-
Title
-
Eng
A Spatial Analysis for Ecologically Conscious Wind Farm Siting in the Pacific Northwest
-
Date
-
2017
-
Creator
-
Eng
Moore, Emily Evangeline
-
Subject
-
Eng
Environmental Studies
-
extracted text
-
A SPATIAL ANALYSIS FOR ECOLOGICALLY CONSCIOUS WIND FARM SITING
IN THE PACIFIC NORTHWEST
by
Emily Evangeline Moore
A Thesis
Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
June, 2017
©2017 by Emily Evangeline Moore. All rights reserved.
This Thesis for the Master of Environmental Studies Degree
by
Emily Evangeline Moore
has been approved for
The Evergreen State College
by
________________________
Kathleen Saul, Ph.D.
________________________
June 12th, 2017
ABSTRACT
A Spatial Analysis for Ecologically Conscious Wind Farm Siting
in the Pacific Northwest
Emily Evangeline Moore
As renewable energy resources are increasing availability, there is evidence to
suggest that the renewable alternatives have numerous ecological impacts that should be
addressed before developers proceed to mass produce energy. Birds are particularly
vulnerable to experiencing population decline as a result of mortality from collision
with wind turbines and displacement from habitat in which the wind farms are
developed. The presently used mortality rate is based on rates at individual wind farms
rather than on a collective rate for wind farms along flyways. Additionally, the present
mortality rate does not account for population declines related to loss of viable habitat or
habitat connectivity during and after construction of wind farms, so the overall
population declines are grossly underestimated. The goal of this study is to locate sites
for new wind farms that will aid in the reduction of bird mortality resulting from wind
farm development in the United States. To accomplish this goal I will perform a spatial
analysis of the United States using Geographic Information Systems (GIS) as an analysis
platform. My analysis resulted in a preliminary siting tool that will address the
conservation of birds by avoiding the important bird areas and protected areas while
encouraging a multipurpose landscape. This tool can be used by the public, politicians
and developers to make informed decisions about wind farm siting to reduce the overall
ecological impacts birds while increasing the availability of renewable energy resources.
Contents
Introduction ..................................................................................................................................... 1
Literature Review ........................................................................................................................... 12
Development and Problem Recognition .................................................................................... 12
Wind Energy Technology Development ................................................................................. 12
Ecological Impact Research Exploration ................................................................................ 14
Impacts of Wind Turbines .......................................................................................................... 17
Displacement ......................................................................................................................... 18
Fatigue ................................................................................................................................... 22
Cumulative Mortality Impacts ............................................................................................... 22
Raptors and Seabirds ............................................................................................................. 24
Scavenger Removal and Carcass Searches............................................................................. 28
Mortality Estimation Accuracy ............................................................................................... 30
Science in Technology-Based Planning Approaches .............................................................. 31
Methods ......................................................................................................................................... 35
Goal/ Products ....................................................................................................................... 35
Data Collection & Resources .................................................................................................. 35
Data Manipulation: ArcGIS Desktop ...................................................................................... 36
Maps .............................................................................................................................................. 39
Results ............................................................................................................................................ 54
Existing Windfarm Sites ......................................................................................................... 54
Potential Windfarm Sites. ...................................................................................................... 54
Discussion ...................................................................................................................................... 58
Limitations ............................................................................................................................. 59
User Recommendations ......................................................................................................... 61
Conclusion ...................................................................................................................................... 63
Future Research ..................................................................................................................... 63
Final Remarks ......................................................................................................................... 65
Bibliography ................................................................................................................................... 67
iv
List of Figures
Figure 1: World Total Primary Energy Supply (TPES) from 1971 to 2014 by fuel (Mtoe). .............. 4
Figure 2: Share of Greenhouse Gas Emissions by Countries with Climate Targets……………………….5
Figure 3: Top 10 Cumulative Wind Energy Capacity, December 2015 ............................................ 8
Figure 4: Projected Non-Hydropower Renewable Energy Generation Increase ........................... 13
Figure 5: Pacific Northwest Site Classification and Important Bird Areas ..................................... 40
Figure 6: Pacific Northwest Site Classification ............................................................................... 41
Figure 7: Washington Site Classification and Important Bird Areas .............................................. 42
Figure 8: Washington Site Classification ........................................................................................ 43
Figure 9: Oregon Site Classification and Important Bird Areas ..................................................... 44
Figure 10: Oregon Site Classification ............................................................................................. 45
Figure 11: Idaho Site Classification and Important Bird Areas ...................................................... 46
Figure 12: Idaho Site Classification ................................................................................................ 47
Figure 13: Montana Site Classification and Important Bird Areas. ................................................ 48
Figure 14: Montana Site Classification........................................................................................... 49
Figure 15: Columbia River Existing Wind Farm and Important Bird Area Overlap ........................ 50
Figure 16: Wildhorse Existing Wind Farm and Important Bird Area Overlap. ............................... 51
Figure 17: Hot Springs Wind Exisiting Wind Farm and Important Bird Area Overlap. .................. 52
Figure 18: Rim Rock Exisitng Wind Farm and Important Bird Area Overlap.................................. 53
Figure 19: Small Important Bird Area Influence on Available Land Classificatios ......................... 56
Figure 20: Detailed Mapping for Areas of Interest ........................................................................ 57
v
List of Tables
Table 1: Area available for each of the classification types developed. ........................................ 55
vi
Acknowledgements
I would like to give a special thank you to my loving fiancé, David Larr IV, and our
beautiful dogs, Sasha and Nova for helping get through this process. Additional thanks to
my roommate and friend Dakota Burt, all of my friends in the Master of Environmental
Studies program, my family, and an honor mention to my thesis advisor Kathleen Saul.
These individuals always took the time to help me think through my work and ease my
worries.
vii
Introduction
Renewable energy technologies represent an opportunity for the United States, as
well as other countries, to reduce global impact on the environment. With early
indications and knowledge development on the ecological impacts that these new
technologies have, there is an additional opportunity to address these problems before
they become crises. However in the United States, "government and industry decision
making on energy questions tends to be event- or crisis-driven" (Heimann, 2004). For
example, the North American Electric Reliability Council (NERC) was created as a result
of the great Northeast blackout in 1965 despite a recognized need for a regulatory council
prior to this energy disaster (Heimann, 2004), This crisis-driven behavior occurs in
regards to climate change mitigation and ecological protection as well. For example, the
Clean Air Act and Endangered Species Act emerged long after the recognition of a threat
to human health and extreme losses of several species in the United States. However,
there are encouraging global actions being taken to reduce the impacts of global climate
change.
As climate change awareness and action increases, global efforts to reduce carbon
emissions by establishing initiatives and policies to introduce an increasing amount of
1
renewable energy for established and developing countries. The United Nations
Framework Convention on Climate Change was developed in 1992 “as a framework for
international cooperation to combat climate change by limiting average global
temperature increases and the resulting climate change” (United Nations, Background on
the UNFCCC). The Kyoto Protocol, adopted in 1997 as a legally binding contract for
developed countries to meet a target for reducing carbon emissions, established a target
for total reduction of greenhouse gases below the 1990 levels by 5 per cent (globally)
between 2008 and 2012 (United Nations, A Summary of the Kyoto Protocol). The actual
reduction targets for individual countries varied based on their present emissions rates.
Since then, new climate change action reforms have been called for.
More recently for the Paris International Conference on Climate Change Action of
2015, nations developed non-binding agreements to reduce carbon emissions based on
what they believe they are capable of achieving by 2020. This conference built upon the
Convention on Climate Change’s goal in an effort to ambitiously combat climate change
and adapt to its effects (United Nations, The Paris Agreement). The agreements will
accelerate the global carbon emission reductions to maintain global temperatures below 2
degrees Celsius greater than the pre-industrial global temperatures (Center for Climate
2
and Energy Solutions, 2015). While the Kyoto Protocol and the Paris Agreement are only
two of the many actions taken, they have provided a basis for directing the existing and
developing energy infrastructure toward renewable energy development across the world.
