environments habitat respectively. This means that there are more
high solar energy resources as well as low-conflict areas in this habitat than any other
habitat type. This outcome represents a potential opportunity to plan and design lowimpact, multifunctional energyscapes for solar energy in the agriculture, pasture, and
mixed environment habitats in the state of Washington.
With regard to the Washington CWCS conservation priority, the
agriculture, pasture, and mixed environments habitat has been categorized as an “other”
priority habitat. This means conservation of this habitat type is not of high importance
compared to other habitat types in the state. This makes agriculture, pasture, and mixed
environment habitats a good place to consider developing solar energy facilities from a
habitat conservation perspective, since land use in this habitat does not have many SGCN
and development impacts are less likely to negatively affect Washington conservation
goals.
As identified in the second chapter of this study, solar energy development does
have negative environmental impacts as a result of facility construction. One of the
primary negative environmental impacts is the near-full conversion of the landscape
required for energy production, which for natural lands results in habitat loss (McDonald
et al. 2009). However, in agriculture, pasture, and mixed-environment habitats the impact
of habitat loss is likely to be much smaller because this habitat type is already a product
of landscape conversion (Washington Department of Fish and Wildlife 2005).
Additionally, the road construction that is often associated with renewable energy
118
development, and which results in habitat fragmentation, is also likely to be minimal.
Agricultural landscapes generally have extensive road networks already in place which, if
utilized in the construction and operation of solar facilities, would greatly reduce or
mitigate the need for new roads. Furthermore, if development were targeted in the
significant low landscape conflict areas of this habitat type, environmental impacts would
again be reduced, since these areas represent landscape clusters of low conservation
value. This means that if negative environmental impacts were incurred in the
development of solar facilities, those impacts would only minimally conflict with
landscape level conservation priorities, if at all.
While the agriculture, pasture, and mixed-environment habitat clearly presents an
opportunity to create multifunctional landscapes benefiting energy production and habitat
conservation priorities on the landscape, there are challenges that are likely to surface that
are unique to this habitat type. To develop solar energy facilities, existing functional
agricultural lands would likely be replaced with solar energy technologies. This would
result in a reduction of existing agricultural and pasture lands. If solar energy
development in this habitat were not strategically designed to fit a multifunctional
landscape model, the landscape conversion would simply transition from one land use
type to another without improving the overall function of the land. This is more of a
concern with solar technologies, since solar panels can be constructed close together,
utilizing the entire landscape footprint of the facility for energy production alone.
In the design of multifunctional landscapes in agriculture, pasture, and mixedenvironment lands, creative construction and placement of solar technologies would be
required. Ideally solar development would strategically minimize the reduction of
119
existing agricultural production while enabling the additional landscape function of
energy production. For example, rather than converting an entire agricultural field to a
solar facility, solar technologies could stretch along the edges of the field and along
existing roads. With this type of landscape design, most of the existing agricultural
production could continue, while adding the production of solar energy to the overall
functionality of the landscape.
In addition to the challenge of strategic, multifunctional landscape design,
multiple stakeholders such as farmers, landowners, and energy companies would all need
to be involved in the design and operation of the facilities. With multiple stakeholders
involved, conflicts over landscape needs and potential influences on neighboring lands
would need to be identified and managed for the least possible impact to all parties
(Harden et al. 2013). As identified in the ecosystem approach to landscape design
management as discussed in Chapter 2 (beginning on page 29), an interdisciplinary
collaboration and partnership would be required to successfully adjust the existing
landscape functions and design in order to introduce permanent solar facilities to the
landscape. However, despite these additional challenges, the opportunity to create
multifunctional solar landscapes should still be explored and, if successful, could
contribute greatly to the growing literature on landscape ecology and planning (Reyers et
al. 2012).
Grassland and Shrubland Habitats
The grassland and shrubland habitats also contained high portions of significant,
high solar energy resource clusters as well as the highest portion of high-conflict
120
landscape clusters with landscape footprints of 33.36% and 6.19% of the total habitat
area, respectively. This means that over one-third of the total grassland and shrubland
habitat is suitable for solar energy development, and some of that land is also significant
most-crucial habitat cluster areas. However, unlike the agriculture, pasture, and mixed
environment habitat, the grassland and shrubland habitats include several CWCS priority
1 habitats, making this habitat type more important to statewide conservation goals.
Solar energy development on this landscape will likely pose higher environmental risks
to the habitat than what would be seen in the agriculture, pasture, and mixed-environment
habitats.
With regard to solar energy development in the grassland and shrubland habitat,
this habitat is again likely to be impacted by habitat loss, habitat fragmentation, and the
threat of invasive species as a consequence of development. However, in this case, the
threat of habitat loss is likely to be the largest threat of solar development. As mentioned
above, solar energy development often requires landscape modification and full
landscape development of the energy facility site. Since over half of the native habitat has
already been lost over the past 125 years and much of the habitat is a CWCS conservation
priority 1 habitat, habitat loss poses a higher environmental risk in this landscape than in
some other habitat types. In addition, if road construction were required, some impact
from habitat fragmentation and threat of invasive species could also be incurred.
All in all, the grassland and shrubland habitats as well as the agriculture, pasture,
and mixed-environment habitats will be impacted the most according to significant
landscape clustering and the interaction between wind and solar energy resources and
crucial habitat in the state of Washington. While the agriculture, pasture, and mixed-
121
environment habitat presents an opportunity to explore multifunctional solar landscapes,
there are substantial risks associated with energy development in the grassland and
shrubland habitats. Given that there are significant, high energy resource clusters for both
wind and solar energy in the grassland and shrubland habitat, it is highly likely that some
renewable energy development will occur. However, knowing there is a significant
interaction within this habitat type should better enable energy developers and
conservation biologist to work together to identify the best mitigation processes and
environmental safeguards for developing in this important habitat. Through this process,
the interests of both initiatives can be optimized by balancing the priorities and trade-offs
for each.
5.4 Policy Implications
Renewable energy development and habitat conservation initiatives are largely
supported and enforced by state and federal policy for successful implementation of state
and national targets and goals. Regarding the land use interactions of wind and solar
energy development and habitat conservation initiatives, there are two policy areas that
relate to the results and discussion sections of this thesis. These are the Washington State
Energy Independence Act, also referred to as the Washington Renewable Portfolio
Standard (RPS), and the Washington State Environmental Protection Act.
