-
extracted text (extracttext:extracted_text)
-
Understanding Amphibian Behavior:
Diel Cover Use Patterns in Alpine Lakes
With and Without Introduced Salmonids
by
Tyler Goodman
A Thesis
Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
June 2018
©2018 by Tyler Goodman. All rights reserved.
This Thesis for the Master of Environmental Studies Degree
by
Tyler Goodman
has been approved for
The Evergreen State College
by
______________________________________
John Withey, Ph. D.
Member of the Faculty
________________________
Date
ABSTRACT
Understanding Amphibian Behavior: Diel Cover Use Patterns in Alpine Lakes
With and Without Introduced Salmonids
Tyler Goodman
Many of the world’s amphibian species are in decline. A large portion of studies on amphibian
populations in alpine environments focus on the factors that affect their success. Past research has
pointed to a seemingly obvious cause of these declines: the presence of native and introduced
salmonid species. While some researchers are confident in this mechanism, far fewer have
examined the complete diel cycle of amphibian behavior in response to salmonid presence as well
as in their absence. This study documents amphibian behavior during both day and night cycles
with an emphasis on how available cover is utilized. All lakes observed in this study showed
greater amphibian detectability during night observations, with variation in preferred substrate
type and activity level. This study points to the possibility that amphibians are adapting to
biological competition and predation within their alpine environments and may be equipped to
adapt to abiotic changes. The assumption of a salmonid–caused decline may not be warranted
under all circumstances as the greatest number of detected amphibians came from study sites with
regularly occurring salmonid stocking. Factors such as substrate type and structure should be
more seriously considered in management decisions, as should the practice of incorporating
complete diel data collection to assess population trends. Determining the extent of salmonid
impact allows a prioritized management effort to focus limited funds where they can provide the
most benefits.
Table of Contents
List of Figures …………………………………………………………………………...iv
List of Tables …………………………………………………………………………….v
Acknowledgements ……………………………………………………………………..vi
Chapter 1: Introduction and Literature Review ……………………………………....1
Introduction ……………………………………………………………………………….1
Literature Review: Interactions Among Amphibians, Salmonids, Biotic, and Abiotic
Influences …………………………………………………………………………………5
History of Fish Stocking ………………………………………………………….5
Motivations for Stocking Salmonids ……………………………………………..7
Amphibians of the Cascades ……………………………………………………...9
Opposition towards fish introduction ……………………………………………11
A Case for Stocking ……………………………………………………………..13
Altered Amphibian Behavior ……………………………………………………14
Gaps in Scientific Knowledge …………………………………………………..16
Reaching Solutions ……………………………………………………………...17
Chapter 2: Methods for Data Collection and Analysis ………………………………19
Study Lake Selection ……………………………………………………………………19
Field Data Collection ……………………………………………………………………21
Visual Encounter Survey (VES) ………………………………………………...21
Statistical Analysis ………………………………………………………………………23
Chapter 3: Results ……………………………………………………………………...24
Relationship Between Fish and Amphibian Abundance ………………………………..28
Other Influencing Variables …………………………………………………………….29
Activity Coefficient ……………………………………………………………………..31
iv
Chapter 4: Discussion and Conclusion ……………………………………………….33
Understanding Lake Ecosystems ………………………………………………………..33
Study Limitations ………………………………………………………………………..37
Next Steps in Research ………………………………………………………………….37
Literature Cited ………………………………………………………………………..38
v
List of Figures
Figure 1. Study Ecoregions ……………………………………………………………...19
Figure 2. Observations of Individual Amphibians ………………………………………25
Figure 3. Species Composition By Observation Period …………………………………26
Figure 4. Comparison of Observed Abundance …………………………………………27
Figure 5. Relative Abundance By Species ………………………………………………28
Figure 6. Activity Coefficient Comparison By Lake ……………………………………32
vi
List of Tables
Table 1. Trout and Char Species Stocked in Alpine Lakes of Washington ………………6
Table 2. Amphibians of the Cascade Range …………………………………………….10
Table 3. Study Lakes with Selected Attributes ………………………………………….21
Table 4. Regression Relationships Between Amphibian Types ………………………...29
Table 5. Pearson’s correlation …………………………………………………………..30
vii
Acknowledgements
Although I was able to do what I love and spend time outdoors for this thesis, it
would not have been possible without support from many individuals. The support of
faculty, peers, friends and family was overwhelming and greatly appreciated. I would like
to start by thanking my faculty reader Dr. John Withey for his tremendous support and
patience through my thesis work, especially with statistical analyses. I’d also like to
thank my all of my MES peers at Evergreen as they gave feedback through the
developmental stages of my work. Thank you Jessica Brown and Leslie Carman for your
peer-review feedback. Lastly, I’d like to thank Becca Goodman and Erik Lentz for
venturing off into the unknown with me to some of the study lake sites.
viii
Chapter 1: Introduction and Literature Review
Introduction
After the last ice age, high elevation lakes were formed as glaciers melted and left
behind ice-scarred landscapes barren of macroscopic life. As ecological communities
established themselves within these alpine systems, amphibians typically became the top
predators in aquatic systems, as fishes were denied access due to steep gradients and
natural obstacles (Bahls, 1992; Pister, 2000). Rugged terrain limited the distribution of
amphibians and concentrated them around reliable sources of water. As humans explored
these montane lakes and ponds, the practice of fish stocking gained popularity as it
provided recreational angling opportunities in pristine settings (Pister, 2000; Landres,
Meyer and Matthews, 2000). The Western North American Ranges of the Sierra
Nevadas, Rockies and Cascades were the primary hotspots for such practices. Little was
understood as to how alpine communities were impacted by such action and not until the
mid 20th century was the impact of fish stocking considered on a management level (Miro
and Ventura, 2013; Downen, 2002).
The topic of amphibian success is multi-faceted and involves many complex
factors including biotic and abiotic inputs. With environmental factors rapidly changing
due to climate change, it is difficult to pinpoint specific root causes for decreased
amphibian success and therefore talk of correlational relationships is commonplace in
alpine aquatic ecology. Amphibians are indicators of environmental health, particularly
water quality and also play an important role in ecosystem health as both predators and
prey (Lunghi, Manenti and Ficetola, 2015). As montane systems are relatively simple and
less diverse in comparison to their lower elevation counterparts, studying more simplified
1
systems allows isolated analysis of variables that may aid in improving the trajectory of
amphibian populations (Carpenter and Kitchell, 1993).
The Cascade Range of Washington State has largely been free from time sensitive
demands to save declining amphibian species. Of the species found at higher elevations
(>1200m), the Cascade frog (Rana cascadae) is the most impacted, listed as near
threatened (NT) by the IUCN (2017). Land managers responsible for the conservation of
species have not yet had to establish urgent action plans, but could benefit from a
preemptive level of preparation for future needs should the species and others decline
further. In order to better understand the needs of amphibian species many agencies at the
federal and state levels conduct routine amphibian surveys to establish baseline data
about the status of various species. North Cascades National Park (NOCA) and Mount
Rainier National Park (MORA) have been leaders in Washington State with yearly
surveys that incorporate citizen science for monitoring (Hoffman, Larson and Brokes,
2002; Downen, 2004). While much effort goes into these survey efforts, most of the
observations occur during daylight hours. This potentially represents a problematic
situation as critical decisions regarding amphibian management and the allocation of
limited funding depends on the data collected. By examining half of the time period of
activity within a body of water, opportunities are missed for additional data collection,
which could lead to more sound representations of amphibian populations.