As the renewable energy sector expands, ideally this would mean that we would see
an overall decrease in non-renewable energy use. However, the alternate scenario
includes the expansion of fossil fuel based energy resources. As seen in Figure 1 below,
through 2014, there has yet to be a decrease in the use of fossil fuels but there is evidence
in the increased use of hydroelectric, biofuels, nuclear, and a category classified as
“other” which consists of geothermal, wind, solar, etc. (International Energy Agency,
2016). The figure is evidence of what I just indicated; the demand for energy continues
to rise. If the energy demand were to decrease or remain consistent, it would give
renewable energy technology and opportunity to grow and command the market.
3
Figure 1: Total world energy supply by energy type. (IEA, 2016 p.7)
In the years following the 2015 United Nations Climate Change Conference (COP21)
held in Paris, we should begin to see decreases in the demand for the non-renewable
resources as commitments agreed upon during that conference come into play. Much of
this decrease will have to occur with the cooperation of major polluters, such as heavily
populated countries like China and the United States (the chart below shows each
participating countries present contributions to greenhouse gas emissions), to reach and
maintain significant drops in carbon emissions, with aid flowing to developing countries
to avoid the coal and fossil fuel consumption and, instead, develop renewable
4
alternatives. Moving directly to clean, renewable energy technology is often referred to as
“leap frog” development.
Share of Greenhouse Gas Emissions by Countries with Climate Targets
Figure 2: Countries’ share of emissions as calculated by the Natural Resources Defense
Council from the 2012 total GHG emissions. Does not include the countries that did not
submit targets for 2012 (NRDC, Dec. 2015).
Leap frog technology development requires money. Prior to implementation of the
Kyoto Protocol “countries agreed to establish the multilateral Green Climate Fund (GCF)
to help mobilize funding in development countries to reduce emissions and adapt to the
impacts of climate change” in Copenhagen, Denmark (NRDC, 2015). The United States
alone pledged three billion dollars of about 11 billion dollars to support the GCF (NRDC,
5
2015). Such donations will be extremely important for the maintaining and/or reducing
the existing carbon dioxide emission levels in those countries as other countries reduce
their pre-existing levels.
A variety of renewable energy options exist for the participating COP countries to
select from as they pursue their individual commitments to reduce per capita emissions
resulting from the Paris Agreement. Technologies related to energy production include
wind turbines, solar panels, and hydroelectric dams used in many innovative ways across
the globe. For example, in Chile, a small fishing village, named Caleta San Marcos, is
developing a renewable energy plant that uses solar and hydroelectric methods to provide
24 hour access to electricity to the community. During the day the solar panels will
provide energy to the community and to a pump that transport ocean water to a storage
area, so that at night the water can be released to spin turbine generators to generate up to
300 Megawatts (MW) daily (Jarroud, 2016). This project creatively takes advantage of
the seaside cliffs found in abundance along the Chilean coast and suggests the potential
for innovation and efficient use of renewable resources as the renewable energy industry
expands.
6
There is an advantage to developing innovative renewable energy projects, such as
the one in Chile, but many of the ecological damages tend to be overlooked by the
developers and the associated energy companies. Hydropower has been used heavily
around the world, but the damming process often creates reservoirs, which disrupts the
preexisting ecology of the location, permanently altering the habitat. Dams prevent fish
from traveling up or down stream, and allow for silt accumulation that eventually clogs
the water intakes and may contain toxic material. Additionally, the reservoirs cover the
original vegetation, causing it to decay and release methane, a potent greenhouse gas. In
fact new studies show that dams may significantly contribute to greenhouse gas
emissions due to this reason (Magill, 2014). Environmental impacts are not limited to
dams, but they occur in all renewable energy projects.
The United States has been exploring wind energy as one of the options for
decreasing carbon emissions. Presently the United States is second in wind energy
production at 17.2% of the shared capacity, following China’s 33.6%, shown below in
Figure 3.
7
Figure 3: Cumulative capacity of the top ten wind energy producers compared to the rest of
the world. Sourced from the Global Wind Energy Council, Global Wind Statistics 2016
(GWEC, 2016).
The United States likely will continue to develop wind energy due to the commitment
made at the Paris Conference to cut greenhouse gas emission by 28% of the 2005 levels
[44,153 metric tons (World Resources Institute, July 2009)] by 2025 (NRDC, December
2015). Wind energy may be a significant actor in achieving this goal based on “the
Department of Energy estimates that if wind energy provides 20% of American Energy
needs by 2030, it can reduce annual emission of carbon dioxide by approximately 825
8
million metric tons” (Evans, 2014). This achievement would result in a 1.8% cut from
2005 emissions, enlarging the total cut so far from 12% (EIA, May 2016) to nearly 14%.
Although wind farms represent an opportunity for the United States to reduce carbon
emissions from the electricity generation sector, they do have known ecological impacts
that must be addressed. For example, wind turbine siting puts birds at a particularly high
risk of mortality and displacement, and prevents a species from remaining in a location
due to the loss of essential resources and/or necessary habitat qualities. As a result, it is
important to mitigate the negative impacts on birds prior to the expansion of the wind
energy industry in the United States. This research outlines one approach to do just that.
In this study, I will perform a spatial analysis of the Pacific Northwest (Washington,
Oregon, Idaho, and Montana) to develop a map series and web application tool using
ArcGIS (Geographic information systems) and ArcMap Online to find potential locations
for new wind farms that will account for:
1. The exclusion of protected areas and other important bird habitat, such as nesting
sites,
2. The exclusion of open water such as rivers and lakes,
9
3. The encouragement of maximum use of agricultural and pasture land to promote a
multipurpose landscape, and
4. The avoidance of existing wind farms sites.
By accounting for all of these restrictions, the potential mortality rates and effects
of displacement on local and migratory bird populations as a result of wind farm
development will be minimized. The result will be the creation of a preliminary tool that
can, and should, be improved upon through additional research to include the other
alternative energy options and account for other species that are experiencing negative
impacts from similar development projects.
In the chapters that follow I will introduce the history of wind powered
technology to put its development into perspective alongside the impacts that the turbines
have on bird communities. The present bird mortality estimates as a result of wind farm
development are very low (0.01% based on Stevens, et al., 2013 and Lucas, et al., 2008),
but that number fails to include other factors that may significantly increase the total
mortality rate. I will describe several of these additional factors in detail and how
mortality may result. Furthermore, I will explain the purpose of using GIS analysis tools
as an alternative to other modeling programs.
10
11
Literature Review
Development and Problem Recognition
Wind Energy Technology Development
The use of wind as an energy resource has been around for thousands of years, but
it was only in the mid 1800’s that wind turbines were optimized for electricity production
and to use as water pumps in the United States. Between 1850 and 1970, over six million
machines were developed for pumping water (Kaldellis & Zafirakis, 2011).The U.S.
Wind Energy Company opened its doors and designed a windmill for the Midwest known
as the Halladay Windmill in 1850, named for Daniel Halladay, one of the company’s
owners (Office of Energy Efficiency & Renewable Energy). In 1890 that metal blades
were introduced to the design for their durability as the structures grew taller to catch
more wind and create more energy. After that, the development of wind turbine
technology took off.