Washington State Energy Independence Act
In 2006, Washington State passed the Energy Independence Act, which states
that, by January of 2020, large utility companies will be required to obtain 15% of the
electrical load they supply from new, renewable energy technologies in Washington
122
State. These renewable energy facilities could produce energy from water (but not large
hydroelectric facilities), wind, solar, geothermal, wave, ocean, tidal, landfill and sewage
treatment gas, and biodiesel (not from crops grown on cleared old-growth or first-growth
forests) (Chapter 19.285 RCW: ENERGY INDEPENDENCE ACT 2006). However, to
constitute a “new” renewable energy facility, operation must have begun after March of
1999 and be operating in Washington State to count toward the 15% target. There are two
incremental targets leading up to the 15% target in year 2020, to encourage a smooth
transition to a more renewable energy platform. These are (1) by January of 2012 3% of
electrical load, and (2) by January of 2016 9% of the electrical load should be supplied by
new renewable energy resources (Chapter 19.285 RCW: ENERGY INDEPENDENCE
ACT 2006). Further, to enforce this legislation, utility companies are required to pay to
the state of Washington a penalty of $50 per megawatt hour (MWh) of energy that fails to
meet the renewable energy targets.
This legislation is one of the primary drivers for the growth of renewable energy
technologies such as wind and non-commercial solar in the state of Washington. Further,
since the final renewable energy target is not until the year 2020, this policy will continue
to encourage the growth of renewable energy technologies across the state. While the
Energy Independence Act does support climate mitigation through the offsetting of
greenhouse gas emissions, it does not consider any potential for other negative
environmental impacts. Given the financial penalty incurred for non-compliance, this
policy effectively spurs development regardless of other landscape concerns such as
habitat conservation.
123
Fortunately, through the analyses conducted in this study, it seems as though
renewable energy development for commercial wind and solar energies could contribute
substantially to this energy target with only moderate to low impact to habitat
conservation initiatives. With a current contribution of 1.8% of Washington’s energy
generation (Northwest Power and Conservation Council 2013), wind energy could
expand to contribute an additional 7.9% of state energy generation within less crucial
habitat lands ranked 3–6. Solar energy could potentially target the least crucial habitat
lands with a rank of 5 and contribute enough energy to meet or exceed the entire 15%
state renewable energy target. While, in this case, renewable energy development can
occur with minimal negative impacts to the environment, future policies should assess
land use impacts when establishing targets and penalties. This would promote land use
management within a multifunctional landscape perspective without prioritizing a single
land use function.
Washington State Environmental Protection Act (SEPA)
The State Environmental Protection Act (SEPA) is the most powerful legislative
tool to protect the environment and aims to “Utilize a systematic, interdisciplinary
approach which will insure the integrated use of the natural and social sciences and the
environmental design arts in planning and in decision making which may have an impact
on man's environment” (Washington State 1971; White et al. 2003). This policy defines
the state regulations for making land use changes in the state of Washington. As a general
regulation for compliance to SEPA, an environmental impact assessment (EIA) is
required to identify probable, significant adverse environmental impacts and discuss
124
mitigation strategies to minimize the final impact. Once identified, it is expected that
development will occur only after all practical means and measures have been employed
to “create and maintain conditions under which man and nature can exist in productive
harmony” (Washington State 1971).
The Washington Energy Facility Site Evaluation Council (EFSEC) is the
regulatory agency that manages and approves the permitting process to site and begin
development for new energy facilities (Energy Facility Site Evaluation Council 2014).
Compliance with SEPA regulations and EIA acceptance largely falls with the
Washington EFSEC. This agency is highly effective in enforcing compliance with
environmental policies, and maintains a strict procedure for obtaining required permits
for energy facility development. However, the policies and EIAs associated with energy
development are generally assessed and reviewed at a local scale unique to the project
site. Industry best practices have been developed over the years to include the assessment
of broader landscape-level impacts and mitigation strategies, but these practices are not
entirely enforceable by law. In fact, the rigor and extent of EIA completion have been
inconsistent, as discussed previously, when comparing the EIA assessments for the
Vantage Wind and Wild Horse Wind Power projects.
Here lies an opportunity to improve the EIA process and include a broader
investigation of land use impacts at the landscape level. Completing EIAs at a local scale
will identify specific threatened species and sensitive habitats such as wetlands.
However, landscape-level conservation priorities such as habitat connectivity are not
often included in the local-level assessments. Using the crucial habitat assessment as an
environmental indicator, the level of rigor and scope of assessment required for EIAs
125
could be better defined. While assessment of crucial habitat is not meant to be a
regulatory requirement, this indicator could be utilized as an industry best practice for the
improvement of the EIA process. This best practice would highlight areas of most-crucial
habitat, so that they receive more thorough environmental surveys and investigation of
impact. Areas of less-crucial habitat could continue with the current requirements. In
addition, landscape-level habitat conservation priorities, such as habitat connectivity,
would also be captured in the EIA process, and mitigation strategies would further
support the creation of multifunctional landscapes.
While both the Energy Independence Act and State Environmental Protection Act
are effective policies encouraging better environmental stewardship, both can continue to
be improved, in order to capture broader landscape-level impacts and perspectives. To
further encourage this transition, other policies, such as incentives contributing to both
energy and habitat conservation goals, could also be employed. For example, the
permitting process could be expedited if proposed renewable energy facilities were sited
on least-crucial habitat lands. This would still encourage renewable energy development
working toward energy independence targets, while focusing on landscapes having the
least conflict with habitat conservation priorities. While there are many creative ways that
could encourage this shift in land use perspective, the idea of multifunctional landscapes
should be greatly considered in the construction of policy incentives. This would instill a
statewide focus on optimizing land use for both renewable energy development and
habitat conservation initiatives.
