ArticlePDF Available

A Surprising Discovery of American Pika Sites in the Northwestern Great Basin

Authors:
  • Nevada Department of Wildlife, Elko, Nevada United States

Abstract and Figures

American pikas (Ochotona princeps) are small mammals related to rabbits and are patchily distributed in broken rock landforms (e.g., talus and lava flows) of mountainous western North America (Smith and Weston 1990). Limits to thermoregulation combined with specific habitat requirements and metapopulation dynamics suggest that pikas ABSTRACT.—Although the American pika (Ochotona princeps) continues to receive attention due to documented declines and range retractions, particularly in the Great Basin, thorough range inventories have yet to be completed in many parts of the region. Here we report on recently discovered populations in northwestern Nevada in areas not suspected to support pika activity under current climate regimes. We describe 238 new locations (" sites ") with evidence of past or current occupancy by pikas which cluster into 31 locales, which we interpret as metapopulations or " demes, " in 15 distinct mountain ranges or geographic areas. We documented twice as many relict sites (sites with evidence of former pika occupancy) as currently occupied sites, a result that supports previous observations of local range retraction and site losses within the pika's range. In the overall site data, median elevation and water year precipitation were higher and minimum and maximum July temperatures were lower for occupied sites than for relict sites. This pattern was repeated in most, but not all, of the 7 mountain ranges where both occupied and relict sites were found. Occupied sites were more likely to be found in areas with a narrow range of water year precipitation, in cooler climates, and on more mesic-facing aspects, but many of these environmental descriptors also describe relict sites. The apparent extirpa-tion of pikas from the range with the highest elevation and lowest temperatures (Black Rock Range) and the continued persistence in some of the lowest and hottest areas of our survey (Home Camp Range) are particularly noteworthy. Since pikas were known from only a handful of early 20th century records in the area, these surveys greatly expand our understanding of both current and historic pika distributions in the northwestern Great Basin and shed light on patterns of pika persistence and extirpation in a region typified by harsher climates respective to other areas with extant pika populations. Furthermore, our results emphasize the importance of conducting spatially extensive fieldwork to better understand site extirpations and species range retractions. RESUMEN.—Aunque la pica americana (Ochotona princeps) continúa recibiendo atención debido a su declive docu-mentado y a su gama de retracciones, en particular en la Gran Cuenca (Great Basin), aún resta completar los inventarios sobre su rango de distribución en muchas partes de la región. Aquí reportamos poblaciones recientemente descubiertas en el noroeste de Nevada, en áreas en las que no se pensaba que las picas estuviesen activas bajo las condiciones climá-ticas actuales. Describimos 238 nuevos lugares (" sitios ") con evidencia de ocupación pasada o actual de las picas que se agrupan en 31 sitios, los cuales interpretamos como metapoblaciones o " demes, " en 15 montañas o áreas geográficas distintas. Documentamos el doble de sitios relicto (sitios que justifican la ocupación histórica de picas) que los sitios habitados actualmente; respaldando observaciones previas de retracción del rango local y la pérdida de sitios dentro del rango de las picas. Al observar los datos generales del lugar, tanto la elevación media como las precipitaciones anuales fueron más altas, y las temperaturas mínimas y máximas en julio fueron más bajas en los sitios habitados actualmente que en los sitios relicto. Este patrón se repitió en la mayoría, pero no en todos, de los siete sitios montañosos en donde se encontraron sitios ocupados y sitios relicto de pica. Los sitios ocupados fueron más propensos a encontrarse entre el límite inferior y el límite superior de precipitaciones anuales, en climas más fríos y en condiciones más mésicas, pero muchas de estas características ambientales también describen sitios relicto. Es particularmente significativa la aparente desaparición de las picas de las montañas con mayor elevación y con temperaturas más bajas (cordillera Black Rock), y su persistencia en algunas de las áreas más bajas y más cálidas de nuestra investigación (cordillera Home Camp). Dado que las picas sólo eran conocidas por unos cuantos registros pertenecientes a principios del siglo XX en el área, estos monitoreos amplían enormemente nuestra comprensión tanto de su distribución actual como de su distribución histó-rica en la Gran Cuenca del noroeste y clarifican sobre los patrones de persistencia y de desaparición de las picas en una región tipificada por climas más severos con respecto a otras áreas con poblaciones de pica. Además, nuestros resultados hacen hincapié en la importancia de llevar a cabo extensas investigaciones de campo para comprender mejor las desapa-riciones en ciertos sitios y las retracciones de estas especies.
Content may be subject to copyright.
American pikas (Ochotona princeps) are
small mammals related to rabbits and are
patchily distributed in broken rock landforms
(e.g., talus and lava flows) of mountainous
western North America (Smith and Weston
1990). Limits to thermoregulation combined
with specific habitat requirements and
metapopulation dynamics suggest that pikas
Western North American Naturalist 77(2), © 2017, pp. 252–268
A SURPRISING DISCOVERY OF AMERICAN PIKA
SITES IN THE NORTHWESTERN GREAT BASIN
Mackenzie R. Jeffress1, K. Jane Van Gunst2, and Constance I. Millar3
ABSTRACT.—Although the American pika (Ochotona princeps) continues to receive attention due to documented
declines and range retractions, particularly in the Great Basin, thorough range inventories have yet to be completed in
many parts of the region. Here we report on recently discovered populations in northwestern Nevada in areas not
suspected to support pika activity under current climate regimes. We describe 238 new locations (“sites”) with evidence
of past or current occupancy by pikas which cluster into 31 locales, which we interpret as metapopulations or “demes,”
in 15 distinct mountain ranges or geographic areas. We documented twice as many relict sites (sites with evidence of
former pika occupancy) as currently occupied sites, a result that supports previous observations of local range retraction
and site losses within the pika’s range. In the overall site data, median elevation and water year precipitation were
higher and minimum and maximum July temperatures were lower for occupied sites than for relict sites. This pattern
was repeated in most, but not all, of the 7 mountain ranges where both occupied and relict sites were found. Occupied
sites were more likely to be found in areas with a narrow range of water year precipitation, in cooler climates, and on
more mesic-facing aspects, but many of these environmental descriptors also describe relict sites. The apparent extirpa-
tion of pikas from the range with the highest elevation and lowest temperatures (Black Rock Range) and the continued
persistence in some of the lowest and hottest areas of our survey (Home Camp Range) are particularly noteworthy.
Since pikas were known from only a handful of early 20th century records in the area, these surveys greatly expand our
understanding of both current and historic pika distributions in the northwestern Great Basin and shed light on patterns
of pika persistence and extirpation in a region typified by harsher climates respective to other areas with extant pika
populations. Furthermore, our results emphasize the importance of conducting spatially extensive fieldwork to better
understand site extirpations and species range retractions.
RESUMEN.—Aunque la pica americana (Ochotona princeps) continúa recibiendo atención debido a su declive docu-
mentado y a su gama de retracciones, en particular en la Gran Cuenca (Great Basin), aún resta completar los inventarios
sobre su rango de distribución en muchas partes de la región. Aquí reportamos poblaciones recientemente descubiertas
en el noroeste de Nevada, en áreas en las que no se pensaba que las picas estuviesen activas bajo las condiciones climá-
ticas actuales. Describimos 238 nuevos lugares (“sitios”) con evidencia de ocupación pasada o actual de las picas que se
agrupan en 31 sitios, los cuales interpretamos como metapoblaciones o “demes,” en 15 montañas o áreas geográficas
distintas. Documentamos el doble de sitios relicto (sitios que justifican la ocupación histórica de picas) que los sitios
habitados actualmente; respaldando observaciones previas de retracción del rango local y la pérdida de sitios dentro del
rango de las picas. Al observar los datos generales del lugar, tanto la elevación media como las precipitaciones anuales
fueron más altas, y las temperaturas mínimas y máximas en julio fueron más bajas en los sitios habitados actualmente
que en los sitios relicto. Este patrón se repitió en la mayoría, pero no en todos, de los siete sitios montañosos en donde
se encontraron sitios ocupados y sitios relicto de pica. Los sitios ocupados fueron más propensos a encontrarse entre el
límite inferior y el límite superior de precipitaciones anuales, en climas más fríos y en condiciones más mésicas, pero
muchas de estas características ambientales también describen sitios relicto. Es particularmente significativa la aparente
desaparición de las picas de las montañas con mayor elevación y con temperaturas más bajas (cordillera Black Rock), y
su persistencia en algunas de las áreas más bajas y más cálidas de nuestra investigación (cordillera Home Camp). Dado
que las picas sólo eran conocidas por unos cuantos registros pertenecientes a principios del siglo XX en el área, estos
monitoreos amplían enormemente nuestra comprensión tanto de su distribución actual como de su distribución histó-
rica en la Gran Cuenca del noroeste y clarifican sobre los patrones de persistencia y de desaparición de las picas en una
región tipificada por climas más severos con respecto a otras áreas con poblaciones de pica. Además, nuestros resultados
hacen hincapié en la importancia de llevar a cabo extensas investigaciones de campo para comprender mejor las desapa-
riciones en ciertos sitios y las retracciones de estas especies.
1Nevada Department of Wildlife, 60 Youth Center Road, Elko, NV 89801. E-mail: mrjeffress@ndow.org
2Nevada Department of Wildlife, 815 E. Fourth St., Winnemucca, NV 89445.
3USDA Forest Service, Pacific Southwest Research Station, Albany, CA 94710.
252
could be vulnerable to climate change. For
example, pikas appear to be in decline in some
low-elevation locations of the Great Basin, and
potential stressors, including decreased snow-
pack and warming summer months, are impli-
cated in these declines (Beever et al. 2003,
2010, 2011). However, pikas continue to be
found outside of described bioclimatic
envelopes, and the species does not often fit
well within species niche models (Martínez-
Meyer et al. 2004). Previously undocumented
populations have recently been found in new
areas, including lower-elevation, non-talus,
and anthropogenic sites (Beever et al. 2008,
Simpson 2009, Rodhouse et al. 2010, Collins
and Bauman 2012, Millar et al. 2013, Varner
and Dearing 2014a, Shinderman 2015). Fur-
thermore, for several populations across the
pika’s range, species distribution models and
projections that account for current and antici-
pated climate change and habitat connectivity
suggest that the impacts on pikas will vary
by place ( Jeffress et al. 2013, Schwalm et al.
2016). Resilience also appears to relate to
pikas’ habitat dependence on broken rock
landforms, which develop thermal regimes at
least partially decoupled from surface temper-
atures and thereby mitigate seasonal tempera-
ture stresses on pikas (e.g., Varner and Dearing
2014b, Hall et al. 2016, Millar et al. 2016).
Therefore, generalizations about pika extirpa-
tions at regional and species-wide scales might
be inaccurate, and place-based assessments
relying on extensive surveys over multiple years
are needed (Jeffress et al. 2013).
Pikas are a Nevada Species of Conservation
Priority and the Nevada Wildlife Action Plan
calls for a thorough statewide inventory of the
species (Wildlife Action Plan Team 2012).