In the mid 1900’s wind farms started to be developed across the United States at
utility scale. Following the oil crisis in 1973, and until 1986, the commercial wind turbine
market evolved primarily from domestic and agricultural applications to encompass
utility interconnected wind farms (Kaldellis & Zafirakis, 2011). It was an option explored
12
with vigor, because it offered a way for the United States to maintain electricity resources
in the midst of oil shortages (Wind Energy Foundation). The first large-scale wind energy
project occurred in California (1981-1990) with over 16,000 installed turbines. Figure 4
below depicts the rapid increase in wind produced energy following 1990, and the
projections for wind energy production, as well as other renewable sources, through
2035.
Figure 4: Projected increase in renewable energy generation (excluding hydropower) from
1990 to 2035. Sourced from the U.S. Energy Administration, Annual Energy Outlook 2012
Early Release (EIA, Today in Energy, February, 2012).
13
The current installed capacity of wind energy in the United States resulted from
developing wind energy technology at the utility scale. However, the understanding of
how renewable energy is classified at the utility scale has been under debate since the
transition to interconnected utility scale wind farms. John Kaldellis and D. Zafirakis
(2011), defined agricultural and domestic applications as 1-25 kilowatts and utility scale
wind farms as 50-600 kilowatts. The California Energy Commission (2016) and
Greentech Media (2013) both define a utility scale renewable energy resource as
producing 10 megawatts or more. However, when the first “utility-scale” wind farm was
developed it supported a local community near Grandpa’s Knob, Vermont for several
months during World War II, year 1941, producing only 1.25 megawatts (Office of
Energy Efficiency & Renewable Energy). For the sake of simplification in the case of
this study, a utility-scale wind farm serves a community through a public utility service,
regardless of its kilowatt capacity.
Ecological Impact Research Exploration
As the development of the turbine technology took off and wind farms started to
spring up, investigation of bird mortality began. “The high number of bird takes each
14
year is largely a result of the design of the wind turbines” (Andrews, 1999). Wind turbine
designs dating back to 5000 B.C.E. were typically wooden structures with fabric or wood
blades used to propel boats, pump water, or (later) grind grain (Wind Energy
Foundation). American colonists used wind turbines for many of the same purposes, but
the development of electrical power brought new applications for turbines (Wind Energy
Foundation).
Engineers have designed modern turbines to be more efficient. The modern
turbines require a greater amount of vertical space than the older ones, and, in the case of
horizontal axis turbines (turbines with long, outward blades), placement to catch higher
wind speeds (Evans, 2014). To support these designs, wood and fabric have been
replaced with sturdier materials including metal and carbon fiber. The height and
materials combined make it harder for birds to avoid turbines in a utility scale setting
where there are many structures in one area and mortality is more likely to result when
the birds collide with the stronger materials. Additionally, because computers assist in
moving the blades to best catch the headwinds, and because the tips of the blades can
move up to 180 mph, it can be difficult for birds to navigate through the farm without
collision (Evans, 2014).
15
Part of the efficiency improvements during siting is that they are placed in areas
with higher wind speeds. “Wind turbines are often arranged in rows, along coasts or
mountain ridges,” which is, typically, where soaring and migrating birds use the air
currents to travel and hunt (Barrios & Rodriguez, 2004). Mortality increases when
turbines are placed in the path of birds which are hunting, gliding or flying using the
wind. For this reason that it is not recommended for wind farms to be built along
important migratory routes or in important areas, such as breeding or hunting grounds
(Everaert, et al, 2014).
The ecological impacts largely went unnoticed until about 30 years ago
(Smallwood, Bell, Snyder, & Didonato, 2010). Around that time, more intensive studies
on wind turbine bird collision mortality were conducted. Ultimately this led to the
realization of the negative impacts of wind turbines on birds, more so among some
species groups than others. Many of these early studies focus on the short-term effects of
individual wind farms on mortality, and concluded that wind turbine collision mortality
contributes to 0.01% of all bird death cases (Lucas, et al., 2008). In later years, other
researchers challenged the technique of the study as insufficient for determining the
16
overall threat on bird posed by wind farms in general and based on species restrictions.
This revelation will be discussed in the following section.
Impacts of Wind Turbines
Numerous impacts associated with wind farm development negatively affect the
conservation of the bird community health. These impacts include displacement resulting
from human disturbance (the presence and activity of humans within or near habitat
areas), collision mortality, barrier effects that disrupt habitat connectivity, habitat loss or
change, and the associated effects on available resources, such as prey (Langston, 2013).
Each of these variables will be discussed in the following below.
Understanding the impacts in detail and considering ways to mitigate these impacts as
the wind energy industry grows will help align the industry with the goals of conservation
groups, environmental advocates, and the public. For example, Europe has a binding
agreement (outlined in Directive 2009/28/EC) to produce 20% of the energy consumed
from renewable energy sources by 2020, but some projects have been cancelled or
postponed due to the potential impacts on wildlife (Johnston, et al., 2014). For countries
17
in Europe, as well as other countries expected to experience an increase in wind energy
(United States and Canada in particular), improving the evidence base reducing the
uncertainty surrounding the impacts on wildlife will benefit the renewable energy
industry and the European national conservation advisors and regulators by preventing
cancellations or delays at a high financial cost (Johnston, et al., 2014). Increasing access
to knowledge on the ecological impacts will ultimately lead to more insight and research
efforts on how to effectively mitigate the problems to promote development of the wind
energy industry.
Displacement
Many species of birds exhibit avoidance behaviors when interacting with tall
man-made, or anthropogenic, structures and will ultimately leave that area completely
(Kuvlesky, et al., 2007). Grassland species for example, demonstrate avoidance behaviors
near anthropogenic structures during and after construction, suggesting that the presence
of a utility-scale wind farm “could degrade habitat to such an extent that it would be
avoided or abandoned by sensitive species” (Stevens, Hale, Karsten, & Bennett, 2013).
This is known as displacement and is “often considered to be a greater threat to bird
18
populations than collision fatalities” (Stevens, et al., 2013) because it reduces the amount
of habitat available for sensitive species and/or reduces the amount of habitat
connectivity. Avoidance behavior may be viewed as positive “in terms of reducing
collision risk and associated direct mortality,” but the consequences of making usable
habitat unavailable tend to be unclear (Langston, 2013). Although displacement is an
issue for most bird species, it is more significant for species that have low human
disturbance tolerance thresholds, such as the Marbled Murrelet or the Northern Spotted
Owl.
Many of the displaced species are listed as endangered or threatened and the
scope of their available habitat tends to be considered critically low. Siting a new wind
farm within an undisturbed area (or an area not used for human purposes, such as
grasslands) poses a risk to birds by reducing the amount of habitat available for the
sensitive species, fragmenting suitable habitat, and making it more dangerous for the
birds to travel between suitable habitat parcels. For example, a bird may avoid the
immediate vicinity of a wind farm, since it may be viewed as an obstacle or barrier
(Madden, et al., 2009). By exhibiting avoidance behavior the bird cannot use potential
feeding habitat and necessitates “additional flight to avoid the obstacle” resulting in
19
excess energy expenditure, which could affect breeding success and survival (Langston,
2013; Madsen, et al., 2009; Kuvlesky, et. Al., 2007).
The amount of displacement due to wind farm construction varies and is
considered species specific, ranging between 100 to 500 meters in displacement from a
single wind farm (Steven, et al., 2013; Langston, 2013; Kuvlesky, et al., 2007). However,
T.K. Stevens et al. (2013) discovered that certain bird behaviors may be indicative of the
amount of displacement that occurred. The presence of wind turbines did not impact three
of the four grassland species studied as heavily, but these species were not limited to the
resources available in undisturbed landscapes and did use those found in altered habitats,
such as agricultural and grazing fields. The single species more severely displaced didn’t
frequent disturbed areas and was limited to locations with low human disturbance,
dominated by native vegetation.