126
5.5 Conclusions
This study was completed as a first attempt to understand the levels of landscape
conflict between wind and solar renewable energy resources and habitat conservation
priorities, according to a landscape-level perspective in Washington State. Previous
studies have largely been restricted to more localized assessments of these topics. This
study is unique in its use of the recently published crucial habitat assessment data, which
have provided the unparalleled opportunity to employ a statewide, landscape-level
analysis. While local levels of assessment are useful for understanding the particular
characteristics of a specific area, a landscape-level assessment is necessary to provide
important insight into the interactions, risks, and opportunities required to optimize land
use planning among multiple landscape functions. With this information, land use
managers, conservation biologists, and energy developers will have greater insight into
the landscape opportunities and conflicts between these two initiatives. This will
encourage future landscape design and management from a multifunctional landscape
approach, in which the priorities of habitat conservation and renewable energy
development are more balanced across the landscape.
Because of the specific scope and types of spatial analysis utilized, the results of
these analyses can only be interpreted within the context of Washington State. However,
this study does provide a methodology for future analysis in different study areas, and
demonstrates an effective way in which areas of significant landscape conflict can be
identified and explored. Regarding the analysis of Washington State, a number of
important outcomes were identified, leading to a lengthy discussion around the
127
opportunity to use crucial habitat as an indicator to optimize land use and move toward
the creation of multifunctional landscapes.
In the context of existing Washington wind farms, the distribution of crucial
habitat was found to vary from one wind farm to another. This provided an opportunity to
analyze the approved environmental impact assessments for the worst sited wind farms
according to crucial habitat distribution. It was found that the rigor of environmental
assessment varied greatly between the two wind farms. However, despite being located in
most-crucial habitat lands, if assessments are thoroughly completed and mitigation plans
are carefully constructed, the environmental impacts of development can be minimized.
Additionally, wind farm projects can contribute to high-level conservation goals by
designating natural lands as preserves and establishing habitat corridors connecting
multiple parcels of natural habitat. This analysis demonstrated the potential to use EIAs
as a springboard for thinking about development within the context of multifunctional
landscapes.
Analysis of suitable wind and solar development locations and the estimate of
future energy contributions found that solar energy development could specifically target
the least-crucial habitat lands. Within these landscapes, solar energy could provide over
50% of total existing state energy production if fully developed. On the other hand, wind
energy development could only be restricted to crucial habitat lands 3–6 and still be able
to quadruple in size, contributing another 7.9% of annual state energy needs. Since there
were some limitations to the solar energy estimate, future analyses—including a more
realistic feasibility study—should be conducted. Preferred renewable energy
development locations could be identified for the assessment according to economic,
128
social, and technological impacts to renewable energy development. This more realistic
model for determining preferred renewable energy development locations would help
land use planners better understand and anticipate land use conflicts between the two
initiatives and optimize land use to the benefit of both.
Finally, the analysis of significant spatial clustering and landscape conflicts
within general Washington habitats showed that the grassland and shrubland habitats, as
well as the agricultural, pasture, and mixed-environment habitats, were impacted the
most. Grassland and shrubland habitats had high energy resources for both wind and
solar, which would make renewable energy development in this habitat highly likely.
However, the solar analysis also showed a significant presence of high conflict clusters.
Since many of the grassland and shrubland habitats are priority 1 conservation habitats
and already threatened by habitat loss, habitat fragmentation, and invasive species,
development should occur cautiously, with a thorough EIA of the proposed development
locations.
The agriculture, pasture, and mixed-environment habitat also had high solar
energy resources, as well as high low conflict areas. These results identified the best
opportunity for solar development, since the agriculture, pasture, and mixed-environment
habitat is not a priority conservation habitat. Additionally, many of the typical
environmental impacts associated with renewable energy development could be
minimized. Existing road networks could be utilized, reducing further habitat
fragmentation and, since this habitat is already a product of landscape conversion, siting
renewable energy development there would reduce the impact of habitat loss, especially
when low conflict areas are targeted. However, since agricultural lands already have a
129
dominant land use function, challenges of strategic multifunctional landscape design and
stakeholder accommodation may be encountered. Despite the additional challenges, solar
development in agriculture, pasture, and mixed-environments presents an outstanding
interdisciplinary opportunity to pilot the construction and management of multifunctional
energyscapes. In this case, land use would be balanced between the environment, energy,
and agriculture.
Overall, the findings of this study have demonstrated the great potential that
landscape-level assessments have in identifying how multiple land use interests can
interact across the landscape. By specifically identifying the levels of conflict between
renewable energy development and habitat conservation priorities at a landscape-level
perspective, the first steps to initiating a multifunctional landscape approach to landscape
design and management have been achieved. If applied to future landscape planning,
lands that would generate energy while also having a low conservation value would be
first identified at a landscape-level and targeted for future energy development. This
would avoid lands having more severe landscape conflicts between renewable energy
development and habitat conservation priorities. Then, the existing EIA process for
development would identify any lesser land use conflicts at a local scale and propose
specific mitigation plans to further optimize the landscape design between these two
initiatives.
While there will be many challenges associated with the design and application of
multifunctional landscapes, this approach to landscape management is becoming critical
for the future. More studies should be conducted with a landscape-level perspective, so
that land use planning can fully optimize the land use requirements of multiple functions
130
across the landscape. This could include similar studies in other states, smaller regional
studies within Washington, and studies that explore different landscape contexts. These
future studies will identify more opportunities and challenges of optimizing the landscape
design between habitat conservation and renewable energy development and further
encourage a multifunctional landscape approach to landscape planning and management.
These efforts will provide the foundation to better balance the needs of society and the
environment for the future across all types of land uses and provide a healthier Earth for
all inhabitants.
131
Works Cited
American Wind Energy Association, Tetra Tech EC Inc., and Nixon Peabody LLP. 2008.
“Wind Energy Siting Handbook”. American Wind Energy Association.
http://www.awea.org/Issues/Content.aspx?ItemNumber=5726.
Anselin, Luc. 1992. “Spatial Data Analysis with GIS - An Introduction to Application in
the Social Sciences”. Technical Report 92-10. Santa Barbara, California: National
Center for Geographic Information and Analysis - UCSB.
Arnett, Edward B., W. Kent Brown, Wallace P. Erickson, Jenny K. Fiedler, Brenda L.
Hamilton, Travis H. Henry, Aaftab Jain, et al. 2008. “Patterns of Bat Fatalities at
Wind Energy Facilities in North America.” Journal of Wildlife Management 72
(1): 61–78. doi:10.2193/2007-221.