Though much of Nevada has been surveyed for
pikas, little was known about pika distribution,
potential connectivity among sites, and site
characteristics of both currently occupied and
seemingly relict sites across the northwestern
portion of the Great Basin, making that region
particularly important for conservation of the
American pika. The area has recently docu-
mented extirpated sites, as well as newly dis-
covered occupied sites. Early 1900s naturalist
E.R. Hall (1946) captured pikas in the area of
Barrel Springs (near the California border),
near Summit Lake, and in the Pine Forest
Mountains. However, resurveys of these sites,
starting in the 1990s, have documented only
old evidence of pikas, suggesting extirpation
(Beever et al. 2011 [App. V], Millar et al. 2014).
Conversely, Collins and Bauman (2012) docu-
mented newly discovered populations on the
Sheldon National Wildlife Refuge (NWR), in
areas just north and west of Hall’s Summit
Lake site and in nearby southern Oregon’s Hart
Mountains, and Beever et al. (2008) found a
small extant population of pikas in Boulder
Canyon in the Hays Canyon Range, Nevada
(Fig. 1). Additionally, Millar et al. (2013) docu-
mented nearby populations to the west in
northern Great Basin portions of California,
above the Madeline Plain, and in the South
Warner Mountains. Interestingly, Hall (1946)
noted that pikas “are much more widely dis-
tributed in northwestern Nevada than our
records indicate.” Yet, aside from these few
areas mentioned, this region has not been
extensively surveyed until now. The northwest-
ern portion of the Great Basin is atypical com-
pared to the mountainous (“alpine”) landscapes
associated with much of the species’ range. The
area is at a lower elevation that is warmer and
more xeric than expected given its latitude,
providing an opportunity to understand how
a species vulnerable to heat stress persists in a
warmer and drier environment. Furthermore,
the region is at the edge of the range for the
Sierra Nevada subspecies (O. p. schisticeps;
Hafner and Smith 2010). In the context of
under standing climate change impacts, examin-
ing populations at the range periphery provides
valuable insight into a species’ adaptive poten-
tial, especially for species like pika, which are
dispersal-limited habitat specialists, with low-
ered ability to avoid or adapt to changes in their
environment (Thomas et al. 2001, Parmesan
2006, Morueta-Holme et al. 2010).
The goal of our study was to conduct a
more complete and spatially extensive survey
of potential pika habitat in northwestern
Nevada to better understand the current and
historic distribution in the area. Here, we
document numerous pika sites over a 2-year
survey period and present our results at mul-
tiple scales, from the smaller “site” level to a
snapshot of potentially connected population
centers (or “demes”) to larger mountain ranges.
At these scales, we describe climatic and envi-
ronmental variables in multiple occupancy
categories to provide insight into factors asso-
ciated with pika persistence and extirpation
within the region. Our hope is that our survey
2017] RECENTLY DISCOVERED GREAT BASIN PIKA SITES 253
results and insights will refine characteriza-
tions of the pika’s ecological niche and further
a comprehensive understanding of population
dynamics and associated conservation needs
for American pikas in the Great Basin.
METHODS
Study Area
Surveys were restricted to northwestern
Nevada but the study area had no other precon-
ceived boundaries; rather, areas were surveyed
opportunistically, as access and the availability
of personnel allowed. Mountain ranges sur-
veyed included the Black Rock Range, Hays
Canyon Range, Home Camp Range, Little
High Rock Mountains, Massacre Range, Mos-
quito Mountains, Painted Point Range, and
Nut Mountain, as well as upland regions that
do not fall into named mountain ranges, such
as the Crooks Lake Hills. These upland areas
are part of the Black Rock Plateau and sites
were dominated by basaltic rock. Upland
vegetation primarily consisted of mountain
big sagebrush (Artemisia tridentata vaseyana),
Wyoming big sagebrush (Artemisia tridentata
wyomingensis), and bitterbrush (Purshia tri-
dentata), with a perennial grass and forb
understory. Some of our sites also had western
juniper ( Juniperus occidentalis) and cheat-
grass (Bromus tectorum) in the vicinity. All
sites surveyed were on public land adminis-
tered by the Department of the Interior
Bureau of Land Management.
Surveys
We visited potentially suitable locations
(i.e., talus, broken lava, and coarse rock matrix
254 WESTERN NORTH AMERICAN NATURALIST [Volume 77
Fig. 1. Sites in northwestern Nevada documented prior to 2014 as American pika habitat (noted as occupied or relict at
the time of observation): (a) Barrel Springs/Fort Bidwell, occupied in 1941 (Hall 1946) but since extirpated (i.e., relict;
Beever et al. 2003); (b) Boulder Canyon in the Hays Canyon Range first documented as occupied in 2002 (Beever et al.
2008); (c) mix of occupied and recently extirpated sites, first documented in 2009, on and near the Sheldon National
Wildlife Refuge (Collins and Bauman 2012); (d) N Summit Lake, occupied in 1936 (Hall 1946) but since extirpated (Beever
et al. 2003); and (e) Duffer Peak in the Pine Forest, occupied in 1935 (Hall 1946) but since extirpated (Beever et al. 2003).
landforms) that had been identified via aerial
imagery or preliminary ground inspection,
and assessed these for pika presence in 2014
and 2015. We surveyed a single potential
habitat patch for at least 30 person-minutes
and documented evidence of current or past
pika occurrence. We also noted when a site
was surveyed but no evidence of pika pres-
ence was found. Survey sites and evidence
encountered were often further documented
by collecting pellets and taking photos. As
pika sign tends to be clustered in small areas,
we combined all detection records within 50
m of each other into a single “site.” Therefore,
a site could contain one to several pika territo-
ries (Smith and Weston 1990) and could be
considered representative of a habitat area or
patch. To evaluate pika occurrence at a coarser
scale, we combined detections that were
within 3 km of each other (which is greater
than the cited average maximum dispersal
distance of 2 km; Smith and Weston 1990) and
called these clusters “demes” (as in Millar et
al. 2013). These demes are meant to present a
snapshot of potentially connected populations.
Finally, we examined results by mountain
ranges and distinct geographic areas defined
by Charlet (unpublished data).
Occupied sites were identified by one or
more detections of hay piles containing cur-
rent year vegetation, pika calls, and pika sight-
ings. Possibly occupied sites were identified
by the detection of questionably fresh hay
(limited green material) or fresh-appearing
scat (green and perched but not abundant or
associated with distinctly fresh hay). Relict
sites were identified by one or more detec-
tions of hay pile remnants (brown and decom-
posing vegetation) and old scat (brown or
moldy, dry, and no longer perched). Sites
where only pika-like urine staining on rocks
was found were described as possibly relict;
however, there is currently no way to confirm
these stains as pika-created. Areas surveyed
that had neither current nor historic evidence
of pika presence were also recorded, with
coordinates taken at the center of the area
searched and annotated at the same scales as
described above. With this single-visit rapid
assessment survey methodology, the potential
for false negative results can be high, particu-
larly when considering possible detectability
issues (e.g., sites that might have had old sign
that we were unable to locate or seasonal
effects on detection). However, previous stud-
ies have found high detection probabilities
(0.90) using these survey methodologies
(Beever et al. 2008, 2010, Rodhouse et al.
2010, Erb et al. 2011, Moyer-Horner et al. 2012,
Ray et al. 2016), and given our use of multiple
sources of conspicuous and persistent sign and
trained observers, we used the detection of
current pika presence to indicate that a site
was occupied (as was done in Jeffress et al.
2013). This allowed us to focus our resources
on increasing survey coverage. Nevertheless,
we acknowledge that multiple surveys of an
area are needed to confirm true extirpation or
absence and encourage the classification of
relict and no-detection sites to be interpreted
with a level of prudence.
Analysis
To help describe our sites, we used several
environmental variables (e.g., elevation, slope,
and aspect) extracted from a 10-m Digital
Elevation Model (DEM) and a 30-m pixel
landform dataset that derived landform posi-
tion classes based on (1) a refined and adjusted
Topographic Relative Moisture Index (TRMI;
accounting for slope position and angle, slope
shape, and slope aspect), (2) landform position
from valley flats to steep slopes, and (3) climatic
zone stratification based on the Soil Adjusted
Vegetation Index, a measure derived from
remote sensing (see table 1 in Manis et al.
[2001] for the slope limits and TRMI values
that define each of the landform position
classes). Aspect was binned into one of eight
45°ordinal classes (e.g., NE, S, SW). We also
calculated total daily solar radiation (both
diffuse and direct) using the solar radiation
tool in ArcMap (Fu and Rich 1999). The solar
analyst tool uses aspect and slope from a 30-m
DEM to derive a measure of clear-sky radiation
over the single-day period of 15 August,
thereby representing the total amount of insola-
tion (indicating heat load from direct sunlight)
and total solar irradiance during peak summer
conditions (Fu and Rich 1999, 2002).
We present our results by occupancy status
and at multiple scales including site, deme,
and mountain range. In addition to the envi-
ronmental variables, we examined several
candidate climatic variables including maxi-
mum and minimum January temperature as a
measure of wintertime stress and maximum
and minimum July temperature as a measure
2017] RECENTLY DISCOVERED GREAT BASIN PIKA SITES 255
of summertime stress (Beever et al. 2011,
Wilkening et al. 2011). The 800-m PRISM
normals from 1981 to 2010 (Daly et al. 2008)
were used to derive temperature variables (as
done in Schwalm et al. 2016). We derived
average water year precipitation over the
1981–2010 period from gridded monthly
PRISM precipitation data. Water year pre-
cipitation is widely used to describe climate
and changes in climatic trends in the interior
western United States, such as changes in
snowpack, snowmelt, seasonal shifts in vegeta-
tion, and vegetation growing period (Regonda
et al. 2005, Wilkening et al. 2011). Water pre-
cipitation is derived from the October through
September period and captures both the season
of high precipitation input (late fall to early
spring), as well as dry periods (late spring to
early fall).
Because many of these variables are highly
correlated (e.g., elevation is highly correlated
with precipitation and with winter and summer
maximum temperatures), we restricted inclu-
sion of variables to those that have Pearson’s
product-moment correlation scores between
0.7 and 0.7. For example, for our analysis we
only used the median maximum and minimum
July temperature with the knowledge that
these variables are highly correlated with
analogous variables from January. We pre-
sented the results of the landform variable in
order to describe the overall environmental
context for these sites, and we included analy-
sis of elevation due to its common use in other
studies. However, neither landform nor eleva-
tion was used in our regression tree analyses
because of their correlation with other envi-
ronmental and climate variables. Because of
the novel settings of many of these locations, it
is unclear exactly how these variables influence
pika occurrence and how that influence changes
depending on differences in macro- and
microsites, differences in the talus structure,
and differences in pika population demograph-
ics. Nonetheless, we present these variables
because they are important in explaining pika
occurrence elsewhere and it seems clear that
climatic stress is important in describing both
pika occupancy and extirpation in many areas
(Beever et al. 2003, Wilkening et al. 2011).
We used a classification tree derived in the
R statistical program (R Development Core
Team 2016) using the “rpart” and “rattle”
libraries to describe the hierarchical nature of
the environmental variables and climate
variables derived at our study sites. A classifi-
cation tree is a type of machine-learning tool
that makes hierarchical binary splits in a
dataset based on a drop in deviance (De’ath
and Fabricius 2000). We restricted the tree to
those sites where definitive sign of current or
historic pika occupancy was found. We used
water year precipitation, both July tempera-
ture variables, solar radiation, and the binned
aspect variable for this analysis. The tree was
not pruned but several controls were used in
fitting the tree. These controls were as follows:
(1) 20 observations needed to exist before a
split was attempted, (2) at least 10 observa-
tions needed to exist in any terminal node,
and (3) each split needed to increase the
overall fit by 0.05 (e.g., a complexity parame-
ter). These limitations were imposed to ensure
parsimony by identifying the most important
splits in our dataset.