Relatively untouched by human development, the North-South Cascade Range is
intersected by four major East-West highways that connect larger cities on either side of
the range. The western side of the range receives more annual precipitation due to coastal
storms and typically exhibits qualities of a temperate rainforest, whereas the east side
2
sees the effects of a rain shadow. The eastern slope of the Cascades is much drier and
water catchments rely more heavily on stored snowpack for water flow (Downen, 2004).
The alpine ecosystems on each side of the Cascade crest differ in community complexity
and needs.
Environmental and Anthropogenic Influences on Amphibian Populations
Some of the most recognized contributing factors to declining amphibian
populations include changes in temperature and precipitation. Amphibians rely on
seasonal and permanent bodies of water for reproduction and development, which may be
at risk due to climatic variation. The challenge with understanding how specific
populations will be impacted and react is that each basin is uniquely influenced by factors
of aspect, snowpack and elevation. A cascading effect results from these factors as faster
snowmelt and more direct exposure to ultraviolet light affects growth and reproductive
success (Nagl and Hofer, 1997; Sommaruga, 2001; Adams et al., 2005). Similarly,
climatic variation coupled with anthropogenic carbon inputs can change water chemistry.
Because of the remoteness and narrow window of seasonal accessibility, these
ecosystems are grossly understood despite their relative simplicity.
Anthropogenic influences range from the above mentioned alteration of the
carbon cycle and atmospheric chemistry to localized impacts of disturbance and overuse.
While the impacts of historic fish stocking are still being mitigated in some alpine lakes,
land managers at the state level are calling for a prioritized mindfulness of amphibians in
decisions to continue these practices (Tyler et al., 2002). While controlling anthropogenic
inputs to alpine systems at a global or national level are impractical, more localized
3
decisions that favor declining species may make a greater difference. The simple task of
reducing impact through low-density stocking or even complete elimination of stocking
where deemed necessary can allow opportunities for recovery where humans have
overstepped in the past.
A Balancing Act
Making management decisions at the crossroads of conservation and recreation
can be very challenging. While the focus primarily falls on obligations to protect or
conserve native amphibian species, the matter also has a social side that is equally as
complex. Land managers, namely the Washington Department of Fish and Wildlife, are
responsible for addressing concerns on both fronts. The balance of conservation efforts
and public recreation opportunity can be seen as conflicting goals. For many people, the
culprits of amphibian decline are introduced salmonids. The recreation side of the
responsibility requires a careful management for a quality over quantity experience. In
many regards these goals can coexist, but the cases where amphibians are in greater
decline could stand to use a management system that is capable of assessing the greatest
threat to those species. By understanding the complete life cycle of an amphibian and all
of the factors that influence its success, managers are equipped to take action where
issues of amphibian decline are the most time sensitive. This thesis seeks to bridge the
gap of current surveying practices by looking to other indicators of amphibian health
within lake ecosystems to then organize lakes in a prioritized list of needed management
action. Understanding the research that exists within this area of study helps to set the
stage and evaluate management practices that are based on these findings.
4
Literature Review: Interactions Among Amphibians, Salmonids, Biotic,
and Abiotic Influences
Setting the Scene
Globally, climate is affecting species that require specific habitat conditions
necessary for breeding, rearing offspring, and acquiring food, as these conditions are
changing at a pace too rapid for species to readily adapt (Carey and Alexander, 2003;
Parmesan, 2006; Case et al., 2015). Amphibians across the world are of particular
concern and often face many other pressures correlated with human impact and recreation
(Ryan et al., 2014). One such impact that is a topic of debate is non-native salmonid fish
stocking in mountain lakes. Amphibians at high elevations are susceptible to changes in
the snow pack and available water for rearing and summer time habitat, and often require
deeper, more permanent bodies of water (Case et al., 2015; Carey and Alexander, 2003;
Taylor, 1983). In certain bodies of water trout and char species such as rainbow trout
(Oncorhynchus mykiss), Westslope cutthroat trout (Oncorhynchus clarki lewisi), Coastal
cutthroat trout (Oncorhynchus clarkii clarkia), golden trout (Oncorhynchus mykiss
aguabonita) brook trout, (Salvelinus fontinalis), and artic grayling (Thymallus arcticus)
have been stocked for recreational fishing opportunities and have replaced amphibians as
the top predators in these alpine ecosystems (Table 1; Pister, 2001). Much of the debate
in the fields of ecology, conservation biology, and public land management are based on
this head-to-head competition between salmonids and amphibians.
5
Table 1. Trout and Char Species Stocked in Alpine Lakes of Washington State (Bahls,
1992; Pister 2001; WDFW, 2018).
Common Name
Scientific Name
Status
Westslope Cutthroat Trout
Oncorhynchus clarki lewisi
Current
Coastal Cutthroat Trout
Oncorhynchus clarki clarki
Current
Rainbow Trout
Oncorhynchus mykiss
Current
Golden Trout
Oncorhynchus aguabonita
Current
Eastern Brook Trout
Salvelinus fontinalis
Historical
Lake (Mackinaw) Trout
Salvelinus namaycush
Historical
Arctic Grayling
Kokanee Salmon
Thymallus arcticus
Onorhynchua nerka
Historical
Historical
Academic literature on the topic often portrays the anti-fish perspective and calls
for actions to remove salmonids in totality (Drake and Naiman, 2000). While there can be
ecological value in this action in certain lakes where the impacts of fish on amphibians
are greatest, this task is not feasible with limited management resources and short seasons
of accessibility in remote areas. This literature review begins with an overview of the
6
historic motivations for fish stocking, with special attention on the Cascade Range of
Washington state. Next, I will discuss prominent literature that has established an antistocking mentality among researchers and land managers. After this I will present
counter-arguments to these viewpoints and elucidate the complex nature of alpine lake
ecosystems.
Historical Motivations for Stocking Salmonid Species
Beginning in the 1800’s fish stocking became common practice among loggers,
trappers and outdoorsmen who visited mountain lakes and wanted recreational
opportunities through sport fishing as well as opportunities for sustenance over prolonged
periods of time in remote locations (Pfeifer et al., 2001). Stocking practices were very
much rooted in the pursuit of catchable trout with a “singular goal to enhance sport
fishing without consideration of ecological ramifications” (Pister, 2001). Since the last
ice age nearly 95% of mountain lakes in the western U.S. were naturally fishless, but
stocking efforts resulted in 60% of these high-elevation lakes containing trout species
(Tyler et al., 1998). While most stocking activity occurred under the direction of various
state departments, illegal stocking activity from unknown parties has been and remains to
be an ongoing concern for land managers. Common legal methods involved airplane
drops of tens of thousands of fish in hopes that a sizeable majority would survive the
traumatic experience (Pfeifer et al., 2001). The most commonly utilized species was the
eastern brook trout (Salvelinus fontinalis), now understood to have detrimental impacts
on lake biota as they are successful breeders in a range of habitats (Knapp et al., 2001).