Athanas, Andrea K., and Nadine McCormick. 2013. “Clean Energy That Safeguards
Ecosystems and Livelihoods: Integrated Assessments to Unleash Full Sustainable
Potential for Renewable Energy.” Renewable Energy 49: 25–28.
Barclay, Robert M.R., E.F. Baerwald, and J.C. Gruver. 2007. “Variation in Bat and Bird
Fatalities at Wind Energy Facilities: Assessing the Effects of Rotor Size and
Tower Height.” Canadian Journal of Zoology 85 (3): 381–87. doi:10.1139/Z07011.
Bellard, Celine, Cleo Bertelsmeier, Paul Leadley, Wilfried Thuiller, and Frank
Courchamp. 2012. “Impacts of Climate Change on the Future of Biodiversity.”
Ecology Letters 15 (4): 365–77.
Bennett, Andrew F., and Denis A. Saunders. 2010. “Habitat Fragmentation and
Landscape Change.” In Conservation Biology for All, edited by Navjot S. Sodhi
and Paul R. Ehrlich, 88–107. New York: Oxford University Press.
Blyth, Ann, Dave Cake, Ian Laing, Martin Andresen, Dr. Gennady Gienko, and Michael
Govorov. 2007. “Spatial Analysis and Modeling: (GII-07) Training Material”.
Training Material for GIS Managers BPD2004-ESF-2.2.0.-02-05/0143. Powell
River, British Columbia: Vancouver Island University.
Bolstad, Paul. 2012. GIS Fundamentals: A First Text on Geographic Information
Systems. Fourth Edition. White Bear Lake, Minnesota: Eider Press.
Burger, Joanna, and Michael Gochfeld. 2012. “A Conceptual Framework Evaluating
Ecological Footprints and Monitoring Renewable Energy: Wind, Solar, Hydro,
and Geothermal.” Energy & Power Engineering 4 (4): 303–14.
doi:10.4236/epe.2012.44040.
Campos, Paulo R. A., Elias D. C. Neto, Viviane M. de Oliveira, and M. a. F. Gomes.
2012. “Neutral Communities in Fragmented Landscapes.” Oikos 121 (11): 1737–
48. doi:10.1111/j.1600-0706.2011.20336.x.
Campos, Paulo R. A., Alexandre Rosas, Viviane M. de Oliveira, and Marcelo A. F.
Gomes. 2013. “Effect of Landscape Structure on Species Diversity.” PLoS ONE 8
(6): 1–10. doi:10.1371/journal.pone.0066495.
Chappell, Christopher B., Rex C. Crawford, Charley Barrett, Jimmy Kagan, David H.
Johnson, Mikell O’Mealy, Greg A. Green, Howard L. Ferguson, W. Daniel Edge,
and Eva L. Greda. 2001. “Wildlife Habitats: Descriptions, Status, Trends, and
System Dynamics.” Wildlife Habitat Relationships in Oregon and Washington.
http://www.cbfwa.org/Committees/WAC/meetings/2010_0622/chapter2cwb.pdf.
132
Chapter 19.285 RCW: ENERGY INDEPENDENCE ACT. 2006. RCW.
http://apps.leg.wa.gov/rcw/default.aspx?cite=19.285&full=true.
Cowardin, Lewis M., Virginia Carter, Francis C. Golet, and Edward T. LaRoe. 1979.
“Classification of Wetlands and Deepwater Habitats of the United States”.
Library of Congress FWS/OBS-79/31. Washington D.C.: U.S. Department of
Interior, Fish and Wildlife Services.
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.364.379&rep=rep1&ty
pe=pdf.
Demirbas, A. 2009. “Global Renewable Energy Projections.” Energy Sources, Part B:
Economics, Planning, and Policy 4: 212–24.
DESIRE. 2013. “Washington Renewable Energy Standard”. Government. July 19.
http://dsireusa.org/incentives/incentive.cfm?Incentive_Code=WA15R&re=1&ee=
1.
Diffendorfer, Jay E., Roger Compton, Louisa Kramer, Zach Ancona, and Donna Norton.
2014. “Onshore Industrial Wind Turbine Locations for the United States through
July 2013: USGS Data Series 817”. Science Organization. Onshore Industrial
Wind Turbine Locations for the United States through July 2013. February.
http://pubs.usgs.gov/ds/817/.
ed. N. S. Sodhi, and Paul R. Ehrlich, eds. 2010. Conservation Biology for All. New York:
Oxford University Press.
Elliott, D. L., C. G. Holladay, W. R. Barchet, H. P. Foote, and W. F. Sandusky. 1986.
“Wind Energy Resource Atlas of the United States”. Government Document
DE86004442. Richland, Washington: National Renewable Energy Laboratory.
http://rredc.nrel.gov/wind/pubs/atlas/atlas_index.html.
Energy Facility Site Evaluation Council. 2014. “Energy Facility Site Evaluation
Council”. Government Site. Energy Facility Site Evaluation Council.
http://www.efsec.wa.gov/default.shtm.
Esri Inc. 2013. ArcGIS 10.2 for Desktop (version 10.2.0.3348). Redlands, CA:
Environmental Systems Research Institute, Inc. (Esri),.
Fetene, Aramde, Kumlachew Yeshitela, and Hayal Desta. 2012. “Approaches to
Conservation and Sustainable Use of Biodiversity- A Review.” Nature & Science
10 (12): 51–62.
Fotheringham, Stewart A., Chris Brunsdon, and Martin Charlton. 2000. Quantitative
Geography: Perspectives on Spatial Data Analysis. 1st Ed. Thousand Oaks, CA:
Sage Publications Ltd.
Gelbard, Jonathan L., and Jayne Belnap. 2003. “Roads as Conduits for Exotic Plant
Invasions in a Semiarid Landscape.” Conservation Biology 17 (2): 420 – 432.
Goble, Dale D., John A. Wiens, J. Michael Scott, Timothy D. Male, and John A. Hall.
2012. “Conservation-Reliant Species.” BioScience 62 (10): 869–73.
doi:10.1525/bio.2012.62.10.6.
Gotelli, Nicholas J. 2001. A Primer of Ecology. 3rd ed. Saunderland, MA: Sinauer
Associates, Inc.