RESULTS
Surveys occurred in September 2014 and
March through October 2015, with most
2015 surveys occurring in August or later.
Raw detections of pikas totaled 977, and an
additional 21 sites were surveyed for a mini-
mum of 30 person-minutes, but no evidence
of pikas was found. The raw detections were
clustered into 238 sites, constituting 31
demes in 15 mountain ranges and distinct
geographic areas (Fig. 2).
Sites
Fifty-six (23.5%) of the sites were occupied,
12 (5.0%) were possibly occupied, 117 (49.2%)
were relict, 32 were possibly relict (13.4%),
and 21 (8.8%) had no detected evidence of
pika occurrence. No currently occupied or
possibly occupied sites fell within habitat
predicted to be occupied by the GAP species
distribution model (USGS Gap Analysis Pro-
gram 2013). Over our study area, only the
relict Black Rock Range sites fell within GAP-
predicted occupied habitat (n= 4). Most sites
were located on gently sloping ridges and hills,
moderately moist steep slopes, and moderately
dry slopes (Fig. 3). The highest percentage of
occupied sites (42.8%) were on moderately
moist steep slopes, while the highest percent-
age of relict sites (36.7%) were on moderately
dry slopes and moderately moist steep slopes
256 WESTERN NORTH AMERICAN NATURALIST [Volume 77
(35.9%). We found relict-only sites on almost
all landform types, while over 90% of occu-
pied sites were found on 3 landform types
characterized by moderate positions on either
slope, moisture gradients, or both. Relict sites
can be found in those places as well but they
2017] RECENTLY DISCOVERED GREAT BASIN PIKA SITES 257
Fig. 2. Map of 2014–2015 pika survey sites and demes, indicated by occupancy status, and associated mountain ranges
in northwestern Nevada. Panel A: The northwest portion of the study area including the Mosquito Mountains, Crooks
Lake Hills, northern Hays Canyon Range, and Coleman and Massacre Rim. Panel B: The southwestern portion of the study
area, from the Hays Canyon Range (furthest west) to the High Rock Canyon Hills and Badger Mountains (to the east).
Panel C: The southeastern portion of the study area encompassing the Black Rock Range. A reference map for the
panels is provided in the upper left and titled Overview.
Fig 3. Percent of sites binned by American pika occupancy status and landform.
Percentage of sites in each
occupancy category
valley flats toe slopes,
bottoms, and
swales
gently sloping
ridges and
hills
nearly level
plateaus or
terrace
very moist
steep slopes
moderately
moist steep
slopes
moderately
dry slopes
very dry
steep slopes
Occupied
Possibly occupied
Relict
Possibly relict
No detections
Landform types
45
40
35
30
25
20
15
10
5
0
dominate the drier landform types. Thirty-nine
percent of occupied sites were on west-facing
slopes and 16% were on northwest-facing slopes
(Fig. 4). Forty percent of currently occupied
sites were distributed over northerly aspects
(north- and northeast-facing together account
for 25%), and approximately 15% of occupied
areas were found on south- and east-facing
slopes. The highest percentage of relict sites
were on southwest-facing slopes (24.7%), fol-
lowed by west-facing (22.2%) and northwest-
facing (18.8%) slopes. However, occupied and
relict sites were found throughout all aspect
categories. Median elevation was highest for
occupied sites (1920 m) and lowest for sites
where no current or historic sign was detected
(1828 m). Median elevation of definitively
occupied sites was 29 m higher than that of
relict sites.
Median water year precipitation was high-
est for occupied sites (311.3 mm) and lowest
for possibly relict sites (297.3 mm). For relict
sites, median water year precipitation was 14.2
mm less than for currently occupied sites;
although, as with elevation, relict sites in the
Black Rock Range had the highest water year
precipitation of all sites in the survey region.
Currently occupied sites had the lowest
median maximum July temperature (26.5 °C)
and lowest median minimum July temperature
(8.3 °C) of all occupancy status categories in
our survey area (Fig. 5). Sites without evidence
of current or historic pika occupation were the
warmest in our survey area, having the high -
est maximum July temperature (27.0 °C) and
highest minimum July temperature (9.0 °C).
Median maximum and minimum July tempera-
tures of possibly occupied sites were about
0.2 °C warmer than occupied sites. Outliers of
both high and low temperatures were noted
at several relict sites. Cold minimum July
temperatures of <7.1 °C were found at occu-
pied and relict sites in the Home Camp
Range and the Hays Canyon Range, at one
occupied site in the Massacre Range, and at
relict sites in the Black Rock Range. Some,
but not all, of these sites were also colder as
defined by a median maximum July tempera-
ture <25.5 °C. Three of the relict Black Rock
Range sites were also the hottest sites of the
survey area using minimum July temperature
as the measure. Median total radiation was
highest for sites where pika sign was not
found (7125 watt-hours [Wh]/m2) and lowest
for possibly occupied sites (5076 Wh/m2).
Definitively occupied sites received much
more solar radiation (median = 6927 Wh/m2)
than relict sites (median = 6138 Wh/m2).
Demes and Mountain Ranges
Eight (25.8%) of the demes were occupied,
2 (6.5%) were possibly occupied, 14 (45.2%)
were relict, 3 (9.7%) were possibly relict, and
4 (12.9%) had no detected evidence of pika
occurrence. Overall, the occupied demes were
found more in the western part of our study
area (Fig. 2). Five demes contained only one
site (7.3 km2), whereas the largest deme
(Carter Reservoir, 74.0 km2) contained 32 sites,
was occupied, and was in close proximity to
2 other occupied demes. Our results indicate
4 population centers in our survey region:
258 WESTERN NORTH AMERICAN NATURALIST [Volume 77
Fig 4. Number of sites binned by American pika occupancy status and aspect.
Occupied
Possibly occupied
Relict
Possibly relict
No detections
N NE E SE S SW W NW
Aspect
Percentage of sites in each
occupancy category
45
40
35
30
25
20
15
10
5
0
(1) Crooks Lake, Carter Reservoir, and South-
ern Vya Rim (Fig. 6) demes (96 sites, 16.7%
occupied), (2) Boulder Canyon and southern
Hays Canyon demes (24 sites, 54.2% occu-
pied), (3) Massacre Ranch and Grassy Canyon
(Fig. 7), and Nut Mountain demes (45 sites,
56.8% occupied), and (4) the Massacre Rim
deme (Fig. 8; 3 sites, 66.7% occupied).
Although we attempted to survey as much
of the study area as possible, we cannot go
2017] RECENTLY DISCOVERED GREAT BASIN PIKA SITES 259
Fig. 5. Boxplots of median maximum (top) and minimum (bottom) July temperature by site occupancy status.
Occupied Possibly occupied Relict Possibly relict No detection
Site Occupancy
Minimum July Temperature (ºC)
30
28
26
24
Minimum July Temperature (ºC)
Occupied Possibly occupied Relict Possibly relict No detection
Site Occupancy
11
10
9
8
7
beyond speculation about the potential for
habitat and connectivity across unsurveyed
areas. Our surveys indicate that the extant
Boulder Canyon and Grassy Canyon demes
may once have been connected via the Pinto
Peak deme, but this area is now likely extir-
pated. It is also possible that the Boulder
Canyon deme was once connected throughout
260 WESTERN NORTH AMERICAN NATURALIST [Volume 77
Fig. 6. Vya Rim looking toward the Carter Reservoir deme, which includes the Mosquito Rim area. The broken rim-
rock in these demes contains a mix of occupied and relict sites. The habitat is sagebrush-steppe and scattered western
juniper, and the view is north. This site had many hay piles and fresh scat and was one of the most densly occupied sites
that we visited.
Fig 7. The Grassy Canyon deme in the Massacre Range. Several patches of pika-occupied basalt outcrops are visible
here and continue down into the canyon. Sagebrush steppe dominates the deme, and the view here is to the southwest.
the Hays Canyon Range north to the Carter
Reservoir deme, but surveys that documented
relict demes inbetween, such as at Forty-niner
Mountain, suggest another possible and recent
break in connectivity. The Massacre Rim
deme is likely much larger than the area we
were able to document due to limited survey
time (see Fig. 8 for photo of habitat that is rep-
resentative of unsurveyed rim areas to north
and south), but our results do indicate that the
Massacre Rim Wilderness Study Area to the
southeast of the rim is now likely extirpated.
The potential for connectivity of these demes
to the previously documented Sheldon sites
(Collins and Bauman 2012, Castillo et al.
2016) needs further exploration. Furthermore,
the relict Black Rock demes (Fig. 9) were likely
at one point connected to each other and pos-
sibly to the early 20th century–occupied Pine
Forest Range to the north and the currently
occupied Sheldon NWR to the west.
Our surveys covered 15 mountain ranges
and distinct geographic areas in northwestern
Nevada, 8 of which had 10 survey sites per
range. We found definitive sign of relict or
current occupation in 11 of these ranges, with
no conclusive evidence of historic occupation
in the remaining 4 ranges (i.e., Badger Moun-
tains, Little High Rock Mountains, Mosquito
Mountains Piedmont North, and the Yellow
Hills). Only 7 of the 11 ranges had sites with
current occupation, and in only 2 of those
ranges, the Home Camp Range and Massacre
Range, more sites with current occupation
were found than relict sites. To examine dif-
ferences between occupied and relict sites,
we summarize elevation and maximum July
temperature in these 7 ranges.
Sites in the Black Rock Range and on Little
Hat Mountain in the Hays Canyon Range
were the highest-elevation survey sites, and
only historical evidence was found at these
sites (Fig. 10). Median elevation of occupied
sites was higher than that of relict sites in 3 of
the 7 ranges and was roughly equal between
the 2 occupancy categories in the Hays
Canyon Range. In 3 of the 7 ranges (Home
Camp Range, Massacre Range, and Painted
Point Range), median elevation of relict sites
was higher than that of occupied sites. The
Home Camp Range had the lowest median
elevation and Painted Point the second lowest
of the 7 ranges; here, pikas were more likely
to be found at the lower elevational limits
(but see comments below about potential
confounding survey effort in the Painted Point
Range). In the Hays Canyon Range, the highest
median elevation in our study area, median
2017] RECENTLY DISCOVERED GREAT BASIN PIKA SITES 261
Fig. 8. The Massacre Rim deme in the Painted Point Range. This area was occupied throughout, and although not
extensively surveyed as part of this study, the rim has similar and likely connected habitat for several miles to the north
and south.
262 WESTERN NORTH AMERICAN NATURALIST [Volume 77
Fig. 9. A deme in the Black Rock Range. Although this was one of the highest elevation areas surveyed, only relict
sites were documented in the demes throughout the range. The view is to the south toward the Colman River drainage.
Fig. 10. Elevational boxplots for the 7 ranges (Crooks Lake Hills, Hays Canyon Range, Home Camp Range, Massacre
Range, Mosquito Mountain, Nut Mountain, and Painted Point Range) where both occupied and relict sites were
detected.