Where most other trout and char species would be unsuccessful breeding, brook trout
thrive and can quickly overpopulate a lake, depleting it of food sources. Without an
7
understanding of the life history, species characteristics, or potential impacts on habitat
and native species, the practice of overstocking lakes continued for many years.
As these practices continued, the Leopold Report of 1963 and the Wilderness Act
of 1964 changed the way that fish stocking was conducted within National Parks and
Wilderness areas respectively (Leopold et al., 1963; Landres et al., 2001). Additionally,
a decline in the quality of the fishery was detected and a re-evaluation of lake
management took place and favored low impact stocking by significantly lowering the
number of fish stocked in each stocking period. In some cases the impacts on native lake
biota were extreme as species, namely S. fontinalis, consumed whatever resources were
available (Pfeifer et al., 2001). Within Washington State there are many agencies
involved in the preservation and restoration of ecosystems as well as providing
recreational opportunities. The agency charged with this balancing act is the Washington
Department of Fish and Wildlife. Efforts to provide “sustainable fishing, hunting and
wildlife viewing opportunities” require adoption of new values surrounding fish stocking
by being responsible stewards of managed land (WDFW, 2015).
By using best available science to assess impacts of introduced fish species on
native biota, there can be responsible implementation of management practices that
provide conservation of species and maintain a high lakes fishery where deemed
appropriate. A major driver of continued stocking is the economic value that recreational
fishing provides. Angler surveys reflect a large percentage of license-buyers that frequent
high lakes for recreation, accounting for $67 to $70 million annually (Pfeifer et al., 2001).
Changing societal values have largely influenced a shift towards a conservationist
mentality as activities such as hunting and fishing have seemingly fallen secondary to the
8
goals of protecting biological diversity (Landres et al., 2001). Despite a changing
paradigm, the economic and cultural values associated with trout and other cultivated
introduced species makes decisions to end stocking programs difficult (Hartman et. al
2014).
Amphibians In Question
Amphibians that inhabit the Cascade Range include various species of frogs and
salamanders, each exhibiting characteristics that have evolved over time. There are
fifteen common species of frog and salamander that inhabit regions found within this
study in what is considered an alpine landscape (Table 2). Each has adapted behaviors
and responses to predatory threats, as well as increased competition with others of their
species, or other amphibians. Historically, the literature classifies frogs as nocturnal
creatures with a tendency to be most active and vocal during evening and nighttime
hours. More inquestion is the behavior of salamanders and newts . With more complex
life history stages there are circumstances in which adults leave a body of water for a
subterranean life. Many of the individuals found to inhabit lakes and ponds long-term are
gilled larvae. Larvae from each species have varying durations of life as a juvenile, based
largely on elevation and snowpack. Some species take as long as two years to complete
their metamorphosis into adults. Others permanently retain gills and other larval
characteristics and live an aquatic lifestyle, a phenomenon known as paedomorphism
(Farner and Kezer, 1953; Tyler et al., 1998b) .
9
Table 2. Amphibians of the Cascade Range in Washington. NT = Near Threatened,
LC = Least Concern (Nussbaum et al., 1983; IUCN, 2004; IUCN, 2017).
10
Common Name
Scientific Name
IUCN Status
Cascade Frog
Rana cascadae
NT
Coastal Tailed Frog
Ascaphus truei
LC
Pacific Tree Frog
Pseudacris regilla
LC
Columbia Spotted Frog
Rana luteiventris
LC
Red-legged Frog
Rana aurora
LC
Western Toad
Bufo boreas boreas
LC
Rough-skinned Newt
Taricha granulosa
LC
Ensatina
Ensatina eschscholtzii
LC
Cope’s Giant Salamander
Dicamptodon copei
LC
Pacific Giant Salmander
Dicamptodon tenebrosus
LC
Long-toed Salamander
Ambystoma macrodactylum
LC
Northwestern Salamander
Ambystoma gracile
LC
Van Dyke’s Salamander
Plethodon vandykei
LC
Western Red-backed Salamander
Plethodon vehiculum
LC
Larch Mountain Salamander
Plethodon larselii
NT
The variation among salamanders and newts in behavior is so great that there is
not one widely accepted time of day to find them active. This is part of what prompted
my interest in studying amphibians in alpine environments. Little work has been done to
understand the full diel life history and activity of salamanders and newts in response to
salmonid presence or absence. One area that the literature has covered particularly well is
the defense response of amphibians, even analyzing the angle at which they flee from
predators. A handful of studies look at the ability of predators to alter the behavior and
use of cover by amphibians. In both lab and field experiments there are tendencies noted
towards increased hiding behavior (Kenison et al., 2016; Pilliod et al., 2010; Hoffman,
Larson and Brokes, 2003; Walls, 1995). The utilization of this cover is an adaptation, as
amphibians were once the top predators in alpine aquatic systems. Similarly, an
adaptation towards nocturnal behavior could represent an adaptation for predator
avoidance.
An Alpine Scapegoat?
With the shift of societal values and scientific papers using combative language
about non-native fish including “biological pollutants” (Schindler and Parker, 2001) and
“alien” (Kats and Ferrer (2003); Cambray, 2003; Winandy et al., 2015), little support
exists for the continued practice of salmonid fish stocking. Further attacking this practice
is the assumed attitude that the presence of fish is synonymous with destruction and
predation. A study by Knapp et al. (2007) in the Sierra Nevada Mountains has been
11
adopted as a seminal paper within the academic community, without careful
consideration for the differences among species and watersheds that were studied. This
study examined a declining population of the mountain yellow-legged frog as a result of
shared wintering habitat under frozen high-elevation lakes as a potential “synergistic”
factor with a variety of environmental influences (Vredenburg and Wake, 2004; Knapp
et. al 2007). Exhibiting this unique characteristic, overwintering tadpoles are subjected to
fish searching for food during the most resource poor time of the year. Fish removal
efforts in affected lakes and streams proved effective initially in rebounding populations
within these bodies of water (Knapp et al., 2007). While this study may prove to be of
great significance in the studied lakes, it becomes problematic to assume this impact
across all bodies of water and amphibian species that do not exhibit the same life history
characteristics.
Similarly, another study conducted in the Sierras attributed terrestrial-aquatic
links with donor and recipient systems. In fishless lakes, mayflies are abundant and
substantial hatches of the insects shape distribution patterns of Gray-crowned RosyFinches. Initial insights into the reduced presence of Rosy-Finches around lakes with
introduced trout pointed to the robbery of resources. While connections could be made
between mayfly availability and fish presence or absence, it was found that additional
factors such as tree cover surrounding the lake was also a determinant of Rosy-Finch
presence (Epanchin 2010).
While various field-based studies draw correlations between fish presence and
amphibian impacts, other efforts have been made to explore these impacts in laboratory
recreations of habitat (Huang and Sih, 1990). One such study conducted by Tyler et al.