Groot, Jeroen C.J., André Jellema, and Walter A.H. Rossing. 2010. “Designing a
Hedgerow Network in a Multifunctional Agricultural Landscape: Balancing
Trade-Offs among Ecological Quality, Landscape Character and Implementation
133
Costs.” European Journal of Agronomy 32 (1): 112–19.
doi:10.1016/j.eja.2009.07.002.
Hamerlinck, Jeff, and Michael Terner. 2013. “Western Governors’ Crucial Habitat
Assessment Tool: Mapping Fish & Wildlife Across the West”. Conference
presented at the NSGIC Annual Conference, Kansas City, MO, October 28.
http://www.nsgic.org/public_resources/AC13_Mon_10303_Hamerlinck_CoordinationCollab.pdf.
Hanski, Ilkka. 2011. “Habitat Loss, the Dynamics of Biodiversity, and a Perspective on
Conservation.” AMBIO - A Journal of the Human Environment 40 (3): 248–55.
doi:10.1007/s13280-011-0147-3.
Harden, Noelle M., Loka L. Ashwood, William L. Bland, and Michael M. Bell. 2013.
“For the Public Good: Weaving a Multifunctional Landscape in the Corn Belt.”
Agriculture & Human Values 30 (4): 525–37. doi:10.1007/s10460-013-9429-7.
Howard, D.C., P.J. Burgess, S.J. Butler, S.J. Carver, T. Cockerill, A.M. Coleby, G. Gan,
et al. 2013. “Energyscapes: Linking the Energy System and Ecosystem Services
in Real Landscapes.” Biomass and Bioenergy 55 (August): 17–26.
doi:10.1016/j.biombioe.2012.05.025.
Hunt, Bruce, Leonard Bauer, Dee Caputo, Dave Anderson, Roger Weed, and Karen
Larkin. 2012. “Urban Growth Area Guidebook: Reviewing, Updating and
Implementing Your Urban Growth Area”. Government Document. Olympia, WA:
Washington Department of Commerce.
http://www.commerce.wa.gov/Documents/GMS-UGA-Guidebook-Final2012.pdf.
Intergovernmental Panel on Climate Change. 2007. “Climate Change 2007: Synthesis
Report”. Synthesis Report Fourth Assessment.
———. 2012. “Renewable Energy Sources and Climate Change Mitigation: Special
Report of the Intergovernmental Panel on Climate Change”. Full report. Potsdam
Institute for Climate Impact Research.
———. 2014. “IPCC WGIII Summary for Policy Makers”. Summary Report IPCC
WGIII. Potsdam Institute for Climate Impact Research.
http://report.mitigation2014.org/spm/ipcc_wg3_ar5_summary-forpolicymakers_approved.pdf.
International Energy Agency. 2012. “Tracking Clean Energy Progress: Energy
Technology Perspectives 2012 Excerpt as IEA Input to the Clean Energy
Ministerial”. International Energy Agency.
http://www.iea.org/publications/freepublications/publication/Tracking_Clean_Ene
rgy_Progress.pdf.
Jay, Stephen. 2010. “Strategic Environmental Assessment for Energy Production.”
Energy Policy 38 (7): 3489–97. doi:10.1016/j.enpol.2010.02.022.
Johnson, David H., and T. A. O’Neil. 2001. Wildlife-Habitat Relationships in Oregon
and Washington. ArcGIS Data (version 1). Corvallis, OR: Oregon State
University Press.
Jones & Stokes, Wind Ridge Power Partners LLC, West Inc., CH2M HILL, Lithic
Analysts, RAM Associates, Nierenberg R., Consulting Meteorologists,
Comsearch, and KTA Associates. 2004. “EFSEC - Wild Horse Wind Power
Project Draft Environmental Impacts Statement”. Draft EIS. Olympia, WA: Wind
134
Ridge Power Partners LLC.
http://www.efsec.wa.gov/wildhorse/deis/whdeis.shtm#deis.
Kiilsgaard, Chris. 1999. “Washington Wildlife Habitat Types: Mapping and
Classification of Landscape-Level Habitats”. NWHI Habitat Program Report for
WDFW. Corvallis, OR: NWHI.
http://www.nwhi.org/nhi/gisdata/docs/washington/wacurhab.pdf.
Kirchhoff, Thomas, Ludwig Trepl, and Vera Vicenzotti. 2013. “What Is Landscape
Ecology? An Analysis and Evaluation of Six Different Conceptions.” Landscape
Research 38 (1): 33–51.
Kittias County Community Development Services. 2008. “Notice of SEPA Action/Public
Hearing - Vantage Wind Power Project”. State Environmental Protection
Assessment No. WSA-07-01. Ellensburg, WA: Invenergy Wind North America,
LLC. https://www.co.kittitas.wa.us/%2Fuploads%2Fcds%2Flanduse%2FWind%20Farm%2FWSA-0701%20Vantage%20Wind%20Power%20Project%20Notices_Staff%20Report/22608%20SEPA%20MDNS%20Vantage%20Wind%20Power%20Project%20WSA07-01.pdf.
Kuvlesky Jr., William P., Leonard A. Brennan, Michael L. Morrison, Kathy K. Boydston,
Bart M. Ballard, and Fred C. Bryant. 2007. “Wind Energy Development and
Wildlife Conservation: Challenges and Opportunities.” Journal of Wildlife
Management 71 (8): 2487–98. doi:10.2193/2007-248.
Lackey, Robert T. 2007. “Science, Scientists, and Policy Advocacy.” Conservation
Biology 21 (1): 12–17. doi:10.1111/j.1523-1739.2006.00639.x.
Laurance, William F. 2010. “Habitat Destruction: Death by a Thousand Cuts.” In
Conservation Biology for All, edited by Navjot S. Sodhi and Paul R. Ehrlich, 73–
87. New York: Oxford University Press.
Lloyd, Michael W., Lesley Campbell, and Maile C. Neel. 2013. “The Power to Detect
Recent Fragmentation Events Using Genetic Differentiation Methods.” PLoS
ONE 8 (5): 1–16. doi:10.1371/journal.pone.0063981.
Lopez, Anthony, Billy Roberts, Donna Heimiller, Nate Blair, and Gian Porro. 2012.
“U.S. Renewable Energy Technical Potentials: A GIS-Based Analysis”.
Government Document. National Renewable Energy Lab.
http://www.nrel.gov/docs/fy12osti/51946.pdf.