Occupied Relict Occupied Relict Occupied Relict
Occupied Relict Occupied Relict Occupied Relict
Occupied Relict
Site Occupancy
2000
1950
1900
1850
1800
2200
2100
2000
1900
2200
2100
2000
2100
2000
1900
1900
1850
1800
1750
1700
1950
1900
1850
1800
1900
1850
1800
elevation of occupied sites was just slightly
higher than that of relict sites.
Occupied sites had higher median July
temperatures in the Home Camp Range and
Painted Point Range than relict sites. For all
other ranges, occupied sites were cooler and
characterized by lower maximum July tem-
peratures than relict sites. At Massacre Range,
although occupied sites tended to be at lower
elevations than relict sites, they were often
in cooler microsites. The same pattern was
reflected in the minimum July temperatures.
Regardless of whether occupied sites trend
warmer or cooler than relict sites, differences
in median temperature (both July minimum
and maximum) between occupied and relict
sites is <1 °C in almost all cases (with mini-
mum July temperature in the Hays Canyon
Range being the exception). Our results
could be due to an inability of the large
800-m PRISM cell size to accurately reflect
conditions in small patches, or it could sug-
gest that minor yet persistent temperature
differences are important in site suitability
for pikas. In the 2 ranges (Home Camp and
Painted Point) that had the lowest median
elevation and that were the hottest ranges in
our study region, occupied sites were more
likely to be found at the lower and hotter
parts of the range. However, the pika popula-
tion on Painted Point covers Massacre Rim, a
large area that we suspect was occupied over
a greater extent than where we surveyed.
Because of this area’s close proximity to the
Sheldon NWR and observations made nearby
by other researchers (Collins and Bauman
2012, Castillo et al. 2016), we did not survey
this likely extensive population; thus, results
presented for this range may change when
the area is surveyed more fully. Despite its
relatively lower elevation and hotter cli-
mates, the Home Camp Range received the
highest median precipitation, which may mit-
igate the relatively warm and lower-elevation
environment. The Painted Point Range had
the third highest median precipitation in the
study area.
Classification Tree Analysis
The classification tree shows a series of
“checks and balances,” with environmental
and climatic variables mitigating potentially
harsh overarching conditions (e.g., a site with
low precipitation will have a higher chance of
remaining occupied if temperatures are cool).
Variable “importance,” a measure of predictive
power of each variable, was ranked in the
following order: water year precipitation, fol-
lowed closely by minimum July temperature,
maximum July temperature, binned aspect
categories, and solar radiation (Fig. 11). The
occupied sites were characterized by wet
(water year precipitation over 292 mm) and
cold (minimum July temperature under 8.8 °C)
climates and were mostly located on aspects
not southerly or easterly.
2017] RECENTLY DISCOVERED GREAT BASIN PIKA SITES 263
Occupied
.67 .33
36%
Relict
.30 .70
6%
Relict
.25 .75
12%
Relict
.05 .95
12%
Relict
.08 .92
34%
Occupied
.62 .38
42%
Occupied
.54 .46
54%
Relict
.45 .55
66%
Relict
.32 .68
100%
yes no
1
2
3
4
5
8
16 17 9
Fig. 11. The classification tree for the variables examined for all the occupied and relict sites in the dataset (n= 173;
WYPPT = average water year precipitation [mm], aspBnd = aspect, JulyTmin = median minimum July temperature
[°C]). The tree can be read as a series of hierarchical characteristics resulting in a terminal node or leaf. Leaves show the
designation of each site (either “occupied” or “relict”) and the probability of membership in each class, with “occupied”
as the first class (on the left) and “relict” as the second class (on the right). Lastly, each leaf presents the percentage of
total observations used in each “branch.” For example, when water year precipitation is under 292 mm, then there is a
92% probability that the sites are relict and an 8% probability that the sites are occupied. There are 59 observations
(34% of 173) falling into this leaf.
DISCUSSION
Survey results presented here make evident
that pika range and distribution in northwest-
ern Nevada have changed dramatically in the
20th and into the early 21st centuries, with
<30% of survey sites showing signs of current
pika persistence. Localized extirpations are
evident on the western side of the range at
high-elevation sites such as Little Hat Moun-
tain and Forty-Niner Mountain and on the
eastern side of the range throughout the high-
elevation Black Rock Range. To the north,
many sites in the Crooks Lake Hills and Mos-
quito Mountains appear extirpated, and the
Barrel Springs area (previously described as
extirpated by Beever et al. 2003, 2008, Wilken-
ing et al. 2011, Millar et al. 2014) remains
extirpated. Despite these range retractions and
calls for more spatially extensive fieldwork
(Beever et al. 2008), we note several previously
undocumented extant populations in a region
formerly described as containing only extir-
pated sites outside of the extant Massacre Rim,
Sheldon NWR, and Boulder Canyon (Hays
Canyon Range) populations (Beever et al. 2003,
2008, Wilkening et al. 2011).
Similar to recent pika inventories on the
Sheldon National Wildlife Refuge (Collins and
Bauman 2012), all of our currently occupied
and most (all but 2) of our relict sites were
below the 2500-m “threshold” considered
typical for pikas in the western Great Basin
(Smith 1974, Smith and Weston 1990, Millar
et al. 2013), although we note that our sites are
near the northern portion of the region and
some latitudinal depression would be expected.
When landform is considered, pikas in our
study are generally occupying more moderate
areas, perhaps indicative of the “goldilocks
effect” (not too hot, not too cold) where envi-
ronmental position (i.e., aspect, slope, and
elevation) and local climate (i.e., precipitation
and temperature) create a more hospitable
and less variable environment. Unlike the
more typical patterns in places such as Crooks
Lake Hills, Mosquito Mountain, and Nut
Mountain, many other areas in northwestern
Nevada do not fit the traditional pattern
where elevation is lower and heat stress is
higher at relict sites than at occupied sites. In
ranges such as Hays Canyon, Home Camp,
and Massacre, occupied sites trend lower and
hotter than relict sites. Despite these atypical
associations, pikas in the Home Camp Range
appeared to be the most connected and largest
population documented in our survey area.
Pikas are hypothesized to be affected in
many ways by decreased precipitation input
and loss of snowpack, including reduced
available forage due to lowered vegetation
productivity, increased spread of disease, loss
of insulating cover, and exposure to more
extreme cold temperatures (Wilkening et al.
2011). Our study did not measure precipita-
tion changes over time in relation to changing
occupancy over time. We report here only the
differences in water year precipitation amounts
across sites overall and across sites within a
range. Thus, we cannot comment on how
precipitation changes have driven changes in
occupancy, especially since we do not know
exactly when sites became extirpated. How-
ever, we do note that in 2 mountain ranges
where pikas are persisting at lower elevations
(Painted Point and Home Camp), precipitation
input over the survey areas within the ranges
is among the highest in our study, suggesting
that there could be some ameliorating influ-
ence of precipitation and the corresponding
vegetation response in warmer areas. Places
that received the highest amounts of precipita-
tion (e.g., Black Rock Range, upper elevations
of Home Camp Range, and Hays Canyon
Range) appear extirpated, and because of this
trend, the classification tree does place occu-
pied sites below a maximum precipitation
threshold. The degree to which this threshold
is actually representative of a true limitation
for pikas is unclear, particularly because we
suspect that some type of stochastic event, such
as fire or a disease outbreak (likely common in
pika populations; Wilkening et al. 2011), and
resulting fragmentation, prompted the loss of
pikas in these areas.
The inconsistency of modeled climate fac-
tors in explaining occupied versus relict sites
in our analyses might relate to the unique
conditions of pikas’ talus habitat and the scale
at which variables were examined. For example,
whereas weather station and modeled data
such as PRISM provide information about
surface free-air conditions, talus and related
broken-rock landforms are widely known to
have thermal regimes that are decoupled from
external surface air (Harris and Pederson
1998, Sawada et al. 2003, Gude et al. 2003,
Delaloye and Lambiel 2005, Zacharda et al.
264 WESTERN NORTH AMERICAN NATURALIST [Volume 77
2007, Millar et al. 2014). Decoupling of talus
from free-air circulation offers potential clues
to the ability of pikas to persist under synoptic
and regional climates that would appear unsuit-
able for them (Millar et al. 2016). Lava flows
and other volcanic taluses, such as those found
in our study area, are known to have particu-
larly cold matrices with highly attenuated
diurnal temperatures and often sustain persis-
tent embedded ice that may provide beneficial
daily and seasonal refuges for pikas (Millar et
al. 2016, Ray et al. 2016, Smith et al. 2016).
Decoupling processes may be responsible for
our solar radiation finding, where mean solar
radiation is higher at occupied than at relict
sites. Higher solar radiation may indicate
increased potential for vegetation growth at
mid- and upper-elevation sites where pikas
are found, while the ameliorating influence
of the belowground structure allows pikas to
remain in these sites. Vegetation growth
within and around the talus may also moder-
ate surface temperatures, and solar radiation
may interact with temperature, precipitation,
subsurface structures, and vegetation at each
site to create beneficial conditions such as
increased herbaceous vegetation and increased
access to food through more frequent and
earlier snowmelt. Unfortunately, our study has
no data regarding the temperatures within the
talus matrices, as might be collected with
data-logging sensors, but future work in the
area would benefit from these and other
microsite data.
A myriad of factors govern patch occupancy
dynamics, including connectivity to source
patches, the amount of suitable habitat in an
area, macro- and microclimate and the link
between the two as mediated by the above-
and belowground talus structure, vegetation
surrounding the patch, landscape variables
that influence connectivity, source-sink dynam-
ics, competition, predation, and disease, as
well as the lag effects of many of these factors.
As pikas are a classic example for metapopula-
tion biology and source-sink dynamics (Kreuzer
and Huntly 2003), it is possible that some site
“extirpations” may only be temporary and
sites might be recolonized after some time.
However, we suspect that numerous relict
sites documented across entire ranges (e.g.,
Black Rock Range) or areas within large por-
tions of each range (e.g., Hays Canyon Range
and northern High Rock Range) are true
extirpations. Although the cause(s) may never
be known, potential reasons for extirpation in
some of these areas include loss of connectivity,
small patch size (Stewart and Wright 2012), or
impacts of stochastic events (e.g., disease). In
other cases, it seems local climates may have
become unsuitable either over space (at rela-
tively lower elevations within or at the far
edge of a local distribution) or over time. For
example, some of the sites we found appeared
to have been recently extirpated, such as in
the Upper High Rock Canyon and the Hang-
ing Rock Canyon in the High Rock Range. In
other areas (e.g., Nellie Springs Mountain),
the few extant sites we found may be vestiges
of more abundant and widespread populations.
Outside of these peripheral areas, we find
currently occupied sites in spatially clustered
areas where climate and connectivity appear
conducive to maintaining pika populations in
the long-term. Although Massacre Rim in the
Painted Point Range was not thoroughly sur-
veyed, we suspect pikas are dense and abun-
dant across the entirety of the almost 7-km
face, which possibly serves as a “mainland
habitat” in the region. Techniques such as
radiocarbon dating of scat samples from relict
sites (as done in Millar et al. 2014) could pro-
vide insight regarding the pattern and causes
of extirpations. For instance, considering that
the Black Rock Range has burned in the last
century, relating relict site ages as obtained
from radiocarbon dating to fire history could
help us better understand the influence of fire
on pikas in more xeric, low-elevation sites.
Radiocarbon dating might also indicate the
direction of local range retractions, which
could consequently be used to predict high-
priority conservation areas and areas of possible
future retractions or (re)colonizations.