While the habitat preference of an individual species or groups of species may be
an indicator of the species’ or group’s reaction to wind farm placement, Steven et al.
discovered evidence to suggest that predator avoidance behaviors may better reflect
responses to wind turbine development on a broader response spectrum. The displaced
species adopts a cryptic evasion strategy, meaning that individuals hide and remain
20
undetected while predators go by. “If cryptic species perceive wind turbines as a
predation risk [as raptor perches for example] then [they] may be more likely to avoid
areas with wind turbines” (Stevens, et al., 2013). Based on these findings, the researchers
hypothesized that “species occupying natural or semi-natural habitat would be more
sensitive to wind energy development than species occupying intensively produced
landscapes” (Stevens, et al., 2013).
Using cryptic species as an indicator species may be an appropriate method for
adjusting turbine siting practices, to avoid negatively impacting the disturbance sensitive
species. This would mean including a suitable buffer zone around the wind turbines, to
account for the amount of unsuitable habitat and the range of displacement post
construction. Since species that already have a high tolerance for human disturbance did
not experience high impacts from displacement, siting wind farms with in agricultural
and grazing lands, or other disturbed landscapes, may not severely impact those
communities.
21
Fatigue
As stated briefly above, another consideration when addressing the impacts of
displacement is the fatigue experienced by birds confronted with an obstacle that leads
them to expend excess energy flying around it, which could affect breeding and survival
when they frequently have to avoid anthropogenic structures (Langston, 2013; Madson,
et al, 2009; Kuvlesky, et al, 2007). This issue particularly affects species which have
large ranges, such as raptors or migratory species, because fatigue can increase mortality
events. (Citation) Placing wind farms within heavily used migratory corridors, forces
species that utilize it to go around the obstruction in the airspace or avoid the location
(Kuvlesky, et al, 2007). The impacted species are then likely to use more energy and
potentially with less for resources to replenish their energy before the next stretch toward
their goal. Unfortunately, this topic has been largely unexplored and the available
evidence is anecdotal, at best.
Cumulative Mortality Impacts
Bird mortality presents another significant concern for wind farm development,
but the implications suggest that the mortality rate per farm is fairly low compared to
other causes for mortality. Newer studies reveal that while the impact of an individual
22
wind farm impact may be low, the cumulative impact on bird populations may be
detrimental to the overall population health of individual species (Brabant, et al., 2015).
This factor is especially important for migratory species with varying densities in a single
space over time and throughout a region.
As the number of wind farms increases in the United States, it will be essential
understand the cumulative effects of wind farms within migratory corridors and along
coastlines, because there may be indications of species health decline from increased loss
of individuals. In an assessment of the cumulative risks of collisions among the North
Sea offshore wind farms, researchers discovered that “up to 1,046 seabirds are expected
to collide with the turbines” and an “estimated 297 thrushes to collide with offshore
turbines in a single night [during migration]” along the North Sea wind farm structures
alone (Brabant, et al, 2015). These results demonstrate that “the cumulative impact of
large-scale wind farms development [has the potential to] cause significant increases in
bird mortality levels” which would put the affected populations under pressure (Brabant,
et al, 2015). The results of this study may be transferable, but there are few cumulative
impact studies available and this type of research is necessary for appropriately modeling
the distances needed between wind farms to reduce the cumulative impacts.
23
Raptors and Seabirds
The discussion of habitat displacement suggests a gap in the investigation of the
contribution that wind farms have in total annual avian mortality. Even so, most of the
mortality analyses “do not acknowledge that some bird species may be affected more by
wind turbines that other anthropogenic sources” (Lucas, et al, 2008). Large bird species,
such as raptors and seabirds, are more vulnerable to fatal collisions with wind turbines
than with other structures. At the Altamont Pass in the United States and Tarifa in Spain,
raptors show some of the highest levels of reported mortality due to their dependence on
thermals to gain altitude for travel and foraging (Wang, et al, 2015). The tendency to fly
lower for foraging and use of the thermals for travel makes it more likely for them to
collide with wind turbines, because these behaviors place them right in their path (Evans,
2014). This behavior also has been observed with seabirds. Seabirds use the wind coming
of the water or other thermals to glide while traveling and feeding, which makes offshore
wind farms a threat for these species as well (Wang, et al, 2015).
Another characteristic of these species that may be of concern when siting wind
farms is that many of them have longer life spans, lower fecundity rates, and smaller
24
healthy population size compared to other species (Ballard & Bryant, 2007; Brabant, et
al, 2015). By having a lower fecundity rate these species already have a lower
reproductive potential than other species, meaning they may wait multiple years before
mating and, once they select a mate, they do not produce many young. That being said, if
too many members of a single population are lost due to collisions with wind turbines,
the reproductive potential will be reduced even further (Evans, 2014; Kuvlesky, et. Al.,
2007).
The potential effects of detrimental population impacts on raptors has been
particularly evident on Somalia Island, Norway. Prior to the construction of a wind farm
in XXXX, 13 White-tailed eagle pairs were present within a 500 meter territory
(Zimmerling, et al, 2013; Nygard, et al, 2010). In the years following construction, only
five pairs remained. Thirty-six white-tailed eagles carcasses were recovered by 2009,
suggesting that the collisions directed impacted the health of the local eagle population
(Zimmerling, et al, 2013; Nygard, et al, 2010). Similar, long-lived species in the United
States with low reproductive rates are likely to experience similar impacts when turbines
are poorly sited--placed in important nesting and hunting territories. Unfortunately, there
25
is very little population level quantitative data to support these concerns (Zimmerling, et
al, 2013).
As for Seabirds, offshore wind farms are far more difficult to monitor for
collisions than onshore wind farms, because carcass searches are impractical in this
particular setting. As a result there is very little collision data for offshore wind farms
(Brabant, et al, 2015). Brabant, et al, (2015) did monitor collisions at multiple offshore
wind farms along the North Sea using visual counts and radar observations to determine
the number of birds flying through the location and the number of collisions each day.
They determined that 98% of all the fatality victims to be seagulls along this stretch, with
a small percentage being the migratory songbirds that utilized this corridor (Brabant, et
al, 2015).
The distance of individual wind farms from the coast may play a significant role
in the high percentage of seagull victims. “The Bligh Bank location compared to OWEZ,
located respectively 46 versus 10 km from the coast, which is inevitably reflected in
lower background densities of gulls” (Brabant, et al, 2015). This study provides some
evidence that proper placement of wind turbines can reduce the number of fatalities of
seabirds.
26
It might be possible to reduce the mortality of birds by turning the turbines off
during migrations or peak breeding seasons if necessary, because mortality theoretically
increases during periods of high population abundance. For some species this idea may
be valid. However, in the case of raptors, this theory is not accurate. Lucas, et al, (2008)
analyzed 10 years of bird mortality data for Tarifa in Cadiz, Spain, and compared those
figures to bird abundance. They found no correlation between abundance and mortality
(Lucas, et al, 2008). Griffon Vultures, for example, had higher collision rates during the
winter even though the period of highest abundance was during pre-breeding season in
the spring (Lucas, et al, 2008). This was the case for the majority of the studied raptors.
This study provides evidence that higher abundance does not always mean higher
instances of collision mortality. Deeper exploration of “the mechanisms involved in
influencing collision risk,” (Lucas, et al, 2008) will be necessary, but will also require the
improvement of mortality estimations and development of consistent protocol for carcass
searches (Wang, et al, 2015; Everaert, et al, 2014; Lucas, et al, 2008).