Lovell, Sarah Taylor, and Douglas M. Johnston. 2009. “Creating Multifunctional
Landscapes: How Can the Field of Ecology Inform the Design of the Landscape?”
Frontiers in Ecology and the Environment 7 (4): 212–20.
Lu, Nan, Chen-Xi Jia, Huw Lloyd, and Yue-Hua Sun. 2012. “Species-Specific Habitat
Fragmentation Assessment, Considering the Ecological Niche Requirements and
Dispersal Capability.” Biological Conservation 152 (August): 102–9.
doi:10.1016/j.biocon.2012.04.004.
Lubowski, Ruben N., Andrew J. Plantinga, and Robert N. Stavins. 2008. “What Drives
Land-Use Change in the United States? A National Analysis of Landowner
Decisions.” Land Economics 84 (4): 529–50.
MacArthur, Robert H., and Edward O. Wilson. 1967. The Theory of Island
Biogeography. Princeton, New Jersey: Princeton University Press.
135
Manwell, J.F., J.G. McGowan, and A.L. Rogers. 2009. Wind Energy Explained: Theory,
Design and Application. 2nd Ed. West Sussex, United Kingdom: John Wiley &
Sons Ltd.
McDonald, Robert I, Joseph Fargione, Joe Kiesecker, William M Miller, and Jimmie
Powell. 2009. “Energy Sprawl or Energy Efficiency: Climate Policy Impacts on
Natural Habitat for the United States of America.” Plos One 4 (8): e6802–e6802.
19707570. doi:10.1371/journal.pone.0006802.
Meine, Curt. 2010. “Conservation Biology: Past and Present.” In Conservation Biology
for All, edited by Navjot S. Sodhi and Paul R. Ehrlich, 7–26. New York: Oxford
University Press.
Mueller, Thomas, Kirk A. Olson, Gunnar Dressler, Peter Leimgruber, Todd K. Fuller,
Craig Nicolson, Andres J. Novaro, et al. 2011. “How Landscape Dynamics Link
Individual- to Population-Level Movement Patterns: A Multispecies Comparison
of Ungulate Relocation Data.” Global Ecology & Biogeography 20 (5): 683–94.
doi:10.1111/j.1466-8238.2010.00638.x.
National Park Service. 2011. “About Us: National Register of Historic Places”.
Government Site. National Register of Historic Places Program: About Us. June.
http://www.nps.gov/nr/about.htm.
———. 2012. “National Register of Historic Places Database and Research Page”.
Government Site. National Register of Historic Places Program: Research - Data
Downloads. http://www.nps.gov/nr/research/data_downloads.htm.
National Renewable Energy Laboratory. 2009. Solar Prospector. GIS (version v2
[2012]). Boulder, CO: National Renewable Energy Laboratory.
http://mapsdb.nrel.gov
———. 2013a. “NREL: Dynamic Maps, GIS Data, and Analysis Tools - Solar Maps”.
Government Site. Solar Data. September. http://www.nrel.gov/gis/solar.html.
———. 2013b. “NREL: Geographic Information System Data Background”.
Government Site. Geographic Information System Data Background. September.
http://www.nrel.gov/gis/data_analysis_background.html.
———. 2013c. “NREL Solar PV Tilts Equals Latitude GIS Data Metadata”. Government
Site. Metadata (Solar PV). September 26.
http://www.nrel.gov/gis/data/GIS_Data_Technology_Specific/United_States/Sola
r/metadata/us_latilt.shp.xml.
———. 2014. “NREL: Dynamic Maps, GIS Data, and Analysis Tools - Wind Maps”.
Government Site. Wind Data. February. http://www.nrel.gov/gis/wind.html.
Nelson, V. 2009. Wind Energy: Renewable Energy and the Environment. 1st ed. Boca
Raton, FL.: Taylor & Francis Group.
Northrup, Joseph M., and George Wittemyer. 2013. “Characterising the Impacts of
Emerging Energy Development on Wildlife, with an Eye towards Mitigation.”
Ecology Letters 16: 112–25.
Northwest Habitat Institute. 2011. “Northwest Habitat Institute: Mission Statement”.
Non-Profit Scientific and Educational. NWHI Mission Statement.
http://www.nwhi.org/index/.
Northwest Power and Conservation Council. 2013. “Power Supply: Adequacy and
Reliability”. State Energy Council. NW Power and Conservation Council: Energy
- Power Supply. July. http://www.nwcouncil.org/energy/powersupply/.
136
Ong, Sean, Clinton Campbell, Paul Denholm, Robert Margolis, and Garvin Heath. 2013.
“Land-Use Requirements for Solar Power Plants in the United States”. NREL
Technical Document NREL/TP-6A20-56290. Golden, Colorado: National
Renewable Energy Laboratory. http://www.nrel.gov/docs/fy13osti/56290.pdf.
Park, Kirsty J., Alex Turner, and Jeroen Minderman. 2013. “Integrating Applied Ecology
Planning Policy: The Case of Micro-Turbines and Wildlife Conservation.”
Journal of Applied Ecology 50: 199–204.
Perez, Richard, Pierre Ineichen, Kathy Moore, Marek Kmiecik, Cyril Chain, Ray George,
and Frank Vignola. 2002. “A New Operational Model for Satellite-Derived
Irradiances: Description and Validation.” Solar Energy 73 (5): 307–17.
doi:10.1016/S0038-092X(02)00122-6.
Pimm, Stewart L., and Clinton N. Jenkins. 2010. “Extinctions and the Practice of
Preventing Them.” In Conservation Biology for All, edited by Navjot S. Sodhi
and Paul R. Ehrlich, 181–98. New York: Oxford University Press.
Protected Areas Database-US Partnership. 2009. “A Map for the Future: Creating the
Next Generation of Protected Area Inventories in the United States”. Government
Document. Idaho: US Geological Survey (USGS).
http://gapanalysis.usgs.gov/padus/files/2012/09/PADUS_FinalJuly2009LowRes.p
df.
Reyers, Belinda, Patrick J. O’Farrell, Jeanne L. Nel, and Kerrie Wilson. 2012.
“Expanding the Conservation Toolbox: Conservation Planning of Multifunctional
Landscapes.” Landscape Ecology 27 (8): 1121–34.