In our surveys, we found relict sites that we
expected to be occupied (e.g., Black Rock
Range and the upper elevations of the Hays
Canyon Range) and occupied sites where we
did not expect pikas to currently exist (e.g.,
Home Camp Range and the lower elevations
of the Hays Canyon Range) based on typical
niche descriptions. In our study area, no sin-
gle variable distinguished currently occupied
sites from relict sites in the same manner
across the mountain ranges, and the concept
that current and historic pika distributions are
contingent on particular combinations of local
site characteristics, regional climate regimes,
2017] RECENTLY DISCOVERED GREAT BASIN PIKA SITES 265
historical context, and connectivity (“idiosyn-
crasies of place”; Jeffress et al. 2013) is repeated
in our work. Our results provide further evi-
dence of the seemingly weak predictive power
of thermal indices alone in many of these
low-elevation and warm mountain ranges of
northwestern Nevada. For example, the classi-
fication tree describes unique environmental
settings that appear to foster persistence or
extirpation. In the tree, the probability of a
site being occupied ranged from 54% to 67%
but the probability of a site being relict
ranged from 55% to 95%, indicating (1) a much
higher predictive power with the variables
used in pinpointing relict sites on the land-
scape, (2) significant overlap in environmental
settings conducive to co-occurrence of both
occupied and relict sites, and (3) the lack of
additional predictive variables needed to dis-
criminate occupied and relict sites in this
study. When linked to environmental and cli-
mate variables, new analysis techniques, such
as measuring stress hormones in individuals
(Wilkening and Ray 2016), might provide
additional insight into pika survival and factors
influencing pika persistence. Furthermore, pika
presence at a site does not necessarily mean
the site has suitable habitat conditions that
will allow for long-term pika persistence
(Kreuzer and Huntly 2003). To more fully
untangle why and how some populations
persist and others do not, a more comprehen-
sive approach beyond occupancy surveys alone
will be needed.
Similar to Millar et al. (2013), we noted that
previous studies often do not examine multiple
scales of pika occupancy, which we attempted
to address. Delineation of possible demes in our
survey area provided insight into the potential
for connectivity and allowed a better under-
standing of the importance of a “patch”-sized
scale in determining area occupancy. We
acknowledge that because we were unable to
survey all potential habitat in every area, not
every currently occupied and relict site was
detected in our survey effort, and this lack of
data may artificially cause some demes to
appear disjunct. However, our approach was a
significant improvement on previous studies
that relied on a survey of only one site or one
habitat patch. For example, if we relied on
survey results from only our first 1 or 2 sites,
our understanding of the occupancy of the
area would have been very different, and we
might have concluded that the area was
recently extirpated when nearby patches were
actually currently occupied.
Understanding pika population stressors
may be particularly important to Bureau of
Land Management managers in northwestern
Nevada, where some districts include the
American pika on sensitive species lists that are
considered in decisions regarding land man-
agement activities and others do not (possibly
because they were not even aware the species
occurred in their district until now!). Beever
et al. (2003) reported a higher probability of
pika presence in places without livestock
activity and in wilderness areas, and Millar
(2011) discussed how grazing might be a factor
influencing population conditions and status
of pikas in the eastern Sierra Nevada and
Great Basin. Many of our sites in the Crooks
Lake Hills, Carter Reservoir, Mountain View
drainage to the west of Boulder Canyon, and
Grassy Canyon had active livestock grazing
as well as a feral horse presence but, due to
difficulty in quantifying, we were unable to
include these variables and others, such as the
presence of invasive plants (e.g., cheatgrass),
in our current analysis. In spite of this, our
study and future studies can help inform land
management decisions and may guide proac-
tive management opportunities to potentially
improve connectivity, such as providing artifi-
cial talus corridors or considering assisted
migration (Hobbs et al. 2010, Morueta-Holme
et al. 2010, Wilkening et al. 2015).
Our surveys reveal new populations across
northwest Nevada in several hotter and drier
places than have been described for the Ameri-
can pika. Because of their isolated nature and
relatively harsh conditions, these sites merit a
long-term and robust monitoring program to
address both population loss at currently
occu pied sites and possible recolonization of
seemingly extirpated sites. Regional patterns
in pika persistence and several seemingly
imperiled populations underscore the need
for more thorough inventory and monitoring
programs that examine not just occupancy at a
single site but also incorporate a more com-
prehensive regional understanding of connec-
tivity and population dynamics at larger scales.
The need for such work is urgent, given that
we found twice as many relict sites as occu-
pied sites in this region, with that ratio exag-
gerated in places like the Crooks Lake Hills.
266 WESTERN NORTH AMERICAN NATURALIST [Volume 77
Many questions remain, most urgently ones
surrounding connectivity, habitat patch size
relative to connectivity, population trends, and
protection and conservation measures for
extant pika populations, such as those that
exist in Home Camp Range, Painted Point
Range, Nut Mountain, and Crooks Lake Hills.
Consequently, we hope that this study is the
first of many in the region.
ACKNOWLEDGMENTS
We especially thank Mark Enders, Rory
Lamp, and Christy Klinger from the Nevada
Department of Wildlife (NDOW) for their
help with surveys across the study region. We
thank Kenny Pirkle, Brad Bauman, and Cody
Byrne from NDOW and Erik Beever with the
U.S. Geological Survey for their knowledge
and recommendations regarding survey areas.
Gail Collins shared shapefiles for the Sheldon
NWR, and Tom Dilts helped with the PRISM
data compilation and derivation. We thank
Elias Flores and staff at the Applegate Bureau
of Land Management field office for insight
and assistance with surveys on Massacre Rim.
The NDOW provided the majority of the sup-
port for biologist field time and equipment,
and the primary funding source was the State
Wildlife Grants Program.
LITERATURE CITED
BEEVER, E.A., P.F. BRUSSARD, AND J. BERGER. 2003. Pat-
terns of apparent extirpation among isolated popula-
tions of pikas (Ochotona princeps) in the Great
Basin. Journal of Mammalogy 84:37–54.
BEEVER, E.A., C. RAY, P.W. MOTE, AND J.L. WILKENING.
2010. Testing alternative models of climate-mediated
extirpations. Ecological Applications 20:164–178.
BEEVER, E.A., C. RAY, J.L. WILKENING, P.F. BRUSSARD,
AND P.W. MOTE. 2011. Contemporary climate change
alters the pace and drivers of extinction. Global
Change Biology 17:2054–2070.
BEEVER, E.A., J.L. WILKENING, D.E. MCIVOR, S.S. WEBER,
AND P.F. BRUSSARD. 2008. American pikas (Ochotona
princeps) in northwestern Nevada: a newly discov-
ered population at a low-elevation site. Western
North American Naturalist 68:8–14.
CASTILLO, J.A., C.W. EPPS, M.R. JEFFRESS, C. RAY, T.J.
RODHOUSE, AND D. SCHWALM. 2016. Replicated
landscape genetic and network analyses reveal wide
variation in functional connectivity for American
pikas. Ecological Applications 26:1660–1676.
CHARLET, D. Unpublished data. Nevada physiography: a
sketch.
COLLINS, G.H., AND B.T. BAUMAN. 2012. Distribution of
low-elevation American pika populations in the
northern Great Basin. Journal of Fish and Wildlife
Management 3:311–318.
DALY, C., M. HALBLEIB, J.I. SMITH, W.P. GIBSON, M.K.
DOGGETT, G.H. TAYLOR, J. CURTIS, AND P.P. PASTERIS.
2008. Physiographically sensitive mapping of clima-
tological temperature and precipitation across the
conterminous United States. International Journal of
Climatology 28:2031–2064.
DEATH, G., AND K.E. FABRICIUS. 2000. Classification and
regression trees: a powerful yet simple technique for
ecological data analysis. Ecology 81:3178–3192.
DELALOYE, R., AND C. LAMBIEL. 2005. Evidence of winter
ascending air circulation throughout talus slopes
and rock glaciers situated in the lower belt of
alpine discontinuous permafrost (Swiss Alps). Norsk
Geografisk Tidsskrift – Norwegian Journal of Geog-
raphy 59:194–203.
ERB, L.P., C. RAY, AND R. GURALNICK. 2011. On the gener-
ality of a climate-mediated shift in the distribution of
the American pika (Ochotona princeps). Ecology
92:1730–1735.
FU, P., AND P.M. RICH. 1999. Design and implementation
of the Solar Analyst: an ArcView extension for mod-
eling solar radiation at landscape scales. Proceedings
of the Nineteenth Annual ESRI User Conference.
FU, P., AND P.M. RICH. 2002. A geometric solar radiation
model and its applications in agriculture and
forestry. Proceedings of the Second International
Conference on Geospatial Information in Agricul-
ture and Forestry I-357-364.
GUDE, M., S. DIETRICH, R. MÄUSBACHER, C. HAUCK, R.
MOLENDA, V. RUZICKA, AND M. ZACHARDA. 2003. Prob-
able occurrence of sporadic permafrost in non-alpine
scree slopes in central Europe. Proceedings of the
8th International Conference on Permafrost, Zurich
2003:331–336.
HAFNER, D.J., AND A.T. SMITH. 2010. Revision of the sub-
species of the American pika, Ochotona princeps
(Lagomorpha: Ochotonidae). Journal of Mammalogy
91:401–417.
HALL, E.R. 1946. Mammals of Nevada. University of
Nevada Press, Las Vegas, NV.
HALL, L.E., A.D. CHALFOUN, E.A. BEEVER, AND A.E.
LOOSEN. 2016. Microrefuges and the occurrence of
thermal specialists: implications for wildlife persis-
tence amidst changing temperatures. Climate Change
Responses 3:8.
HARRIS, S.A., AND D.E. PEDERSON. 1998. Thermal
regimes beneath coarse blocky materials. Permafrost
and Periglacial Processes 9:107–120.
HOBBS, R.J., D.N. COLE, L. YUNG, E.S. ZAVALETA, G.H.
APLET, F.S. CHAPIN, P.B. LANDRES, D.J. PARSONS,
N.L. STEPHENSON, P.S. WHITE, ET AL. 2010. Guiding
concepts for park and wilderness stewardship in an
era of global environmental change. Frontiers in
Ecology and the Environment 8:483–490.
JEFFRESS, M.R., T.J. RODHOUSE, C. RAY, S. WOLFF, AND
C.W. EPPS. 2013. The idiosyncrasies of place: geo-
graphic variation in the climate-distribution relation-
ships of the American pika. Ecological Applications
23:864–878.
KREUZER, M.P., AND N.J. HUNTLY. 2003. Habitat-specific
demography: evidence for source-sink population
structure in a mammal, the pika. Oecologia 134:
343–349.
MANIS, G., J. LOWRY, AND R.D. RAMSEY. 2001. Pre-classifi-
cation: an ecologically predictive landform model.
GAP Analysis Bulletin Volume 10. U.S. Geological
Survey, Biological Resources Division. http://www
.gap.uidaho.edu/Bulletins/10/preclassification.htm
2017] RECENTLY DISCOVERED GREAT BASIN PIKA SITES 267
MARTÍNEZ-MEYER, E., A.T. PETERSON, AND W.W. HAR-
GROVE. 2004. Ecological niches as stable distributional
constraints on mammal species, with implications for
Pleistocene extinctions and climate change projections
for biodiversity. Global Ecology and Biogeography
13:305–314.