12
(1998) recreated various substrate and habitat cover scenarios, controlling water
temperature, period of light exposure, and feeding intervals. The ratio selected by the
researchers was one fish per twenty larval salamanders. The conclusions of this study
pointed out that there is correlation between fish presence and amphibian use of cover.
While the scientific method for testing hypotheses is a valid approach to understanding
behavioral patterns, fabricating predation pressure in a laboratory setting overlooks
various biotic and abiotic factors that may favor either fish or amphibian. In all cases
amphibians utilized cover structure that was made available to them.
A Case For Stocking
With past mistakes recognized and scientific inquiry informing management
decisions, this is the most responsible period in the history of fish stocking. Efforts by
managers within the North Cascades National Park Complex (NOCA) show that a
balance can exist between conservation and stocking. These examples can serve as a
framework of responsible high lakes fisheries management during a time in which funds
are limited for removal efforts that are rarely successful. The methods outlined in High
Lakes Management documentation carefully analyze lake biota and habitat before fish
are considered for stocking (Downen, 2004). Many view the pursuit of a successful high
lake fishery to be connected with past approaches of vast quantities of fish, but efforts to
minimize ecosystem disturbances through low-density fish stocking achieve responsible
management while appealing to both conservationists and recreationalists (Downen,
2004; Pope, 2008). Stocking that once was conducted with aircraft depositing tens of
thousands of fish has shifted to efforts by volunteer backpackers carrying as few as fifty
fish. With a decrease in the number of fish in lakes there is a better “quality over
13
quantity” approach as lake biota are minimally disturbed with a low number of nonreproducing fish living for a finite period of time. Should ecological impacts be greater
than initially anticipated the stocking regime can simply be halted and the fish will die off
naturally (Liss et al., 2002).
Studies on amphibians at various elevations and global locales have pointed to the
ability of certain species to adapt to changing conditions and environmental stressors.
The North Cascades National Park Complex (NOCA) has noted such behavioral shifts as
the development of the North Cascades National Park High Lakes Fishery Management
plan has called for in-depth baseline surveys of lake habitat. One observation of long-toed
salamander larvae shows that larger bodied (i.e. more developed, older) larvae were more
prone to finding refuge in substrate than their smaller, younger siblings (Downen, 2004).
Additionally, lakes that were observed during afternoon hours with few observed
amphibians came to life at night, pointing to a shift in behavior towards nocturnal
activity. The locations of amphibian sightings also reflected tendencies towards more
shallow nearshore areas or areas with significant bottom cover (Sih et al., 1992; Downen,
2004).
Others have pointed to the ability of different life history stages of amphibians to
utilize available structure and alter activity levels in response to stress (Winandy et al.,
2015; Walls and Wiliams, 2001). Additionally, increased predation stress can cause
amphibians to be selective with breeding and feeding sites, which may not be possible in
isolated watersheds that lack additional ponds and lakes to which amphibians can migrate
(Winandy et al., 2016).
14
Beyond the Adaptation Theory
While the controversy is focused on the amphibian-fish interaction, a host of other
biotic and abiotic variables may be at work as either the primary forces of amphibian
success or synergistic contributors to the impacts of salmonids. Returning again to
NOCA, a study conducted by Tyler et al. (1998) attributed the abundance of crustacean
zooplankton (e.g. copepods and gammarus) to Total Kjeldahl Nitrogen (TKN)
concentrations within a lake. This in turn contributed to an increased abundance of longtoed salamanders (Tyler et al., 1998). Other impacts of trout on alpine ecosystems
remove the direct conflict between amphibian and fish and look into trophic cascade
impacts at lower trophic levels. Such studies look at fish predation on phytoplankton and
zooplankton and other invertebrates within lake ecosystems. An Italian study conducted
by Tiberti et al. (2014) examined the feeding ecology of introduced S. fontinalis and
concluded that adult fish were more impactful on zooplankton depletion within a lake
than younger fish, with more limited mouth gape (Tiberti et al., 2014).
Additional explanations exist for amphibian behavior that is viewed to have
driving factors outside of predation response. A study by Pough (2007) analyzed
amphibian activity in response to lighting, UVB radiation, and temperature. While weak
correlations were made between use of cover in response to the intensity of lighting and
use of UVB radiation, temperature variation proved to be a significant driver of cover
use. Because amphibians have glandular skin that facilitates gas exchange and water
transport, extreme temperature fluctuations are stressful. Pough concluded that
amphibians chose shelter under benthic cover as a response to elevated temperatures and
fled in search of new cover when temperatures exceeded 32 degrees C (Pough, 2007).
15
Another perspective comes from research from Walls and Williams (2001), Wildy et al.
(1998), and Taylor (1983) who all review impacts of interspecific competition among
amphibians of the order Urodela and Anura. These interactions ranged from segregation
of habitat due to differences in life history traits and avoidance of competition to
cannibalistic behavior influencing growth rates of juvenile amphibians. The valuable
information gleaned from these studies is that given a host of potential influential
variables, there are a number of explanations for the decreased success of amphibians in
alpine ecosystems.
Gaps in Scientific Knowledge
With such variation in opinion on the topic, little progress is made pitting
conservation and recreation against one another. Land managers have a two-pronged
mission of conserving native species while also providing recreational opportunities for
license buyers, who ultimately fund agency programs. The issue that has come about is
the way in which the data is gathered to make the determinations of fish impacts and
recommended management intervention. With priorities increasingly placed on
conservation of native species, land managers within state and federal agencies have
developed detailed land management plans that outline clear policies and criteria for
surveying lakes for amphibian presence and fish impact. The gap that consistently comes
to the surface of this survey data is the fact that these presence-absence surveys are
largely conducted during daytime hours, disregarding amphibian activity that may take
place after dark.
16
Half of a species’ behavioral repertoire potentially goes unstudied because of
existing protocol. There are many benefits to incorporating both day and night surveys
for amphibian presence and activity. When we look to life history characteristics of
various amphibian species we find that many frog species exhibit tendencies towards
nocturnal behavior, however salamander activity is largely understudied and what little
research exists on alpine species varies by life stage. My work aims to bridge this gap in
knowledge by studying amphibian behavioral interactions with available habitat cover in
both the presence and absence of introduced salmonid species during daytime and
nighttime observational periods. While the reasons behind these protocols can be
understood due to the remoteness and relatively short study period of alpine areas,
understanding the complete picture within a given body of water can better inform how to
use limited resources.
Historically, if daytime observations conclude that amphibian presence is low it is
assumed that the population is under stress from predation. Methods for fish removal are
both time intensive and expensive, so being sure of the need is critical especially with the
underfunding that many agencies face. Based on studies that have anecdotally observed
increased salamander activity at night and European studies of lowland salamanders and
newts (Hartman and Lawler, 2014; Liss et al., 2002), I hypothesize that in instances
where amphibians are exposed to increased predation stress, species will adopt nocturnal
behavior to avoid predatory salmonid species. This would further point to the need for
nighttime surveys as populations may not be fully represented with daytime observations.