27
Scavenger Removal and Carcass Searches
Many other variables need to be considered when determining the collision
influence on total annual bird mortality. Some carcasses get removed by mammalian and
avian scavengers, or are not recovered because they fall outside of the search radius
(Johnson, et al, 2016; Wulff, et al, 2016; Smallwood, et al, 2010). As a recognized
problem for mortality estimation, some studies have adjusted for the relative collection
biases. Johnson et al (2016) understood several of the inaccuracies that need to be
accounted for when comparing avian fatalities at different wind farms. They stated that
the number of carcasses found is a biased estimator of the actual number of fatalities due
to spatial incompleteness, temporal incompleteness, incomplete availability, and
imperfect perceptibility (Johnson, et al, 2016). In lay-man’s terms, the entire expanse is
not usually investigated for fatalities, the searches only occur for a part of the year,
scavengers remove some of the carcasses, and some carcasses are missed by the
observers (Johnson, et al, 2016, Wulff, et al, 2016, Smallwood, et al, 2010, Johnson, et al,
2002).
To account for these biases there are a few recommendations given by the authors
of these studies. First of all, covering the entire area rather than preforming searches on a
28
fraction of the area on occasion may give a more accurate adjustment in the estimators
currently used (Johnson, et al, 2016). Additionally, Johnson et al (2016) mentioned that
the searches are temporally incomplete, only occurring six to nine months of the year, so
continuing the searches for the whole year will improvement mortality estimates as well.
Some of the solutions already used are to customize adjustment factors for estimator
method used, search interval, and classification for carcass removal (Johnson, et al, 2016;
Erickson, et al, 2014).
While there are solutions for some of the biases, it is difficult to develop more
precise estimates for the removal of carcasses by scavengers. A study using scavenger
removal trials recognized that previous scavenger removal trial studies may have
experienced “scavenger swamping,” which means that the researchers placed too many
carcasses per the area to accurately estimate the rate at which scavengers take their meals
(Smallwood, et al, 2010). The resulting mortality rate estimation tool underestimates the
removal rate of birds killed by turbines.
29
Mortality Estimation Accuracy
“It is assumed that wind farms are less harmful to birds that other energy
industries or other human-made structures” (Barrios & Rodriguez, 2004), but this is
largely because of the limited available knowledge on the mechanisms driving bird
collision mortality (Zimmerling, et al, 2013; Masden, et al, 2009; Barrios & Rodriguez,
2007). As mentioned above, the mortality rates of birds at individual wind farms cannot
be determined simply based on the number of carcasses with characteristic impact
injuries found present after a period of time, a common practice. Additionally, early
estimates of bird mortality from collision with wind turbines may have been grossly
underestimated. Some of the earliest estimates place total annual mortality from wind
turbine collisions at less than 0.1% (Stevens, et al, 2013; Lucas, et al, 2008), but more
recent studies reveal that it could be as high as 0.8% when factoring habitat loss
associated with wind turbines (Zimmerling, et al, 2013). This suggests that the overall
mortality estimation may increase as more factors are included and models more
accurately depict the influences of collision mortality.
As discussed previously, scavenger removal is only one of many issues that may
influence the low mortality rate estimates. Displacement and fatigue are often left out of
30
the equation and are difficult, and nearly impossible, factors for wind farm operators to
monitor effectively. However, developing a more accurate fatality rate estimate is
necessary for assessing the effectiveness of impact reduction measures, but “remain
imprecise and potentially biased by common field methods” (Smallwood, et al, 2010).
This practice, combined with technological modeling ingenuity, may represent an
opportunity to “balance technical requirements with environmental considerations,” and
refine spatial planning techniques accordingly (Langston, 2013).
Science in Technology-Based Planning Approaches
One of the key components for completing this research to is develop a tool that is
easily accessible to all community members, as well as to key players in wind farm
development, a tool that will suitably describe the potential for minimizing ecological
impacts of wind farms while encouraging sustainable infrastructure development. Using
technology to reach a greater audience aligns with the goals of the National Science
Foundation (NSF), according to Nalini Nadkarni and Amy Stasch in How broad are our
broader impacts? An analysis of the National Science Foundation's Ecosystem Studies
Program and the Broader Impacts requirement (2013). This goal appears under the
31
Broader Impacts Creterion (BIC) with the intention to broaden the use and understanding
of science and technology for enhancement of research and education, while creating
benefits to society and include the participation of underrepresented groups (Nadkarni,
Stash, 2013).
As researchers develop the knowledge base on the broad scope of bird mortality
risk, there lies potential to develop a tool that can semi-accurately calculate the impact of
wind turbines on birds to find suitable wind farm locations based on flight patterns and
population density models along a temporal scale. Spatial analysis is a useful tool for
visualizing the potential for wind farm siting in the Pacific Northwest while considering
the reduction of bird mortality. In order to make wind farm siting a more participatory
process it is important to make the tools and models easy to understand, otherwise the
model is essentially useless because people cannot understand what is being proposed
(Al-Kodmany, 2000). For example, in a study in the UK and Europe, Alison Johnston et
al (2014) modeled flight heights of 25 seabird species from 32 sites to estimate collision
rates with wind turbines for proposed offshore wind farms (OWF’s). Based on the flight
height ranges and different turbine designs for hub height and rotor diameter, the
resulting “distribution makes it possible to consider how different turbine designs and
32
collision risk with the rotor-swept area may affect collision rate estimates” (Johnston et
al, 2014). This is a great way to assess design options in sensitive areas, but is not readily
available for public education and interpretation. The result of the Johnston et al study
was a matrix of graphs for each species. Visually, it is complex and difficult to
understand without prior explanation of the significance of the graph matrix and the
modeling process.
In a similar study, Liechti et al (2013) created a model for Switzerland through a
simulation and recognizes that birds have specific spatial and temporal patterns that they
follow when migrating and assert that "the best way to mitigate conflicts between birds
and wind turbines is to avoid their spatial concurrence." This is one way to account for
impact on birds by predicting collision potential from this model, and it resulted in a
product that was more user-friendly for people outside of the field of study. However a
series of assumptions underlying the model could have influenced the collision rate
estimates. These assumptions included uniform flight patterns year to year with account
for variation (based on species or just general pattern changes, fixed bird concentration
migrating along a route, and did not consider changes in flight patterns from other
construction or proximity to nearby habitat. The sole focus on migratory pattern and
33
exclusion of roosting sites exhibits endangerment risk for unknown population damages,
based on this modeling technique.
Although the Liechti et al modeling example may provide a sufficient conflict
mitigation solutions, the result is not readily accessible to the general population because
it requires a deeper understanding of the modeling process and its limitations in order to
understand the output. Al-Kodmany (2000) explains that GIS offers "a way to provide all
participants with full access to a large amount of spatial data in the form of easily
digestible, non-threatening graphics and maps." Using GIS provides the opportunity to
create an interactive tool that will allow a broad audience to understand why some
locations should be selected as potential wind farm sites, while others are not ideal,
without requiring the viewer to understand the process of building the geodatabase and
the resulting maps.
34
Methods
Goal/ Products
The overall goal of this study was to create an easy to understand tool that can be
used by a broad audience to site/understand the siting of potential wind farms in the
Pacific Northwest for new wind farms that do not interfere with open water areas and the
Important Bird Areas (IBA’s) that have been established by the National Audubon
Society. To accomplish this, I prioritized agricultural lands to promote a multipurpose
landscape, while avoiding the IBA’s. The end product will include a preliminary map
series created in ArcMap.
Data Collection & Resources
As indicated, the data used in this study came from existing resources, The United
States Geographic Survey (USGS) which provides GIS point data for all of the existing
turbines in the United States. Land use data tends to come in a variety of forms, but the
most reliable is raster data. A raster file consists of a matrix of pixels that are organized in
to a grid, where each pixel represents information (ESRI, 2008). In this case, the
information represents land use types. The National Land Cover Land Use data for 2014
35
is the most recent raster data available. Accompanying the raster in a separate file is a
table that contains the necessary metadata used during the manipulation process. Lastly,
the bird habitat data came from the National Audubon Society’s Important Bird Area
polygon data. This data shows important nesting, feeding, and resting habitat for local
and migratory bird species.