MILLAR, C.I. 2011. Influence of domestic livestock grazing
on American pika (Ochotona princeps) haypiling
behavior in the eastern Sierra Nevada and Great
Basin. Western North American Naturalist 71:
425–430.
MILLAR, C.I., K. HECKMAN, C. SWANSTON, K. SCHMIDT,
R.D. WESTFALL, AND D.L. DELANY. 2014. Radiocar-
bon dating of American pika fecal pellets provides
insights into population extirpations and climate
refugia. Ecological Applications 24:1748–1768.
MILLAR, C.I., R.D. WESTFALL, AND D.L. DELANY. 2013.
New records of marginal locations from American
pika (Ochotona princeps) in the western Great Basin.
Western North American Naturalist 73:457–476.
MILLAR, C.I., R.D. WESTFALL, AND D.L. DELANY. 2016.
Thermal components of American pika habitat—how
does a small lagomorph encounter climate? Arctic,
Antarctic, and Alpine Research 48:327–343.
MORUETA-HOLME, N., C. FLØJGAARD, AND J. SVENNING.
2010. Climate change risks and conservation impli-
cations for a threatened small-range mammal species.
PLOS ONE. http://journals.plos.org/plosone/article
?id=10.1371/journal.pone.0010360
MOYER-HORNER, L., M.M. SMITH, AND J. BELT. 2012. Citi-
zen science and observer variability during Ameri-
can pika surveys. Journal of Wildlife Management
76:1472–1479.
PARMESAN, C. 2006. Ecological and evolutionary responses
to recent climate change. Annual Review of Ecology,
Evolution, and Systematics 37:637–669.
RAY, C., E.A. BEEVER, AND T.J. RODHOUSE. 2016. Distri -
bution of a climate-sensitive species at an interior
range margin. Ecosphere 7:e01379.
R DEVELOPMENT CORE TEAM. 2016. R: a language and
environment for statistical computing. R Foundation
for Statistical Computing, Vienna, Austria. https://
www.R-project.org
REGONDA, S.K., B. RAJAGOPALAN, M. CLARK, AND J.
PITLICK. 2005. Seasonal cycle shifts in hydroclima-
tology over the western United States. Journal of
Climate 18:372–384.
RODHOUSE, T.J., E.A. BEEVER, L.K. GARRETT, K.M. IRVINE,
M.R. JEFFRESS, M. MUNTS, AND C. RAY. 2010. Distri-
bution of American pikas in a low-elevation lava
landscape: conservation implications from the range
periphery. Journal of Mammalogy 91:1287–1299.
SAWADA, Y., M. ISHIKAWA, AND Y. ONO. 2003. Thermal
regime of sporadic permafrost in a block slope on
Mt. Nishi-Nupukaushinupuri, Hokkaido Island,
North ern Japan. Geomorphology 52:121–130.
SCHWALM, D., C.W. EPPS, T.J. RODHOUSE, W.B. MONAHAN,
J.A. CASTILLO, AND M.R. JEFFRESS. 2016. Habitat
availability and gene flow influence diverging local
population trajectories under scenarios of climate
change: a place-based approach. Global Change
Biology 22:1572–1584.
SHINDERMAN, M. 2015. American pika in a low-elevation
lava landscape: expanding the known distribution of
a temperature-sensitive species. Ecology and Evolu-
tion 5:3666–3676.
SIMPSON, W.G. 2009. American pikas inhabit low-elevation
sites outside the species’ previously described bio-
climatic envelope. Western North American Natu-
ralist 69:243–250.
SMITH, A.T. 1974. Distribution and dispersal of pikas: influ-
ence of behavior and climate. Ecology 55:1368–1376.
SMITH, A.T., J.D. NAGY, AND C.I. MILLAR. 2016. Behavioral
ecology of American pikas (Ochotona princeps) at
Mono Craters, California: living on the edge. West-
ern North American Naturalist 76:459–484.
SMITH, A.T., AND M.L. WESTON. 1990. Ochotona princeps.
Mammalian Species 352:1–8.
STEWART, J.A.E., AND D.H. WRIGHT. 2012. Assessing per-
sistence of the American pika at historic localities in
California’s northern Sierra Nevada. Wildlife Society
Bulletin 36:759–764.
THOMAS, C.D., E.J. BODSWORTH, R.J. WILSON, A.D. SIM-
MONS, Z.G. DAVIES, M. MUSCHE, AND L. CONRADT.
2001. Ecological and evolutionary processes at
expanding range margins. Nature 411:577–581.
USGS GAP ANALYSIS PROGRAM. 2013. Ochotona princeps.
U.S. Geological Survey Gap Analysis Program
Species Distribution Models.
VARNER, J., AND M.D. DEARING. 2014a. Dietary plasticity
in pikas as a strategy for atypical resource land-
scapes. Journal of Mammalogy 95:72–81.
VARNER, J., AND M.D. DEARING. 2014b. The importance of
biologically relevant microclimates in species distri-
bution models. PLOS ONE 9(8):e104648.
WILDLIFE ACTION PLAN TEAM. 2012. Nevada Wildlife
Action Plan. Nevada Department of Wildlife, Reno,
NV.
WILKENING, J.L., AND C. RAY. 2016. Characterizing pre-
dictors of survival in the American pika (Ochotona
princeps). Journal of Mammalogy 97:1366–1375.
WILKENING, J.L., C. RAY, E.A. BEEVER, AND P.F. BRUS-
SARD. 2011. Modeling contemporary range retraction
in Great Basin pikas (Ochotona princeps) using data
on microclimate and microhabitat. Quaternary Inter-
national 27:1–12.
WILKENING, J.L., C. RAY, N. RAMSAY, AND K. KLINGLER.
2015. Alpine biodiversity and assisted migration:
the case of the American pika (Ochotona princeps).
Biodiversity 16:224–236.
ZACHARDA, M., M. GUDE, AND V. RU
˚ŽIˇ
CKA. 2007. Thermal
regime of three low elevation scree slopes in central
Europe. Permafrost and Periglacial Processes 18:
301–308.
Received 8 July 2016
Accepted 23 March 2017
Published online 19 July 2017
268 WESTERN NORTH AMERICAN NATURALIST [Volume 77
... In the GB, resurveys of historic sites document a steep extirpation curve in recent decades at low elevations and other marginal sites where temperature is typically warmest Stewart et al. 2015). In addition, extirpation at some northwestern GB sites has been assumed when only old sign was found during single-visit surveys (Jeffress, Van Gunst, and Millar 2017). In other timeseries studies of this region, however, population status does not correlate consistently with temperature or elevation (Millar et al. 2014a;Smith and Nagy 2015; CONTACT Constance I. Millar cmillar@fs.fed.us ...
... Stewart and Wright 2012). Surveys in parts of the western GB (Massing 2012;Millar and Westfall 2010;Millar, Westfall, and Delany 2013;Stewart et al. 2015) documented the locally widespread presence of pikas, including sites at low and climatically marginal GB sites, which suggest thermal tolerance (Beever et al. 2008;Collins and Bauman 2012;Jeffress, Van Gunst, and Millar 2017;Millar, Westfall, and Delany 2013;Smith, Nagy, and Millar 2016). ...
... This distance was used as a conservative haypile nearest-neighbor distance, representing an individual animal territory (Smith and Weston 1990). Records that overlapped 50 m buffers were combined into "50 m buffer clusters," which in some cases resulted in larger polygons (as in Jeffress, Van Gunst, and Millar 2017). ...
Article
To advance understanding of the distribution, climatic relationships, and status of American pikas (Ochotona princeps) in the Great Basin, United States, we compiled 2,387 records of extant pika sites surveyed since 2005, 89 records of documented extirpated sites (resurvey of historic sites), and 774 records of sites with old sign only. Extant sites extended across five degrees latitude and ten degrees longitude, encompassed six subregions, traversed forty mountain ranges, spanned 2,378 m in elevation (1,631-4,009 m), and comprised three of five currently described pika subspecies. A climate envelope for extant sites using the PRISM climate model expands the range of temperature and precipitation values that have been previously described. Extirpated and old-sign sites were mostly found within the geographic and climatic space of extant sites, but often in warmer and drier portions. Considerable overlap of extirpated, old, and extant groups within the same climate space suggests that nonclimatic factors have also contributed to population losses. The broad distribution and enlarged climate envelope of extant pika sites indicate that despite some localized extirpations, pika populations are persisting across Great Basin mountains, and appear to be able to tolerate a broader set of habitat conditions than previously understood.
... The species is thought to be undergoing contraction in some portions of its range due to accelerated warming (Beever, Brussard, & Berger, 2003;Beever et al., 2016) and in general is thought to be range-restricted by summer heat stress (Galbreath, Hafner, & Zamudio, 2009;. However, it continues to exhibit enigmatic persistence in apparently inhospitable locations (Beever, Wilkening, McIvor, Weber, & Brussard, 2008;Jeffress, Van Gunst, & Millar, 2017;Millar & Westfall, 2010;Millar, Westfall, & Delany, 2013;Ray, Beever, & Rodhouse, 2016;Rodhouse et al., 2010;Shinderman, 2015;Smith & Nagy, 2015;Smith, Nagy, & Millar, 2016), fuelling ongoing investigations into its vulnerability to contemporary climate change (Calkins, Beever, Boykin, Frey, & Anderson, 2012;Schwalm et al., 2016). ...
... Range-wide species distribution models and inspections of historically occupied sites arrayed along elevational gradients have provided evidence of contraction (Beever et al., 2003;Moritz et al., 2008;Calkins et al., 2012;Stewart et al., 2015;2017;Beever et al., 2016). But, finer-grained studies and those that have been replicated across multiple large landscapes using random sampling have presented evidence of enigmatic persistence in unusually low and hot locations, or otherwise have not found a consistent heat stress signal in their analyses Jeffress et al., 2013Jeffress et al., , 2017Millar & Westfall, 2010;Millar et al., 2018;Ray et al., 2016;Rowe et al., 2015;Schwalm et al., 2016;Shinderman, 2015;Smith & Nagy, 2015). Most pertinent to our study was Smith and Nagy's (2015) conclusion that heat stress apparently did not contribute to metapopulation dynamics in their long-term study of pikas in a California ore dump. ...
Article
Full-text available
Aim: Environmental changes that amplify rates of site or patch occupancy turnover can increase risks of decline in spatially-structured populations. We asked whether local habitat and meso-scale climate influenced site occupancy turnover rates in four American pika (Ochotona princeps) metapopulations. We focused on winter cold stress, which is a proposed driver of American pika extinction risk but has been rarely studied. Location: Oregon, Northern California, and Idaho. Methods: We developed Bayesian hierarchical multi-season site occupancy models that accounted for both false-negative and false-positive survey detection errors to explore the winter stress turnover hypothesis. We used remotely-sensed meso-scale (1 km) snowpack and temperature data and fine-grained local habitat attributes as covariates to model site persistence and colonization rates. Results: The estimated magnitude of imperfect detection was greater than previously reported for the species. After accounting for imperfect detection, we found no evidence of declines in site occupancy over the 5-year study period, but our models provided evidence that pika site occupancy turnover can be high (>50% between some years) and apparently exacerbated by winter cold stress, summer heat stress, and variation in site habitat quality. However, strength of evidence varied among metapopulations, suggesting influential local contingencies, as reported previously for the species. Main conclusions: Our empirical results suggest that the American pika may be vulnerable to wintertime turnover, negatively exacerbated by climatic events, with implications for future persistence given forecasted snowpack declines across the species’ range. Our models suggest a more nuanced dynamism to persistence and extinction risk than the simple scenario of inexorable, monotonic range contraction offered by static range-wide distribution models and we suggest several ways to strengthen these insights with future studies.