17
Reaching Realistic Solutions
While the literature that specifically studies alpine environments largely disfavors
the continuation of recreational fish stocking, it is unrealistic to expect a complete purge
of stocked salmonids. It is clear that many view salmonids as the sole culprit to declines
in amphibian populations in alpine lakes, but with increasing impacts from climate
change, there is a greater need for the synthesis of values between governing agencies
and the scientific community rather than the creation of a dividing fissure (Case et al.,
2015). This issue is not about whether or not fish belong, but how to deal with fish that
are already in place and may have detrimental impacts to amphibian species who share
the same lake ecosystem. In many cases the removal of fish would prove to be
advantageous for conservation as well as recreational opportunity as values are reflected
in the sustainability of the activity (Aasetre and Gundersen, 2012). Through my thesis
work I hope to be able to help shape and streamline the process of deciding which lakes
should receive prioritized attention in an effort to improve responsible stewardship when
managing native species. If we can move past placing blame and make progress towards
better understanding these alpine ecosystems, the scientific community and amphibians
would benefit alike.
18
Chapter 2: Methods For Data Collection and Analysis
Study Lake Selection
The data collection sites that I selected represent alpine lakes in three ecoregions
that intersect the Washington Cascade Range, including the North Cascades, West
Cascades, and East Cascades (Figure 1). These three adjacent ecoregions were selected to
represent variations in annual precipitation and temperature inputs for much of the
mountainous region within Washington. To select the lakes to be studied, I generated a
random list of lakes based on a set of minimal criteria to ensure an equivalent probability
to observe amphibians.
Figure 1. Ecoregions of Washington State, with the three ecoregions included in this
study (North, West, and East) in color.
Criteria for selection of individual lakes included a minimum elevation of 4,000
feet (1,219 meters), overall lake depth greater than 1 meter, and accessibility on foot. By
setting a lower limit on elevation, I was able to survey primarily alpine environments as
aligned with the methods of other researchers in alpine studies (Case et al., 2015; Knapp
19
et al., 2007; Liss et al., 2002; Tyler et al., 1998). The minimum depth threshold made it
more likely that the selected body of water would be permanent throughout the year and
therefore suitable for amphibians and/or fishes. Lastly, accessibility was a key
consideration as many of the surveyed lakes are located off trail, requiring up to three
days of travel on foot.
I used a random number generator to produce a numeric value for township,
range, and section, which I compared to a database of alpine lakes from the Washington
Department of Fish and Wildlife (WDFW). Once a lake’s coordinates matched the
random number, I noted any additional permanent bodies of water within 0.5 miles of the
lake, and included any such lakes in my study. I did this to maximize observation
opportunities in remote locations as well as to provide a more detailed snapshot of each
lake basin. I repeated this process until six lakes were selected in each ecoregion (Table
3, n=18 in total), four containing fish and two void of fish as a control (presence or
absence of fish based on WDFW fish stocking archives). I noted the lake elevation, size
of the lake in hectares (ha), and presence or absence of fish. In many cases the lakes
generated were unnamed in mapping publications and software, therefore I coded each
lake to represent its ecoregion, elevation rank, and salmonid presence (Table 3).
In addition to statistical analysis I calculated an activity coefficient to compare
overall amphibian activity level within a given lake between the two observation periods.
I assigned active behavior a value of “1”, resting behavior a “0”, and hiding behavior a “1”. The net activity values for each observation period were then divided by the number
of amphibians observed during that period to obtain the activity coefficient.
20
Table 3. Study lakes with selected attributes. Lake Codes contain data about lake
ecoregion and presence of introduced salmonids. W-West Cascades, E-East Cascades, NNorth Cascades, F-Fish present, NF- No fish present. The numbers are the rank from
lowest to highest elevation within each ecoregion.
Lake Code
Township-Range-Section (T-R-S)
Elevation (m) Size (HA)
W1-F
16N-07E-32B
1405
5.0
W2-NF
06N-08E-10P
1417
1.5
W3-F
06N-08E-10E/M
1433
13.0
W4-F
06N-08E-10L/P
1448
8.0
W5-NF
13N-10E-27H
1469
0.5
W6-F
13N-10E-13Q
1554
3.7
E1-F
22N-13E-24B
1429
4.9
E2-F
13N-11E-03B
1587
12.2
E3-NF
14N-11E-34G/K
1614
7.2
E4-NF
23N-14E-27C/F
1676
1.4
E5-F
27N-15E-34F
1792
3.5
E6-F
23N-14E-27F/L
1826
1.1
N1-F
24_1/2N-11E-32Q
1408
2.8
N2-NF
24_1/2N-11E-32Q
1420
0.5
N3-F
30N-12E-04F
1533
1.5
N4-F
32N-14E-27L
1700
6.4
N5-NF
32N-14E-26B
1713
2.5
N6-F
32N-14E-26G
1814
14.7
Field Data Collection
The process for collecting field data from each of the study lakes included
observing amphibians during daytime and nighttime hours. Equipment used for the
observation process included a 2-meter folding ruler, GoPro camera, field notebook,
21
pencil, survey flag, packraft, dive mask, dive light, and headlamp. Additionally, I utilized
a personal flotation device when rafting was necessary. Data collection at each lake
involved a visual encounter survey as is common within organizations in Washington
State management (Downen, 2014). A visual encounter survey (VES) involved one
revolution around each lake, observing each individual amphibian encountered. I
conducted two surveys on each lake, once during the day, and once at night within a
single 24-hour period.
Upon initiation of the survey, I recorded water temperature and began walking
clockwise around the lake. Upon the observation of an individual amphibian I recorded
its level of activity, which I classified as active, resting, or hiding. I observed for two
minutes to note any potential changes in this activity. In addition, I recorded its position
within the water column: top, mid, or bottom. After two minutes I placed a survey flag in
the substrate where the amphibian was first observed and measured the distance from the
shoreline. I set a maximum study radius of 5 feet from shoreline to accommodate
observation on foot where possible. This is consistent with literature that questions the
possible interference of observer with natural, uninterrupted behavior by wading (Rocha
et al., 2014). I recorded the substrate type, which I classified as: rock, woody debris,
sand, or mixed. After I completed one interval I waited until the subsequent observation
period (i.e., day or night, depending on arrival time and first observation period) to
conduct the second half of the survey for a given lake.
Night observations utilized additional equipment including a headlamp and dive
light. The underwater camera was utilized during the day and night to record activity
levels and aid in the identification of species upon later review away from the field site.
22
In instances where the shoreline was too steep to traverse on foot, a raft was utilized to
continue observation around the complete circumference of the lake. When the shoreline
was again accommodating to foot travel, I exited the raft and resumed the VES on foot.
Off site, the field data was transferred to an online spreadsheet for statistical analyses.
Statistical Analysis
I used JMP software (Version 14.0.1) to perform statistical tests on my collected
data, including 1-way ANOVA, multiple linear regression modeling and Chi-Squared
tests of independence.. For this study I used an alpha of 0.10 to consider results as
statistically significant, but take care to point out specific test results where the p-value is
between .05 and .10. I used a Bonferroni correction for multiple comparisons when
appropriate (Cabin and Mitchell, 2000). I generated graphs and charts to visually
represent the data using Google Sheets, Microsoft Excel and Microsoft Word.