Data Manipulation: ArcGIS Desktop
Gathering together the various data sets acted as a starting point, but several steps
were required before the desired result emerged. After downloading the data into a
geodatabase I first reduced the data to the study area (Washington, Oregon, Montana, and
Idaho or the Pacific Northwest) to make them more manageable. Following the reduction,
I created wind farm polygons based on the wind turbine point data and added a buffer
around the IBA’s of 500 meters. This distance was selected based on a series of studies
that found evidence suggesting bird displacement from habitat areas between 100 and
800 meters (Steven, et al., 2013; Langston, 2013; Kuvlesky, et al., 2007). The majority of
these studies stated that the greatest distance for displacement is 500 meters, so I chose
36
this number as the buffer distance; there was little support in the literature for the 800
meter distance.
The raster file required the greatest amount of manipulation to be used for my
study. The initial raster file contained data for the entire United States, so the first step
was to reduce the file to my study area to make it easier to manipulate. I used the Raster
Clip analysis tool to limit it to the general area of the PNW. To read the raster file there is
a key available on USGS website in the same location as the raster file download. I put
the key into excel and turned it into a flat file so that I could use it to designate land use
types. To connect to the two files I “joined” them by classification number.
Next, I created additional layers to narrow down possible wind farm locations.
One of my goals was to suggest multipurpose landscapes. By selecting the agricultural
and pasture/hay classifications I created a new layer that removed all other ineligible land
types. To further reduce this file, I created a second new layer of all the open water and
placed a second buffer around it to protect riparian corridors as a sensitive, protected
area. Finally, I eliminated the polygons that intersected with the IBA buffer, the open
water buffer, and the existing wind farm polygon set. The end result was several
37
polygons considered eligible for wind farm placement by the standards I had set for this
study.
With a file that could now be analyzed, I aggregated the polygons to reduce the
number of borders present and performed a near analysis. A near analysis creates the
final new layer file that had to be developed prior to creating the map series. In this final
file, I reclassified the symbology to reflect the following site quality levels:
●
High- agricultural lands, farthest from IBA’s or greater than 4000 m (meters)
away,
●
Moderate- agricultural lands or minimally developed lands, Relatively far from
IBA’s or between 2000 m and 4000 m away, and
● Low- agricultural lands or minimally developed lands, close to IBA’s less than
2000 m away.
38
Maps
This chapter displays some of the map products that may be available by using my
completed geodatabase as a resources for finding and developing new wind farms with
ecological consciousness. The legend given on this page is applicable to all of the maps
in this section. There are two copies of all of the land classification maps, with and
without the IBA’s, followed by close up images of the intersections of existing wind
farms with the IBA’s. This map set includes full scale images of the Pacific Northwest
Available Lands, followed by close up image of each
state in this order: Washington, Oregon, Idaho, and
Montana. The intersection maps follow the same
order.
In the upper left corner of each map is a scale
equation so that the viewer can develop a spatial
understanding of the land classifications, IBA’s, and
Wind farm/IBA intersection points. The north end of
the each map is oriented in the same direction of the
title with the map description at the south end.
39
Figure 5: Available land for the entire study area for wind
farm siting with classifications for poor, intermediate, and
good locations with the Important Bird Areas for
comparison.
40
Figure 6: Available land for the entire study area for wind farm
siting with classifications for poor, intermediate, and good
location without Important Bird Areas.
41
Figure 7: Available land classifications for Washington State with Important
Bird Areas for comparison.
42
Figure 8: Available land classification for Washington State without Important
Bird Areas.
43
[Grab your reader’s attention with a
great quote from the document or
use this space to emphasize a key
point. To place this text box anywhere
on the page, just drag it.]
Figure 9: Available land classifications for Oregon States with Important Bird
Areas for comparison.
44
Figure 10: Available land classifications for Oregon State without Important Bird
Areas.
45
Figure 11: Available land classifications for Idaho State with Important Bird Areas
for comparison.
46
Figure 12: Available land classifications for Idaho State without Important Bird Areas.
47
Figure 13: Available land classifications for Montana State with Important
Bird Areas for comparison.
48
Figure 14: Available land classifications for Montana State without Important
Bird Areas.
49
Figure 15: Columbia River Scenic Area along the border of Washington and
Oregon States. Overlap of existing wind farms with Important Bird Areas.
50
Figure 16: Wildhorse wind farm in Washington State overlap with
Important Bird Area.
51
Figure 17: Hot Springs Wind wind farm overlap with Important Bird Areas in
Idaho.
52
Figure 18: Rim Rock wind farm overlap with Important Bird Areas in Montana.
53
Results
Existing Windfarm Sites
In the PNW existing windfarms cover 1,398 square kilometers (km2) of the
landscape, with 446 km2 in conflict with the IBA's. This is about 32% of the windfarms
presently in operation, conflicting with 0.2% of the National Audubon Society’s IBA's.
The areas of conflict are primarily located along the Washington/Oregon border with
single instances in Mid-Washington, Northern Montana, and South-western Idaho
(Figures 15-18). The names of the conflicting sites are the Windy Point/Windy Flats in
the Columbia Hills, Shepards Flat North in the Boardman Grasslands, Wild Horse in the
Quilomene-Colockum Wildlife Area, Rim Rock at Kevin Rim, and Hammet Hill
Wind/Hot Springs Wind in the Snake River Birds of Prey NCA/CJ Strike Reservoir.
Potential Windfarm Sites.
The total area available for multipurpose landscape potential is 171,872 km2. This
total area is divided into the three class described in the methods chapter.
54
Area per Quality Classification Type
Quality
Poor
Intermediate
Good
Area (kmsq)
143,514
14,916
13,442
Table 1: Area available for each of the classification types developed.
The majority of the available land area falls into the “Poor” category, since it is
close to an IBA. The smallest amount of land is “Good” for wind farm development,
based on the standards of this study.
At a distance it appears some of the potential sites should not be accounted for as
"Poor" quality. For example, Figure 5, does not contain detail seen in the close-up maps
of the individual states. In Washington, it appears from a distance that no IBA's should
cause the large amount of "Poor" quality land. Upon closer examination, however, IBA's
appear. In the following map, Figure 19, the large map shows Washington and the circle
identifies a location that cannot be seen clearly from the full view distance.
55
Figure 19: Example of a location where closer images are important for seeing the
details that caused the resulting classification of the available land for future wind
farm development. Location is found in Mid-Washington.
The smaller image is a close-up view of the location that shows the amount of
detail missed in the larger images. This view shows that there are, in fact, IBA's,
influencing the classification.
On a similar note, the more area a map covers, the less defined the border of the
available lands. It can give a sense of continuity where there is none. In the example
below, Figure 20, the circled area in the Oregon view appears to have a contiguous line of
“good” available land for wind farm development. However, when zoomed to the same
56
area at a 1cm to 8 km scale, we can see that there are several small polygons instead of a
few contiguous polygons.
Figure 20: Example of a location where connecting polygons are deceiving at a
distance and a zoomed in map reveals greater boarder details.
57
Discussion
The mapping tool proposed in this study takes several concerns regarding
ecological impacts in to consideration; that the ecological impact of future wind farm
development may be reduced if this map product is used when making siting decisions.