... While pikas occur abundantly on high mountains, often above local tree line and extending to peaks as high as 4300 m (Figure 1b), what is often overlooked is that they may occupy their obligate habitat, talus or piles of broken rock (Smith & Weston, 1990), far downslope. Where suitable rocky environments occur, pikas can be found at mid-montane elevations and lower, such as in semi-arid sagebrush (Artemisia spp.) steppe communities as low as 1450 m in the Great Basin (Jeffress et al., 2017;Millar et al., 2013Millar et al., , 2018Smith et al., 2016). Pikas even occupy talus near sea level in the Columbia River Gorge, Oregon (Horsefall, 1925;Simpson, 2009). ...
Article
Full-text available
American pikas (Ochotona princeps), small mammals related to rabbits, occur in mountainous regions of western North America, where they live in shattered‐rock habitats (talus). Aspects of their physiology and life history create situations that appear to put pikas at risk from warming climates. Some low‐elevation, warm sites that historically harbored pikas have become extirpated, and the assumption is that these will not be re‐colonized under current climate trends. Unexpectedly, in 2021, we found that pikas had re‐colonized two very warm, low‐elevation, dry sites in eastern California, USA, in the Bodie Mountains and Mono Craters. Resident pikas appear to have been absent at both sites for ≥10 years. These findings suggest that pikas, which are normally diurnally active, are able to overcome thermal dispersal barriers and re‐colonize long‐extirpated sites, perhaps by moving during cool nights. Our data also highlight the often unrecognized suitability of pika habitat in warm regions where the interiors of taluses can remain stably cool even when external air temperatures are hot. Here we show two locations where American pikas were found in 2021 to have re‐colonized long extirpated, very marginal sites. The sites are in the Bodie Mountains and Mono Craters, eastern California, USA.
... Pikas are currently experiencing range retractions , Jeffress et al., 2017 and local extirpations in many parts of their range (Nichols et al., 2016, Jeffress . Figure 1: Examples of hypothesized variation in stress for two hypothetical individuals. ...
Article
Full-text available
Temporal variation in stress might signify changes in an animal’s internal or external environment, while spatial variation in stress might signify variation in the quality of the habitats that individual animals experience. Habitat-induced variations in stress might be easiest to detect in highly territorial animals, and especially in species that do not take advantage of common strategies for modulating habitat-induced stress, such as migration (escape in space) or hibernation (escape in time). Spatial and temporal variation in response to potential stressors has received little study in wild animals, especially at scales appropriate for relating stress to specific habitat characteristics. Here, we use the American pika (Ochotona princeps), a territorial small mammal, to investigate stress response within and among territories. For individually territorial animals such as pikas, differences in habitat quality should lead to differences in stress exhibited by territory owners. We indexed stress using stress-associated hormone metabolites in feces collected non-invasively from pika territories every 2 weeks from June to September 2018. We hypothesized that differences in territory quality would lead to spatial differences in mean stress and that seasonal variation in physiology or the physical environment would lead to synchronous variation across territories through time. We used linear mixed-effects models to explore spatiotemporal variation in stress using fixed effects of day-of-year and broad habitat characteristics (elevation, aspect, site), as well as local variation in habitat characteristics hypothesized to affect territory quality for this saxicolous species (talus depth, clast size, available forage types). We found that temporal variation within territories was greater than spatial variation among territories, suggesting that shared seasonal stressors are more influential than differences in individual habitat quality. This approach could be used in other wildlife studies to refine our understanding of habitat quality and its effect on individual stress levels as a driver of population decline.
... The tool uses aspect and slope from a 30-m digital elevation model to derive a measure of total daily clear-sky radiation (diffuse and direct). We selected the single-day period of August 15 to represent total solar radiation during peak summer heat, a date used in other GB analyses (Van Gunst et al., 2016;Jeffress et al., 2017). We calculated solar loading for all sampled watersheds in the Wassuk Range. ...
Article
Annually dated tree-rings of 509 live and deadwood limber pine ( Pinus flexilis ) samples from the semi-arid Wassuk Range, Nevada, yielded a 3996-yr record extending from 1983 BC to AD 2013. Correlations of radial growth with climate were positive for water relations and negative for summer temperatures. Long-term trends of ring-width corresponded to climate variability documented from other proxies, including low growth during the Late Holocene Dry Period and Medieval Climate Anomaly (MCA) and elevated growth during cool, wet periods of the Neoglacial and Little Ice Age. Spline fit of the data indicated that growth decrease in the last 20 years was second lowest on record, surpassed by lowest growth at 20 BC—AD 150. Demographics of limber pine by aspect and elevation were not strongly related to long-term climate dynamics, except in the case of extirpations on all but north aspects at the end of the MCA. Pines occurred persistently on north aspects, where a continuous record existed to present. Elevation shifts were not obvious on any aspect, and no evidence existed for migration above current treeline. Non-climatic factors appear to interact with climate to make north slopes refugial for upland pines in semi-arid regions across four millennia.
Preprint
Declines and extirpations of American pika ( Ochotona princeps) populations at historically occupied sites started being documented in the literature during the early 2000s. Commensurate with global climate change, many of these losses at peripheral and lower elevation sites have been associated with changes in ambient air temperature and precipitation regimes. Here, we report on a decline in available genetic resources for an iconic American pika metapopulation, located at the southwestern edge of the species distribution in the Bodie Hills of eastern California, USA. Composed of highly fragmented habitat created by hard rock mining, the ore dumps at this site were likely colonized by pikas at the end of the 19th century from nearby natural talus outcrops. Genetic data extracted from both contemporary samples and archived natural history collections allowed us to track population and patch-level genetic diversity for Bodie pikas across three distinct sampling points during the last half- century (1948, 1988–1991, 2013–2015). In addition to declines in within-population allelic diversity and expected heterozygosity, we observed an increase in population structure and a reduction in effective population size from more extensive sampling of extant patches during 1988–1991 and 2013–2015, respectively. Furthermore, census records from the last 50 years as well as archived museum samples collected in 1947 from a nearby pika population in the Wassuk range (NV, USA) provide further support of the increasing isolation and genetic coalescence occurring in this region. This study highlights the importance of museum samples and long-term monitoring in contextualizing our understanding of population viability.
Article
The American pika (Ochotona princeps) is commonly perceived as a species that is at high risk of extinction due to climate change. The purpose of this review is two-fold: to evaluate the claim that climate change is threatening pikas with extinction, and to summarize the conservation status of the American pika. Most American pikas inhabit major cordilleras, such as the Rocky Mountain, Sierra Nevada, and Cascade ranges. Occupancy of potential pika habitat in these ranges is uniformly high and no discernible climate signal has been found that discriminates between the many occupied and relatively few unoccupied sites that have been recently surveyed. Pikas therefore are thriving across most of their range. The story differs in more marginal parts of the species range, primarily across the Great Basin, where a higher percentage of available habitat is unoccupied. A comprehensive review of Great Basin pikas revealed that occupied sites, sites of recent extirpation, and old sites, were regularly found within the same geographic and climatic space as extant sites, and suggested that pikas in the Great Basin tolerated a broader set of habitat and climatic conditions than previously understood. Studies of a small subset of extirpated sites in the Great Basin and in California found that climate variables (most notably measures of hot temperature) were associated more often with extirpated sites than occupied sites. Importantly, upward contraction of the lower elevation boundary also was found at some sites. However, models that incorporated variables other than climate (such as availability of upslope talus habitat) often were better predictors of site persistence. Many extirpations occurred on small habitat patches, which were subject to stochastic extinction, as informed by a long-term pika metapopulation study in Bodie, California. In addition, several sites may have been compromised by cattle grazing or other anthropogenic factors. In contrast, several low, hot sites (Bodie, Mono Craters, Craters of the Moon National Monument and Preserve, Lava Beds National Monument, Columbia River Gorge) retain active pika populations, demonstrating the adaptive capacity and resilience of pikas in response to adverse environmental conditions. Pikas cope with warm temperatures by retreating into cool interstices of their talus habitat and augment their restricted daytime foraging with nocturnal activity. Pikas exhibit significant flexibility in their foraging tactics and are highly selective in their choice of available vegetation. The trait that places pikas at greatest risk from climate change is their poor dispersal capability. Dispersal is more restricted in hotter environments, and isolated low-elevation sites that become extirpated are unlikely to be recolonized in a warming climate. The narrative that American pikas are going extinct appears to be an overreach. Pikas are doing well across most of their range, but there are limited, low-elevation losses that are likely to be permanent in what is currently marginal pika habitat. The resilience of pikas in the face of climate change, and their ability or inability to persist in marginal, hot environments, will continue to contribute to our understanding of the impact of climate change on individual species.
Article
Full-text available
As regional climates continue to warm, climate-sensitive species might increasingly depend on microclimates to maintain normal activities. The American pika (Ochotona princeps) has been suggested as a sentinel of montane climate change, largely due to its sensitivity to warm temperatures and high exposure to changing climatic conditions. Pikas use talus to thermoregulate and have been shown to persist more often in talus slopes that create stable microclimates. However, many studies of pika behavior and persistence use metrics of climate measured above the talus surface, outside of the proposed microclimates that pikas can access. We paired fine-scale microclimate measurements with behavioral observations of uniquely tagged pikas in the Colorado Rocky Mountains to model how pikas might use talus subsurface habitat as a microrefuge. Our findings suggest that pikas spend more time on the surface of the talus when there is a stronger gradient in temperatures between "shallow"(0.1 m below the surface) and "deep"(1 m below the surface) regions of the talus, especially during the middle of the day when shallow temperatures are warmer. These results underscore the importance of pairing fine-scale data on microclimate with behavioral observations of tagged individuals to more accurately evaluate how animals might use climate refugia to persist in changing conditions.
Article
Full-text available
Genetic variation is the basis upon which natural selection acts to yield evolutionary change. In a rapidly changing environment, increasing genetic variation should increase evolutionary potential, particularly for small, isolated populations. However, the introduction of new alleles, either through natural or human-mediated processes, may have unpredictable consequences such as outbreeding depression. In this study, we identified a contact zone and limited gene flow between historically separated genetic lineages of American pikas (Ochotona princeps), representing the northern and southern Rocky Mountain subspecies, within Rocky Mountain National Park. The limited spatial extent of gene flow observed may be the result of geographic barriers to dispersal, selection against hybrid individuals, or both. Our fine-scale population genetic analysis suggests gene flow is limited but not completely obstructed by extreme topography such as glacial valleys, as well as streams including the Colorado River. The discovery of two subspecies within this single protected area has implications for monitoring and management, particularly in the light of recent analyses suggesting that the pikas in this park are vulnerable to fragmentation and local extinction under future projected climates. Future research should focus on the fitness consequences of introgression among distinct genetic lineages in this location and elsewhere, as well as within the context of genetic rescue as a conservation and management strategy for a climate sensitive species.