23
Chapter 3: Results
Relationship between Fish and Amphibian Abundance
The hypothesized results from my experiment were that there would be a distinct
correlation between the presence fish and the tendency for amphibians to exhibit
nocturnal behavior. I recorded a total of 756 individual amphibians from 7 species of
frogs, salamanders and newts (Figure 2). Of the observations made, 239 occurred during
diurnal survey periods and 517 during nocturnal surveys [Figure 3]. When examining fish
presence impacts on abundance of amphibians across all lakes I began at a wide scale and
worked down towards a narrower field of view to assess significance. Beginning with all
amphibian abundance data, there was no difference by fish presence (lakes with vs. lakes
without fish, F1, 16=.588, p=0.454). I further broke abundance down into categories of
“salamander/newt” (F1,16=0.3645, p=0.555) and “frog” (F1,16=1.597, p=0.225), which also
generated non-significant results. Lastly, to examine fish impact on specific species
during different times of the day I examined salamander/newt abundance during the day
(F1,16=0.049, p=0.826), salamander/newt abundance during the night (F1,16=0.087,
p=0.772), frog abundance during the day (F1,16=4.408, p=0.052) and frog abundance
during the night (F1,16=0.573, p=0.460). Of these comparisons frog abundance during the
day was the comparison with any indication of a significant difference (albeit at 0.05 < p
< 0.10) between lakes with fish vs. lakes without fish, and this significant difference
disappears with a Bonferroni correction for multiple comparisons (corrected alpha =
0.10/7 = 0.0143).
24
Figure 2. Observations of individual amphibians during both survey periods.
The relative abundance of species observed varied between diurnal and nocturnal
surveys as well as between lakes with and without salmonids [Figure 4]. Relative
abundance was greatest in the West Cascades ecoregion (West mean=7.35; East mean =
5.46, North mean = 6.41, F6, 11=3.523, p = 0.056).
25
Figure 3. Species composition by observation period.
26
Figure 4. Aggregate Relative Abundance of All Amphibians Observed. Comparison
of observed abundance of frogs and salamanders in lakes with and without salmonids.
Relative abundance equals individual amphibians per hour of observation.
27
Figure 5. Relative Abundance By Species. Columns represent aggregate abundance
across observation periods with individual values for each day period noted.
28
Beyond Salmonid Influence
Using each lake’s recorded elevation, average lake water temperature during
sampling periods, size in hectares, ecoregion, and salmonid presence to predict overall
amphibian abundance did not result in a statistically significant regression model (Table
4). When using species richness as a response variable, elevation had a significant
negative influence (Table 4).
Table 4. Regression model results. Report of results with total abundance and total
richness as response variables.
Total Abundance
Total Richness
(SE)
p
(SE)
p
Elevation
-0.004 (.001)
0.391
-0.005 (0.002)
0.009*
Temperature
-0.076 (0.167)
0.657
-0.012 (0.055)
0.838
Size
-0.064 (0.161)
0.700
0.020 (0.530)
0.708
Ecoregion (N)
-0.009 (0.813)
0.992
-0.3618 (0.269)
0.205
Ecoregion (E)
-0.591 (0.874)
0.513
-0.064 (0.289)
0.829
Fish (n)_
0.266 (0.680)
0.703
-0.020 (0.931
0.931
Adjusted R2
-0.137
0.544
* - denotes statistically significant value
29
Examining other data collected during observation periods, I compared amphibian
activity level and substrate type during day and night separately. Activity level was
significantly associated with substrate type, both for daytime observations (χ²6 = 27.8, N
= 239, p = 0.0001) as well as at night (χ²6 = 36.9, N = 517, p < 0.001). Of these analyses,
hiding behavior on woody debris was higher than expected (expected 8.00, deviation
+9.99, cell χ² = 12.47). Active behavior on sand was also higher than expected (expected
16.82, deviation +5.18, cell χ² = 1.59). When examining night activity, active behavior on
rock was lower than expected 50.15, deviation -25.15, cell χ² = 12.61), and resting
behavior on rock was higher than expected 71.8, deviation +23.1, cell χ² = 7.45).
Lastly, I examined the correlation of the relative abundance at different times of
the day across different groups of amphibians (Table 5).
Table 5. Pearson’s correlation coefficient for Relative Abundance (Day and
Night) across Amphibian Types.
Total
Abundance
30
Sal/Newt
Day
Sal/Newt
Night
Frog
Day
Total
Abundance
1.00
Sal/Newt
Day
0.645
1.00
Sal/Newt
Night
0.597
0.407
1.00
Frog Day
0.441
-0.185
0.010
1.00
Frog Night
0.244
-0.171
-0.371
0.752
Frog
Night
1.00
Activity Coefficient
To examine the variability in activity from day to night with and without fish I
calculated an activity coefficient to quantify the difference. Active behavior received a
value of “1”, resting a “0” and hiding a “-1”, for the purpose of comparison and
relatability. I compared daytime and nighttime activity levels in each lake (Figure 6). . A
coefficient for both day and night was calculated in addition to the direction of movement
(positive or negative) towards more or less activity at night. Overall during diurnal
observations the dominant activity behavior was resting (n=139, 58.2%), followed by
active (n=67, 28.0%) and hiding (n=33, 13.8%). During nocturnal observations resting
was the dominant characteristic (n=288, 55.7%), followed by active (n=201, 38.9%) and
hiding (n=28, 5.4%).
31
Figure 6. Activity Coefficient Comparison By Lake Between Day and Night
32
Chapter 4: Discussion and Conclusion
Towards a Complete Understanding
All of the lakes exhibited increased amphibian detectability at night, regardless of
the presence or absence of fish. I hypothesized that there would be greater tendencies for
hiding behavior in lakes with established salmonid populations due to predator prey
dynamics, but this did not appear to be the case. The presence of fish did not play a
significant role in predicting the presence or abundance of amphibians during daytime or
nighttime hours. This finding challenges many assumptions that past researchers have
based their research upon. While fish alone did not predict presence or abundance, there
could be a synergistic effect at play with another variable that I did not collect during my
field study. Increased detectability at night, however points to a change in how
amphibians are utilizing available habitat. It is unknown whether available cover,
availability of deep-water refuge, or a combination of these and other habitat structural
elements play a role (Welborn, Skelly and Werner, 1996). Elevation was a significant
predictor of species richness (Table 4).
The Chi-Squared tests of independence that I conducted point to significant
correlations between activity level and substrate type and could play a greater role in the
predictability of amphibian behavior than predator prey dynamics alone. This changes in
expected versus observed behavior generally point towards more cautious behavior
during daytime hours and increased activity and exposure during nighttime observations.
This could potentially be linked to community composition as larval salamanders may
avoid larger salamanders and frogs. This correlation is represented during both daytime
33
and nighttime observation periods through Pearson’s correlation coefficient (Table 5),
which displays negative correlations of frog presence on salamander detectability.