Rueda, Marta, Bradford A. Hawkins, Ignacio Morales-Castilla, Rosa M. Vidanes, Mila
Ferrero, and Miguel Á Rodríguez. 2013. “Does Fragmentation Increase Extinction
Thresholds? A European-Wide Test with Seven Forest Birds.” Global Ecology
and Biogeography 22 (12): 1282–92.
Schaub, Michael, and Fitsum Abadi. 2011. “Integrated Population Models: A Novel
Analysis Framework for Deeper Insights into Population Dynamics.” Journal of
Ornithology 152 (July): 227–37. doi:10.1007/s10336-010-0632-7.
Shirk, A. J., D. O. Wallin, S. A. Cushman, C. G. Rice, and K. I. Warheit. 2010. “Inferring
Landscape Effects on Gene Flow: A New Model Selection Framework.”
Molecular Ecology 19 (17): 3603–19. doi:10.1111/j.1365-294X.2010.04745.x.
The World Bank. 2014. “About Us - Renewable Energy Database - The World Bank”.
International Financial Institution. The World Bank - About the Renewable Energy
Database. http://ppi-re.worldbank.org/About-Us.
Trombulak, Stephen C., and Christopher A. Frissell. 2000. “Review of Ecological Effects
of Roads on Terrestrial and Aquatic Communities.” Conservation Biology 14 (1):
18–30. doi:10.1046/j.1523-1739.2000.99084.x.
Trombulak, Stephen C., Kristian S. Omland, Julie A. Robinson, Jeffrey J. Lusk, Thomas
L. Fleischner, Glenn Brown, and Meg Domroese. 2004. “Principles of
Conservation Biology: Recommended Guidelines for Conservation Literacy from
the Education Committee of the Society for Conservation Biology.” Conservation
Biology 18 (5). Conservation Education: 1180–90.
US Department of Agriculture, National Agriculture Statistics Service, Research and
Development Division, Geospatial Information Branch, and Spatial Analysis
Research Section. 2014. USDA, 2013 Washington Cropland Data Layer. ArcGIS
137
(version 2013 Growing Season). Washington D.C.: USDA, NASS.
http://www.nass.usda.gov/research/Cropland/metadata/metadata_wa13.htm.
US Department of Energy. 2010. “2009 US State Clean Energy Data Book”. Government
Document. National Renewable Energy Lab.
http://www.nrel.gov/tech_deployment/state_local_activities/pdfs/48212.pdf.
———. 2011. “2011 Renewable Energy Data Book”. Government Document. National
Renewable Energy Lab. http://www.nrel.gov/docs/fy13osti/54909.pdf.
———. 2013. “DSIRE (Database of State Incentives for Renewables and Efficiency)”.
National Database. DSIRE USA. Accessed October 22.
http://dsireusa.org/incentives/incentive.cfm.
US Energy Information Administration. 2012. “Washington Renewable Electricity State
Profiles”. Government Site. State Renewable Electricity Profiles.
http://www.eia.gov/renewable/state/Washington/.
US Fish & Wildlife Service. 2013. “U.S. Fish and Wildlife Service”. Government Site.
About the U.S. Fish and Wildlife Service. December.
http://www.fws.gov/help/about_us.html.
———. 2014. “National Wetlands Inventory: Limitations, Exclusions and Precautions”.
Government Site. Data Limitations, Exclusions, and Precautions. February.
http://www.fws.gov/wetlands/Data/Limitations.html.
US Fish and Wildlife Service, National Standards and Support Team. 2012. National
Wetlands Inventory Metadata. ArcGIS (version 1.0). Washington D.C.: U.S. Fish
and Wildlife Services.
http://www.fws.gov/wetlands/Data/metadata/FWS_Wetlands.xml.
US Geological Survey. 2013. “About United States Geological Survey”. Science
Organization. About USGS. June. http://www.usgs.gov/aboutusgs/.
US Geological Survey, Gap Analysis Program. 2012. Protected Areas Database of the
United States (PADUS). GAP (version 1.3 Combined Feature Class). Idaho:
USGS.
———. 2013. “Standards and Methods Manual for State Data Stewards: PAD-US”.
University of Idaho.
ftp://ftp.gap.uidaho.edu/outgoing/PADUS_download/PADUS_Standards.pdf.
Vandergast, Amy G., Richard D. Inman, Kelly R. Barr, Kenneth E. Nussear, Todd C.
Esque, Stacie A. Hathaway, Dustin A. Wood, et al. 2013. “Evolutionary Hotspots
in the Mojave Desert.” Diversity (14242818) 5 (2): 293–319.
doi:10.3390/d5020293.
Vejre, Henrik, Jens Abildtrup, Niels Kærgaard, Bo Fritzbøger, Anne Gravsholt Busck,
and Søren Bøye Olsen. 2012. “Revitalisation of Common Use in Management of
Modern Multifunctional Landscapes.” Landscape Research 37 (6): 637–57.
doi:10.1080/01426397.2012.705821.
Villard, Marc-André, and Jean Paul Metzger. 2014. “REVIEW: Beyond the
Fragmentation Debate: A Conceptual Model to Predict When Habitat
Configuration Really Matters.” Journal of Applied Ecology 51 (2): 309–18.
doi:10.1111/1365-2664.12190.
WA Department of Ecology. 2011. “Washington State Cities and Urban Growth Area
Boundaries Metadata”. Government Site. Washington State City/UGA Areas
Metadata. April. http://www.ecy.wa.gov/services/gis/data/polsub/cityuga.htm.
138
Waltner-Toews, David, James J. Kay, and Nina-Marie E. Lister. 2008. The Ecosystem
Approach - Complexity, Uncertainty, and Managing for Sustainability. 1st ed.
Chichester, West Sussex New York: Columbia University Press.
Washington Department of Ecology. 2014. “About Us | Washington State Department of
Ecology”. Government Site. About Us. http://www.ecy.wa.gov/about.html.
Washington Department of Fish and Wildlife. 2005. “Washington’s Comprehensive
Wildlife Conservation Strategy”. Olympia, WA.
Washington State. 1971. Chapter 43-21 RCW STATE ENVIRONMENTAL POLICY.
RCW. http://www.ecy.wa.gov/programs/sea/sepa/43_21C.htm#z030.