By taking the precaution to not only exclude IBA's, but to include an excluded buffer
zone around them, I have narrowed down the potential sites as much as possible to
protect birds from heightened mortality risk using currently available knowledge. As a
result, the majority of the potential sites fall within the "Poor" classification. This also
indicates that there are many important habitat sites for birds within or near agricultural
and pasture/hay land types.
Fortunately, there are several square kilometers of available land within the
"Good" and "Fair" classifications. Pursue wind farm development in the "Good"
polygons, would increase the total land area in the PNW for wind energy by 961% versus
the current area used. This would not require use of undeveloped lands; this number
solely reflects the amount of farmland that may be available for multipurpose use.
Despite the fact that there is plenty of land available in a multipurpose landscape without
58
interfering with the IBA's, there were a few instances where the existing farms overlap
with an IBA (See Figures 11-14).
Based on the evidence from the literature, we can assume that some level of
displacement of local and migratory species occurs within the sites where existing wind
farms and IBA's conflict. These IBA’s are important to pay attention to when using the
maps, because they should be avoided whenever possible to prevent total displacement of
the local species. This is part of the reason we need to establish multipurpose landscapes
whenever possible and must do so as far away from important habitat zones as possible.
Creating buffers allows some leniency for placing wind farms near IBA's, but the "Poor"
quality of the close potential locations signals caution when evaluating those sites.
Limitations
Despite the classification into which each polygon is placed, it should not be
excluded from investigation for wind farm siting. Errors that can be expected when using
GIS that can influence the results. In this study I needed to simplify the data by
aggregating the polygons, because the data was too dense to classify. As a result, the
polygons are much larger than they were initially. There may be parts of the "Poor"
59
classification that may be the same quality as the "Intermediate" classification, but is not
shown at that level since it was a part of a contiguous polygon in the "Poor" quality
classification. In the end, the size of the polygons could affect the classifications.
However, this is not always true. As displayed in the results section (Figure 15),
some of the IBA’s are very small and ill-defined at a distance, which makes them
difficult to identify. Using the zoom function in the GIS program is important for
understanding why some of the expansive “Poor” zones are so large where is seems there
should be more “Intermediate” and “Good” zones.
A similar concept applies to understanding the expanse of individual polygons.
Figure 16 shows how a map at a greater scale may provide deceiving information on the
continuity of the available lands polygons. Where they appear to be connected, they are
actually several smaller polygons in a concentrated area. Therefore, depending on the use
of the map products, the user may need to have closer images for the maps to be
effective.
While this map series can provide valuable information on the potential locations
for placing wind farms, there were a number of other limitations involved with
60
completing this study. First of all, these maps were produced strictly based on raster data
relationships. The study has not incorporated data or maps for wind speed, flight
behaviors of the local avifauna, or individual state regulations. Additionally, the map
scale for the printed series will not be suitable for planning site assessments because the
detail is not refined enough to find the border. Solutions to these problems will be
discussed in the following section.
User Recommendations
After completing the study I have developed a few user recommendations for this
map series so that the best, most informed wind farm placement decisions can be made.
First and foremost, these maps should always be used as a preliminary resource for
beginning a wind farm siting assessment. This spatial analysis will be most effective for
reducing impact of wind farm development on the bird community when combined with
other spatial analysis on flight and feeding behaviors. Additionally, the user should
measure wind speed and complete any other siting regulations before selecting a site for
development.
61
It is possible to create a printed map series using the same tools presented in this
study, however since there is so much detail to the data it is better to use the map product
on the ArcGIS software. By doing so, the user can zoom in further on a location if they
are interested in pursuing it as a wind farm location.
62
Conclusion
Future Research
Continuing to develop the available knowledge on wind turbine impacts on birds
is key to improving the available resources for siting wind farms in an ecologically
sensitive manor. One of the major research gaps that will be important for understanding
the impact of wind farms on birds, is the cumulative impact of wind farms along
individual corridors and within individual regions. As discussed in the study of the North
Sea Migratory Corridor, the cumulative impact of the wind farms along the corridor
likely causes greater impact on bird mortality than the individual wind farm estimates can
predict (Brabant, et al, 2015). If we understand the cumulative impacts of wind farms in
the PNW, then we can more deeply understand the risks associated with bird mortality as
a result of wind turbine collisions.
Another major knowledge gap is on fatigue resulting from avoidance behaviors.
As discussed in the Literature Review "Fatigue" section, there is very little research
specifically focusing on how fatigue from avoiding wind turbines specifically can impact
bird populations. While we know that extra expenditures of energy can damage otherwise
healthy birds, there are other dynamics at play when looking at the situation with a wind
farm in the mix. Since wind farms can and have displaced bird species from their now
unsuitable habitat, food, water, and nesting resources are reduced significantly
(Kuvlesky, et al, 2007; Langston, et al, 2013; Steven, et al, 2013). With little knowledge
63
on the subject there is no way of knowing what the population impacts are, so exploring
it will provide the opportunity to develop more accurate mortality estimates for existing
and future wind farms. Once we know what the impacts of fatigue are more precisely,
the depth of the buffer used for the mapping process may need to be adjusted to reduce
the impact of fatigue on the local and migratory bird species throughout the PNW.
Lastly, developing a spatial flight pattern analysis for birds in regions of the
United States where there are existing or proposed wind farms is one way to improving
available data. Liechti et al (2013) developed a 3-dimensional model for the flight
behaviors of birds along their migratory routes in Sweden. The key aspect that the study
in Sweden addresses, but this study does not, is the third dimensional space. Without
considering the patterns of the PNW bird species, we cannot fully understand the impact
of wind turbines along migratory corridors or between IBAs. Creating a similar spatial
flight pattern analysis will give the map product greater depth and will better define the
best areas for new wind farm development.
64
Final Remarks
Understanding the ecological impacts of the technology we develop is one of the
most important things we can do as develop sustainable options for energy resources.
Birds are particularly susceptible to wind farm development because they are predisposed
to heightened mortality risk. While the current mortality impact is estimated very low, the
number is only based on carcass removal surveys at individual wind farms. There are
several other factors that need to be considered when discussing the impact of wind
turbines on bird mortality, including the collective impact of wind farms along corridors
or regions, displacement, fatigue.
Since we are aware of these problems it is entirely within our power to preemptively
strike and a take an active stance on minimizing these problems, while encouraging
continued renewable energy resource development. Avoiding key habitat areas, such as
key nesting sites and locations with high bird population densities, and protected lands
will be the best way to start dealing with ecological impacts of wind farms while
promoting the development of renewable energy resources. To successfully do this we
must plan sufficiently based on the available knowledge about local birds species.
65
This study acts as a starting point for successfully planning wind farms in an
ecologically conscious manor. By avoiding establish important habitat (IBA's), riparian
areas, and undeveloped land for the development of wind farms the PNW now has some
direction for developing the local wind industry. However, as stated previously, this
study is a starting point and is not meant to be used without additional spatial
understanding. Performing a flight behavior analysis for local birds, as was completed in
Sweden, for the identified potential wind farm and using it in tandem with the map series
is highly recommended, because it will help provide more accurate information about
where the birds are the most vulnerable so that we can avoid those locations.
66
Bibliography
Al-Kodmany, K. (2000, May). Public Participation Technology and Democracy. Journal
of Architechtural Education, 53(4):220-228
Andrews, R.N.L. (1999). Managing the environment managing ourselves: A history
of American environmental policy. Yale University Press.
Barrios, L., Rodriguez, A. (2004). Behavioural and environmental correlates of soaring-bird
mortality at on-shore wind turbines. Journal of Applied Ecology, 41:72-81.