Without a complete story as to why amphibian activity level fluctuates across
lakes, I calculated an activity coefficient to help bring a better understanding of lakes that
experience that greatest diel swings in activity level. Figure 6 displays the comparative
activity coefficients by lake during daytime and nighttime observations. The coefficient is
a measurement of the net activity per individual observed and can serve as a snapshot of
general activity level for a given time of day. The activity coefficient gains power as
observations are repeated over time to gain a more complete picture of amphibian
activity. By looking at the variation from day to night, extreme swings may warrant
additional study to understand the factors influencing the change. Land managers can
then begin to understand if these fluctuations come from direct predator threat, or another
biotic or abiotic factor within the lake environment. There is not currently a detection
system in place for managers to identify priority lakes with amphibian populations in the
greatest need. The snapshot provided through the activity coefficient could potentially
lead to more efficient management practices and use of funding.
While fish alone did not predict presence or abundance, there could be a
synergistic effect at play with another variable that I did not collect during my field study.
Some potential explanations from the literature for variation in amphibian activity
include factors not considered in this study, such as shoreline vegetation and habitat, nonaquatic predators, and other abiotic factors. Shoreline vegetation can create respite from
hot summer sun and can cool water temperature in the immediate vicinity of the shoreline
(Hossack et al., 2013; Warren and Buttner, 2008). This could be a factor in drawing
34
amphibians up from greater depths and become more detectable during daytime hours.
Other landscape disturbance such as avalanche activity can alter shoreline vegetation
structure and disturb surrounding trees and vegetation on a micro or macro scale.
Ultraviolet exposure at high elevations may also affect the use of cover by
amphibians throughout the day (McCaffery and Maxell, 2010). Thinner atmosphere at
higher elevations can cause prolonged exposure to potentially deadly UV-B rays (Hatch
and Blaustein, 2000). Larval stage frogs and amphibians may be especially vulnerable to
UV radiation, explaining the finding that larvae were most active at night. Winter
severity also plays a role in the level of exposure to ultraviolet radiation. More severe
winters provide a longer lasting snowpack that reflects UV light and protects frogs and
salamanders under the ice (Blaustein, 2000; McCaffery and Maxell, 2010). The rearing
timeframe for larvae begins in early summer and a longer lasting snowpack allows larvae
to get a head start on development without harmful UV rays penetrating the shallow
water that they typically occupy.
Non-aquatic predators deserve consideration as an influence on amphibian
behavior, due to the fact that many of the observed salamanders were in larval form and
an easy target for many predators. There are relatively few large predatory birds that
frequent alpine lakes, aside from ospreys and eagles. However, the spread of the
American dipper into higher elevations due to climate change can also be considered.
Normally seen as a riparian bird at lower elevations and even the ocean, this bird would
be a prime suspect for the role of diminishing the vulnerable larval stage salamanders and
tadpoles (Garwood et al., 2009). Another threat of predation that is frequently overlooked
as an explanation for behavior is that of inter and intra species predation by other
35
amphibians. Within the salamander community, cannibalism is common occurrence as
individuals eliminate competition for valuable food sources (Wildy et al., 1998; Walls,
1995). Lastly, the lunar phase could play a role in activity level due to the fact that when
there is increased lunar light, salmonids are able to continue to search for food in
conditions that normally offer safety for amphibians and other lake inhabitants.
Additional considerations for what could be seen as confident behavior would
need to be made when specifically analyzing individual species. The Northwestern
salamander displays combative behavior when threatened, as does the Rough-skinned
newt that is armed with a powerful neurotoxin. The increased activity displayed by these
species alone could be less of a holistic representation of amphibian species as a whole.
While none of the observed species are currently listed as threatened or endangered, there
could potentially be valuable data collected for future struggles, which are sure to come.
The umbrella approach to studying species favors some and is a disservice to others.
Ideally a study specific to each species would be conducted, but due to the widely
understudied nature of alpine species aquatic interactions and behavior, it isn’t plausible
in many circumstances.
The uniform tendency for nocturnal behavior is likely to be more than just a life
history characteristic for some of the amphibians that were observed in this study
(Kenison et al., 2016; Pearson and Goater, 2009). Another possible explanation for this is
that food sources have diel migration patterns. The tendency to find an amphibian at the
top or bottom may be a factor linked to its own hunting strategies and less of an action
based in self-preservation. A study conducted by Dolmen (1983) initially examined this
diel movement of species through the water column as prey and microhabitat benefits
36
changed with temperature and consumption needs. Another consideration linked to food
is competition with salmonid species for phytoplankton and zooplankton that inhabit the
lake (Parker et al., 2001).
Potential changes to analysis
If I were to continue or adapt this study I would narrow my scope to fewer lakes
and focus on the repeatability of the data to examine long-term trends that occur in a
given lake. With continued observation of one location, the external variables may
become more evident. This is one of the main challenges that researchers face in the field
today as studies may take several years for these variables to emerge. Lake basins are
very diverse and there are a lot of factors that can synergistically affect behavior of biota.
Additionally, by studying all amphibians present I was introducing the factor of potential
immigration and emigration of species between bodies of water. By solely focusing on
the larval stage salamanders and tadpoles, there would be fewer influences and the study
could purely examine the seasonally permanent inhabitants of the lake.
Future implications and next steps
Responsible management action is required to be able to meet the dual mission of
state agencies in Washington State to conserve native species and provide recreational
opportunity for citizens. The fish stocking practices of the past changed because of a
recognition that overpopulation was destructive to both native biota and a healthy fishery.
Without careful management of both goals, tensions are likely to rise between supports of
each side. Some value the right to recreate on public land while others value preservation
of species at nearly any cost. While there is clearly variation among alpine lakes, there
37
will also be varied need for intervention and active management by land managers
(Whittaker, Willis and Field, 2001). It is an implausible feat to move to one extreme or
another of the spectrum of involvement (Fellers et al, 2008). All future decisions will
likely be based on how to accommodate and consider the other side. Another danger of
this duality of purpose is creating a division among people that care for the outdoors and
need to support one another. With limited funding for projects and research another
benefit that is underutilized is the incorporation of citizen science and volunteer efforts.
Conclusion
The discussion on threatened and declining amphibian species in alpine lakes is
often accompanied by mention of introduced salmonid species. The logical step for many
is to place responsibility on the misdeeds of ill-managed alpine fisheries of the past.
While predation will occur when a predator is introduced into an ecosystem, the extent of
that blame is not fully understood and often involves other variables that may have
synergistic effects on the outcome. My work finds strong correlations between substrate
type and the behavior displayed by amphibians. The interactions that take place in aquatic
environments by nature are less observed than ecological interactions on land. Coupled
with the remote locations of these lakes, researcher accessibility is limited and therefore
little is known about the complete diel cycle of interactions and behaviors that take place.
My hope is to initiate conversations that examine how field research is conducted with a
larger emphasis on nocturnal observations. The aquatic environment, through substrate
and other physical cover, provides physical protection for many amphibian species and is
a necessary buffer in battling climate change and other changing biotic and abiotic
factors. As managers are able to understand the relationship that exists between
38
amphibians and their natural environment, there can be a more thoughtful allocation of
resources to assist populations that face these challenges without ability to adapt or
adjust. By understanding the diurnal and nocturnal patterns of movement within the
environment managers can better meet the challenge of conserving biodiversity and
managing established fisheries that aid in funding such actions.