Washington State Department of Agriculture. 2014. “About Washington State
Department of Agriculture (WSDA)”. Government Site. About WSDA.
http://agr.wa.gov/AboutWSDA/.
Western Governors’ Association. 2013. “Western Governors’ Crucial Habitat
Assessment Tool Westwide Metadata”. Western Governors’ Association.
http://www.westgovchat.org/data/metadata.
Western Governors’ Wildlife Council. 2013a. “Western Governors’ Wildlife Council
White Paper - Version III”. White Paper. Denver, CO: Western Governors’
Association.
———. 2013b. “Western Wildlife Crucial Habitat Assessment Tool (CHAT): Vision,
Definitions and Guidance for State Systems and Regional Viewer.”
http://www.westgov.org/initiatives/wildlife.
White, Gordon, Neil Aaland, Healther Ballash, Patty Betts, Rebecca Inman, Elizabeth
Phinney, Peter Riley, Barbara Ritchie, and Marv Vialle. 2003. “State
Environmental Policy Act Handbook”. Government Document 98-114. Olympia,
WA: Shorelands and Environmental Assistance Program, Washington State
Department of Ecology.
https://fortress.wa.gov/ecy/publications/publications/98114.pdf.
139
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Windy Point/Windy Flats
Lower Snake River
Big Horn
Marengo Wind
White Creek
Palouse Wind Farm
Juniper Canyon
Wild Horse
Hopkins Ridge
Stateline Wind
Nine Canyon
Vantage Wind
Kittitas Valley Wind
Goodnoe Hills
Harvest Wind
Linden Ranch
Coyote Crest Project
Lower Davenport
Coastal Energy Wind
Swauk Valley Ranch…
Appendix
Crucial
Habitat
Rank
1
2
3
4
5
Figure 19. Crucial habitat on each existing Washington wind farm
140
Table 9. Crucial Habitat Ranking Factors
(Source: Crucial Habitat Assessment Tool metadata prioritization worksheet (Western Governors’ Association 2013))
Ranking
Factors
Aquatic
Species of
Concern
Terrestrial
Species of
Concern
Natural
Vegetation
Communities
1
Endangered and
threatened species
spawning area
(documented).
Confirmed
locations for
threatened and
endangered plant
and animal species;
level 1 priority
habitat species.
Ecological systems
of concern with
level 1 priority
habitat and
confirmed heritage
vegetation
locations.
2
Endangered and
threatened species
documented,
presumed,
transported, or
artificial presence.
Confirmed locations
for level 2 priority
habitat species;
federal and state
candidate and
sensitive species.
Ecological systems of
concern with level 2
priority habitat and
confirmed heritage
vegetation locations;
federal and state
candidate and
sensitive species.
Crucial Habitat Rank
3
4
Federal species of
Endangered and
concern documented or
threatened
presumed presence; state species potential
candidate species of
presence.
concern documented or
presumed presence.
Confirmed locations for Level 3 modeled
level 3 priority habitat
species of
species; level 2 modeled concern
species of concern
locations.
locations.
Presence of at least one
global ranked G1 or G2
ecological systems of
concern with level 3
priority habitat and
confirmed heritage
vegetation locations.
5
Endangered
and threatened
species
potential
presence.
6
None of the
aforementioned
factors
applies.
Level 4
modeled
species of
concern
locations.
None of the
aforementioned
factors
applies.
Presence of at
least 1 global
ranked G3
ecological
systems of
concern.
None of the
aforementioned
factors
applies.
141
Ranking
Factors
1
High integrity
estuaries.
2
Moderate integrity
estuaries.
Wetland and
Riparian
Areas
Aquatic
Species of
Economic &
Recreational
Importance
Freshwater
Integrity
Large
Natural
Areas
142
Terrestrial
Species of
Economic &
Recreational
Importance
Other salmonid
spawning areas
(documented).
Crucial Habitat Rank
3
4
Low integrity estuaries;
Presence of good
priority species wetland condition flood
and riparian habitats;
plain.
National Wetland
Inventory; presence of
excellent condition flood
plain.
Other salmonid
Non-native game
documented or
fish presence.
presumed presence;
white sturgeon presence.
5
Presence of fair
condition flood
plain.
6
None of the
aforementioned
factors
applies.
Other salmonid
potential
presence.
None of the
aforementioned
factors
applies.
None of the
aforementioned
factors
applies.
None of the
aforementioned
factors
applies.
Catchment in relatively
excellent condition.
Catchment in
relatively good
condition.
Catchment in
relatively fair
condition.
Large intact blocks
greater than 1000Ha and
with best (top 33%)
integrity.
Large intact
blocks less than
10,000Ha or
greater than
50,000Ha and
with integrity less
than top 33%
(best).
Level 2 priority
game habitats
and concentration
areas.
Large intact
blocks greater
than 10,000Ha
or less than
50,000Ha and
with integrity
less than top
33% (best).
Level 3 priority
game habitats
and
concentration
areas.
Level 1 priority game
habitats and
concentration areas.
None of the
aforementioned
factors
applies.
Ranking
Factors
Landscape
Connectivity
Wildlife
Corridors
1
2
Crucial Habitat Rank
3
4
Connectivity
zones with score
of 1.
WA wildlife
habitat
connectivity
modeled network
with overlap of at
least three focal
species.
5
Connectivity
zones with
score of 2.
6
None of the
aforementioned
factors
applies.
None of the
aforementioned
factors
applies.
143
144
Figure 20. Spatial interaction of wind energy resources and crucial habitat in Washington agriculture,
pasture, and mixed-environment habitats
145
Figure 21. Spatial interaction of wind energy resources and crucial habitat in Washington
urban and mixed-environment habitats
146
Figure 22. Spatial interaction of wind energy resources and crucial habitat in Washington wetland,
rivers, lakes, and reservoir habitats
147
Figure 23. Spatial interaction of solar energy resources and crucial habitat in Washington forest
and woodland habitats
148
Figure 24. Spatial interaction of solar energy resources and crucial habitat in Washington urban
and mixed-environment habitats
149
Figure 25. Spatial interaction of solar energy resources and crucial habitat in Washington wetland,
rivers, lakes, and reservoir habitats
Keese_KMESthesis2014.pdf