Bennett, V. J., Hale, A. M., Karsten, K. B., Gordon, C. E., Susan, B. J. (2014) Effect of Wind
Turbine Proximity on Nesting Success in Shrub-nesting Birds. The American Midland
Naturalist, 172(2):317-328. Retrieved
from: http://www.bioone.org/dog/full/10.1674/0003-0031-172.2.317
Brabant, R., Vanerman, N., Steinen, E. W. M., Degraer, S. (2015, February). Towards a
cumulative collision risk assessment of local and migrating birds in North Sea offshore
wind farms. Hydrobiolgia, 756:63-74.
C2ES (2015, December). Essential elements of a Paris climate agreement. Center for climate and
energy solutions. Retrieved from: http://www.c2es.org/
California Energy Commission. Utility Scale Renewable Energy. Retrieved from:
www.energy.ca.gov/research/renewable/utility.html
EIA. (2012, February). EIA projects U.S. non-hydro renewable power generation increases, led
by wind and biomass. Today in Energy. Retrieved from:
https://www.eia.gov/todayinenergy/detail.php?id=5170
EIA. (2016, May). U.S. energy-related carbon dioxide emissions in 2015 are 12% below their
2005 levels. Today in Energy. Retrieved from:
https://www.eia.gov/todayinenergy/detail.php?id=26152
Erickson, W.P., Wolfe, M.M., Bay, K.J., Johnson, D.H., Gehring, J.L. (2014). A comprehensive
analysis of small-passerine fatalities from collision with turbines at wind energy facilities.
9(9): e107491
67
Evans, R.K. (2014). Wind turbines and migratory birds: Avoiding a collision between the energy
industry and the migratory bird treaty act. North Carolina Journal of Law & Technology.
32
GWEC. (2017, November). Global wind statistics 2016. Global wind energy council. Retrieved
from: www.gwec.net/wp-content/uploads/vip/GWEC_PRstats2016_EN_WEB.pdf
Heiman, M.K., Solomon, B.D. (2004, March). Power to the people: Electric utility restructuring
and the commitment to renewable energy. Annals of the Association of American
Geographers, 94(1):94-116. Retrieved from: http://www.jstor.org/stable/3694070
Johnson, D.H., Loss, S.R., Smallwood, K.S., Erickson, W.P. (2016). Avian fatalities at wind
energy facilities in North America: a comparison of recent approaches. HumanWildlife Interactions, 10(1):7-18
Johnson, G. D., Erickson, W. P., Strickland, M. D., Shepard, M. F., Shepard, D. A., Sarappo, S.
A. (2002, Autumn). Collision mortality of local and migrant birds at a large-scale windpower development on Buffalo Ridge, Minnesota. Wildlife Society Bulletin, 30(3):879887. Retrieved from: http://www.jstor.org/stable/3784243
Johnston, A., Cook, A.S.C.P., Wright, L.J., Humphreys, E.M., Burton, N.H.K. (2014). Modelling
flight heights of marine birds to more accurately assess collision risk with offshore wind
turbines. Journal of Applied Ecology, 51:31-41.
Kaldellis, J. K., Zafirakis, D. (2011). The wind energy (r)evolution: A short review of a long
history. Renewable Energy, 36:1887-1901.
Kristoferson, L. (1992, February). Global Energy and Sustainability: A Complicated
Matter. Ambio, 21(1):88-89. Retrieved from: http://www.jstor.org/stable/4313893
Kuvlesky, W.P., Brennan, L.A., Morrison, M.L., Boydston, K.K., Ballard, B.M., Bryant, F.C.
(2007, November) Wind Energy Development and Wildlife Conservation: Challenges
and Opportunities. The Journal of Wildlife Management, 71(8):2487-2498. Retrieved
from: http://www.jstor.org/stable/4496368
Langston, R. (2013, March) Birds and Wind Projects Across the Pond: A UK Perspective.
Wildlife Society Bulletin, 37(1):5-18. Retrieved
from: http://www.jstor.org/stable/wildsocibull2011.37.1.5
68
Lapena, B.P., Wijnberg, K.M., Hulscher, S., Stein, A. (2010, October). Environmental impact
assessment of offshore wind farms: a simulation-based approach. Journal of Applied
Ecology, 47(5):1110-1118. Retrieved from: http://www.jstor.org/stable/40835769
Liechti, F., Guelat, J., Komenda-Zehnder, S. (2013, April). Modelling the spatial concentrations
of bird migration to assess conflicts with wind turbines. Biological Conservation, 162:2432. Retrieved from: www.elsevier.com/locate/biocon
Lucas, M., Janss, G. F. E., Whitfield, D. P., Ferrer, M. (2008, December). Collision fatality of
raptors in wind farms does not depend on raptor abundance. Journal of Applied
Ecology, 45(6):1695-1703. Retrieved from: http://www.jstor.org/stable/20144148
Magill, B. (2014, October). Methane emissions may swell from behind dams:
hydropower, which is increasing worldwide, may prove a huge source of the potent
greenhouse gas. Scientific American. Retrieved
from: https://www.scientificamerican.com/article/methane-emissions-may-swell-frombehind-dams/
Masden, E. A., Fox, A. D., Furness, R. W., Bullman, R., Haydn, D. T. (2009, May). Cumulative
impact assessments and bird/wind farm interactions: Developing a conceptual
framework. Environmental Impact Assessment Review.
McCall, M.K., Minang, P.A. (2005, December). Assessing participatory GIS for communitybased natural resource management:claiming community forests in Cameroon. The
Geographical Journal, 171(4):340-356
Nadkarni, N.M., Stasch, A.E. (2013, January 16). How broad are our broader impacts? An
analysis of the National Science Foundation’s ecosystem studies program and the
broader impacts requirement. The Ecological Society of America, 11(1):13-19. Retrieved
from: http://about.jstor.org/term
National Audubon Society. () Important Bird Areas. National Audubon Society. Retrieved from:
www.audubon.org/important-bird-areas
NRDC. (2015, December). The Paris Agreement on Climate Change. NRDC. IB:15-11-Y
Past Present and Future: Understanding Wind Power. Retrieved from:
www.brynmawr.edu/geology/206/tomich.htm
69
Smallwood, K. S., Bell, D. A., Snyder, S. A., Didonato, J. E. (2010, June). Novel scavenger
removal trials increase wind turbine-caused avian fatality estimates. The Journal of
Wildlife Management, 74(5):1089-1097. Retrieved from: http://www.jstor.org/stable/406
Stevens, T. K., Hale, A. M., Karsten, K.B., Bennett, V. J. (2013, June). An analysis of
displacement from wind turbines in a wintering grassland bird community. Biodiversity
Conservation. 22:1755-1767.
USGS. Onshore Wind Turbine Locations for the United States. Retrieved from:
https://eerscmap.usgs.gov/windfarm/
USGS. (2011). NLCD Land Cover Land Use 2011. Retrieved from:
https://www.mrlc.gov/nlcd2011.php
Van den Bergh, J.C.J.M. (2013, February). Policies to enhance economic feasibility of
sustainable energy transition. Proceedings of the National Academy of Sciences of the
United States of America, 110(7)2436-2437. Retrieved
from: http://www.jstor.org/stable/41992347
Wind Energy Foundation. Retrieved from: windenergyfoundation.org/wind-at-work/wildlifeenvironment/
Wulff, S.J., Butler, M.J., Ballard, W.B. (2016).Assessment of diurnal wind turbine collision risk
for grassland birds on the sourthern great plains. Journal of Fish and Wildlife
Management, 7(1)129-140
Zimmerling, J. R., Pomeroy, A. C., d’Entremont, M. V., Francis, C. M. (2013). Canadian
estimate of bird mortality due to collisions and direct habitat loss associated with wind
turbine developments. Avian Conservation and Ecology, 8(2):10. Retrieved
from: http://doi.org/10.5751/ACE-00609-080210
70