39
Literature Cited
Bahls, P. (1992). The status of fish populations and management of high mountain lakes
in the western United States. Northwest Science, 66(3):183–193.
Blaustein, A. R., and Wake, D. B. (1990). Declining amphibian populations: a global
phenomenon?. Trends in Ecology & Evolution 5(7):203–204.
Cabin, R.J., and Mitchell, R.J. (2000). To Bonferroni or Not to Bonferroni: When and
How Are the Questions. Bulletin of the Ecological Society of America. 81(3): 246248.
Carpenter, S. R., and Kitchell, J. F. (1993). The trophic cascade in lakes. Cambridge
University Press, New York.
Carey, C., and Alexander, M.A. (2003). Climate change and amphibian declines: Is there
a link? Diversity and Distributions. 9(2): 111-121.
Case, M. J., Lawler, J. J., & Tomasevic, J. A. (2015). Relative sensitivity to climate
change of species in northwestern North America. Biological Conservation, 187,
127–133.
Dolmen, D. (1983). Diel Rhythms and Microhabitat Preference of the Newts Triturus vulgaris
and T. cristatus at the Northern Border of Their Distribution Area. Journal of
Herpetology, 17(1), 23-31.
Downen, M. R. (2004). North Cascades National Park High Lakes Fishery Management.
Washington Department of Fish and Wildlife, (May).
40
Drake, D. C., and Naiman, R. J. (2000). An evaluation of restoration efforts in fishless
lakes stocked with exotic trout. Conservation Biology 14(6):1807–1820.
Farner, D. S., and Kezer, J. (1953). Notes on the amphibians and reptiles of Crater Lake
National Park. The American Midland Naturalist 50(2):448–462.
Fellers, G. M., Pope, K. L., Stead, J. E., Koo, M. S., and Welsh, H. H. (2008). Turning
population trend monitoring into active conservation: can we save the Cascades
frog (Rana cascadae) in the Lassen region of California? Herpetological
Conservation Biology 3:28–39.
Garwood, J.M, Pope, K.L., Bourque, R.M., Larson, M.D. (2009). High Mountain Lakes
Provide a Seasonal Niche for Migrant American Dippers. The Wilson Journal of
Orithology. 121(3). 600-609.
Hoffman, R. L., Larson, G. L., and Brokes, B. J. (2003). Habitat segregation of
Ambystoma gracile and Ambystoma macrodactylum in mountain ponds and lakes,
Mount Rainier National Park, Washington, USA. Journal of Herpetology 37(1):
24–34.
Huang, C., & Sih, A. (1990). Experimental Studies on Behaviorally Mediated , Indirect
Interactions through a Shared Predator. Ecology, 71(4), 1515–1522.
IUCN. (2017) The IUCN Red List of Threatened Species. Version 2017-3.
http://www.iucnredlist.org. Downloaded on 30 April 2018.
IUCN, Conservation International, and NatureServe. (2004). Global Amphibian
41
Assessment. IUCN, Conservation International, and NatureServe, Washington
DC and Arlington, Virginia, USA.
Kats, L. B. and Ferrer, R. P. (2003). Alien predators and amphibian declines: review of
two decades of science and the transition to conservation. Diversity and
Distributions 9(2):99–110.
Knapp, R. A., Boiano, D. M., & Vredenburg, V. T. (2007). Removal of nonnative fish
results in population expansion of a declining amphibian (mountain yellow-legged
frog, Rana muscosa). Biological Conservation, 135(1), 11–20.
Knapp, R. A., Corn, P. S., and Schindler, D. E. (2001). The introduction of nonnative fish
into Wilderness Lakes: Good intentions, conflicting mandates, and unintended
consequences. Ecosystems 4(4):275–278.
Landres, P., Meyer, S., and Matthews, S. (2001). The Wilderness Act and fish stocking:
An overview of legislation, judicial interpretation, and agency implementation.
Ecosystems 4(4):287–295.
Leopold, A. S., Cain, S. A., Cottam, C. M., Gabrielson, I. N., and Kimball, T. L. (1963).
Wildlife management in the national parks. Transactions of the North American
Wildlife Conference (28):28–45.
Liss, W. J., Larson, G. L., & Hoffman, R. L. (2002). Ecological Impact of Introduced
Trout on Native Aquatic Communities in Mountain Lakes Phase III Final Report.
42
Nussbaum, R. A., Brodie, E. D., & Storm, R. M. (1983). Amphibians and reptiles of the
pacific northwest. Moscow, Idaho: University Press of Idaho.
Parmesan, C. (2006). Ecological and Evolutionary Responses to Recent Climate Change.
Annual Review of Ecology, Evolution, and Systematics. 37(12): 637-669.
Pilliod, D. S., and Peterson, C. R. (2001). Local and landscape effects of introduced trout
on amphibians in historically fishless watersheds. Ecosystems 4(4): 322–333.
Pister, E. P. (2001). Wilderness fish stocking: History and perspective. Ecosystems
4(4):279–286.
Pope, K. L. (2008). Assessing changes in amphibian population dynamics following
experimental manipulations of introduced fish. Conservation Biology, 22(6), 1572–
1581.
Rocha, C.F.D., Siqueira, C.C., Ariani, C.V., Vrcibradic, D., Guedes, D.M., Kiefer, M.C.,
Almeida-Gomes, M., Goyannes-Araujo, P., Borges-Junior, V., Van Sluys, M.
(2014). Differential success in sampling of Atlantic Forest amphibians among
different periods of the day. Brazilian Journal of Biology, 75(2), 261-267.
Sih, A., Kats, L. B., & Moore, R. D. (1992). Effects of Predatory Sunfish on the Density,
Drift, and Refuge Use of Stream Salamander Larvae. Ecology, 73(4), 1418–1430.
Tyler, T., Liss, W. J., Ganio, L. M., Larson, G. L., Hoffman, R., Deimling, E., &
Lomnicky, G. (1998a). Interaction between introduced trout and larval
salamanders (Ambystoma macrodactylum) in high-elevation lakes. Conservation
43
Biology 12(1):94–105.
Tyler, T. J., Liss, W. J., Hoffman, R. L., & Ganio, L. M. (1998b). Experimental analysis
of trout effects on survival, growth, and habitat use of two species of
ambystomatid salamanders. Journal of Herpetology 32(3):345–349.
Vredenburg, V. T., and Wake, D. B. (2004). Reversing introduced species effects:
Experimental removal of introduced fish leads to rapid recovery of a declining
frog. Proceedings of the National Academy of Sciences 101(20):7646–7650.
Welborn, G.A., Skelly, D.K., and Werner, E.E. (1996). Mechanisms creating community
structure across a freshwater habitat gradient. Annual Review of Ecology and
Systematics, (27):337–363.
Whittaker RJ, Willis KJ, and Field R. 2001. Scale and species richness: towards a
general, hierarchical theory of species diversity. Journal of Biogeography 28:
453–70.
44