Article
Full-text available
Advances in understanding the factors that limit a species’ range, particularly in the context of climate change, have come disproportionately through investigations at range edges or margins. The margins of a species’ range might often correspond with anomalous microclimates that confer habitat suitability where the species would otherwise fail to persist. We addressed this hypothesis using data from an interior, climatic range margin of the American pika (Ochotona princeps), an indicator of relatively cool, mesic climates in rocky habitats of western North America. Pikas in Lava Beds National Monument, northeastern California, USA, occur at elevations much lower than predicted by latitude and longitude. We hypothesized that pika occurrence within Lava Beds would be associated primarily with features such as “ice caves” in which sub-surface ice persists outside the winter months. We used data loggers to monitor sub-surface temperatures at cave entrances and at non-cave sites, confirming that temperatures were cooler and more stable at cave entrances. We surveyed habitat characteristics and evidence of pika occupancy across a random sample of cave and non-cave sites over a 2-yr period. Pika detection probability was high (~0.97), and the combined occupancy of cave and non-cave sites varied across the 2 yr from 27% to 69%. Contrary to our hypothesis, occupancy was not higher at cave sites. Vegetation metrics were the best predictors of site use by pikas, followed by an edge effect and elevation. The importance of vegetation as a predictor of pika distribution at this interior range margin is congruent with recent studies from other portions of the species’ range. However, we caution that vegetation composition depends on microclimate, which might be the proximal driver of pika distribution. The microclimates available in non-cave crevices accessible to small animals have not been characterized adequately for lava landscapes. We advocate innovation in the acquisition and use of microclimatic data for understanding the distributions of many taxa. Appropriately scaled microclimatic data are increasingly available but rarely used in studies of range dynamics.
Article
Full-text available
The measurement of stress hormone (glucocorticoid [GC]) concentration is increasingly used to assess the health of wildlife populations. However, for many species, we do not have a good understanding of the range of GC concentrations that might indicate a compromised individual. A temporary increase in GC concentration can prompt the adoption of behavior or activities to promote individual survival. However, chronic GC elevation results in deleterious effects on health that can reduce survival. In order to use GC concentration as a metric of individual fitness for a given species, it will be necessary to relate individual demographic rates to GC concentration. We related survival in American pikas (Ochotona princeps) to 2 different stress metrics, glucocorticoid metabolite (GCM) concentration in fecal samples and GC concentration in plasma samples. Annual survival was analyzed in relation to each of these stress metrics as well as other physiological metrics and habitat characteristics at several sites in the Rocky Mountains. Among the predictors considered, GCM concentration was by far the strongest predictor of annual survival in pikas, and individuals with higher baseline GCM were less likely to survive. Our metric of flea load was also negatively related to annual survival. Given the limited time and resources that characterize many wildlife conservation projects, it is important to establish which endocrine metrics are the most informative for a species. American pikas have been identified as a sentinel species for detecting effects of climate change, and several correlational studies have projected range contraction for the species. Our results suggest that more mechanistic projections might be possible given further study of the relationship between GCM and climate. Our approach contributes to a better understanding of factors affecting survival in this species and provides a basis for further research relating individual stress response and survival to environmental change.
Article
Full-text available
Anticipating the response of small mammals to climate change requires knowledge of thermal conditions of their habitat during times of the day and year when individuals use them. We measured diurnal and seasonal temperatures of free air and of six habitat components for American pikas (Ochotona princeps) over five years at 37 sites in seven mountain ranges in the western Great Basin, United States. Talus matrices (subsurfaces) had low daily variances and, in the warm season, remained cool during the hottest times of the day relative to surfaces and free air. During winter, matrices were warmer than free air. Talus surfaces were warmer than free air in the warm and cold seasons, and had large daily variances. Summer forefield and dispersal environments were warmest of all habitat components. Talus surfaces in summer were highly responsive to solar radiation over the course of the day, warming quickly to high midday temperatures, and cooling rapidly in the evening. By contrast, matrices lagged the daily warm-up and remained warmer than free air at night. These differences afford diurnal and seasonal opportunities for pikas to adapt behaviorally to unfavorable temperatures and suggest that animals can accommodate a wider range of future climates than has been assumed, although warming of the dispersal environment may become limiting. Climate envelope models that use or model only surface air measures and do not include information on individual thermal components of pika habitat may lead to errant conclusions about the vulnerability of species under changing climates.
Article
Full-text available
Landscape connectivity is essential for maintaining viable populations, particularly for species restricted to fragmented habitats or naturally arrayed in metapopulations and facing rapid climate change. The importance of assessing both structural connectivity (the physical distribution of favorable habitat patches) and functional connectivity (how species move among habitat patches) for managing such species is well understood. However, the degree to which functional connectivity for a species varies among landscapes, and the resulting implications for conservation, have rarely been assessed. We used a landscape genetics approach to evaluate resistance to gene flow and, thus, to determine how landscape and climate-related variables influence gene flow for American pikas (Ochotona princeps) in eight federally managed sites in the western United States. We used those empirically-derived, individual-based landscape resistance models in conjunction with predictive occupancy models to generate patch-based network models describing functional landscape connectivity. Metareplication across landscapes enabled identification of limiting factors for dispersal that would not otherwise have been apparent. Despite the cool microclimates characteristic of pika habitat, south-facing aspects consistently represented higher resistance to movement, supporting the previous hypothesis that exposure to relatively high temperatures may limit dispersal in American pikas. We found that other barriers to dispersal included areas with a high degree of topographic relief, such as cliffs and ravines, as well as streams and distances greater than one to four kilometers depending on the site. Using the empirically-derived network models of habitat patch connectivity, we identified habitat patches that were likely disproportionately important for maintaining functional connectivity, areas in which habitat appeared fragmented, and locations that could be targeted for management actions to improve functional connectivity. We concluded that climate change, besides influencing patch occupancy as predicted by other studies, may alter landscape resistance for pikas, thereby influencing functional connectivity through multiple pathways simultaneously. Spatial autocorrelation among genotypes varied across study sites and was largest where habitat was most dispersed, suggesting that dispersal distances increased with habitat fragmentation, up to a point. This study demonstrates how landscape features linked to climate can affect functional connectivity for species with naturally fragmented distributions, and reinforces the importance of replicating studies across landscapes.
Article
Full-text available
Ecological niche theory holds that species distributions are shaped by a large and complex suite of interacting factors. Species distribution models (SDMs) are increasingly used to describe species' niches and predict the effects of future environmental change, including climate change. Currently, SDMs often fail to capture the complexity of species' niches, resulting in predictions that are generally limited to climate-occupancy interactions. Here, we explore the potential impact of climate change on the American pika using a replicated place-based approach that incorporates climate, gene flow, habitat configuration, and microhabitat complexity into SDMs. Using contemporary presence-absence data from occupancy surveys, genetic data to infer connectivity between habitat patches, and 21 environmental niche variables, we built separate SDMs for pika populations inhabiting eight US National Park Service units representing the habitat and climatic breadth of the species across the western United States. We then predicted occurrence probability under current (1981-2010) and three future time periods (out to 2100). Occurrence probabilities and the relative importance of predictor variables varied widely among study areas, revealing important local-scale differences in the realized niche of the American pika. This variation resulted in diverse and - in some cases - highly divergent future potential occupancy patterns for pikas, ranging from complete extirpation in some study areas to stable occupancy patterns in others. Habitat composition and connectivity, which are rarely incorporated in SDM projections, were influential in predicting pika occupancy in all study areas and frequently outranked climate variables. Our findings illustrate the importance of a place-based approach to species distribution modeling that includes fine-scale factors when assessing current and future climate impacts on species' distributions, especially when predictions are intended to manage and conserve species of concern within individual protected areas. This article is protected by copyright. All rights reserved.
Article
Full-text available
Alpine mammals are predicted to be among the species most threatened by climate change, due to the projected loss and further fragmentation of alpine habitats. As temperature or precipitation regimes change, alpine mammals may also be faced with insurmountable barriers to dispersal. The slow rate or inability to adjust to rapidly shifting environmental conditions may cause isolated alpine species to become locally extirpated, resulting in reduced biodiversity. One proposed method for mitigating the impacts of alpine species loss is assisted migration. This method, which involves translocating a species to an area with more favourable climate and habitat characteristics, has become the subject of debate and controversy in the conservation community. The uncertainty associated with climate change projections, coupled with the thermal sensitivity of many alpine mammals, makes it difficult to a priori assess the efficacy of this technique as a conservation management tool. Here we present the American pika (Ochotona princeps) as a case study. American pikas inhabit rocky areas throughout the western US, and populations in some mountainous areas have become locally extirpated in recent years. We review known climatic and habitat requirements for this species, and also propose protocols designed to reliably identify favourable relocation areas. We present data related to the physiological constraints of this species and outline specific requirements which must be addressed for translocation of viable populations, including wildlife disease and genetic considerations. Finally, we discuss potential impacts on other alpine species and alpine communities, and overall implications for conserving alpine biodiversity in a changing climate.
Article
Full-text available
We describe 46 new site records documenting occupancy by American pika (lOchotona princeps) at 21 locations from 8 mountain regions in the western Great Basin, California, and Nevada. These locations comprise a sub- set of sites selected from regional surveys to represent marginal, isolated, or otherwise atypical pika locations, and to provide information for assessing environmental tolerance limits. Several locations are known from historic observations (Madeline Plain, Bodie Mtns., Wassuk Mtns., Mono Craters) and are included here to update current status. Site eleva- tions range from 1848 m to 3392 m; relative to the broad range of pika sites in the region, the new locations have cli- mates that are 2—4 °C warmer and receive approximately half the annual precipitation. Sites are located in lava flows and domes, inselbergs (isolated, rocky exposures on a small hill), eroding bedrock, rock-glacier till, talus slopes, and anthro- pogenic roadbed armaments and mining ore dumps. Several sites are situated in uncommon vegetation contexts, for example, montane desert scrub communities or locations where vegetation adjacent to taluses is sparse or lacking. Prox- imity to surrounding pika habitats (as a measure of marginality) was evaluated based on relative talus distribution pat- terns for 0.5-km, 2.5-km, and 5.0-km circular areas nested around each site. Seven idealized, schematic spatial patterns were used to assess potential connectivity among sites, ranging from “island” (no other talus within the respective areas) to “even” (many talus patches regularly distributed). Applying this approach to the 21 sites demonstrated a simple method for qualitatively assessing pika habitat relative to dispersal potential and metapopulation viability and also revealed complexities of biogeographic patterns related to marginality.
Article
The behavioral ecology of the American pika (Ochotona princeps) was investigated at a relatively hot south-facing, low-elevation site in the Mono Craters, California, a habitat quite different from the upper montane regions more typically inhabited by this species and where most prior investigations have been conducted. Mono Craters pikas exhibited a behavioral profile that contrasted significantly with that of pikas found in upper montane regions. Mono Craters pikas were less surface active than pikas in studies at high-elevation sites, although their rate of short-call vocalizations was similar. Mono Craters pikas did not exhibit typical foraging behavior: they fed and collected hay at significantly reduced rates, and did not construct large central-place hay piles. Social behaviors (conspecific aggression, social tolerance, avoidance) were infrequent compared with data from prior studies in upper montane environments. The Mono Craters site appears to be one of the warmest localities in which pikas have been observed. Recorded talus surface temperatures consistently exceeded 30 °C, and temperatures >40 °C were commonly recorded. In contrast, temperatures measured in the matrix of the talus were consistently cooler, and the apparent insulating effect of talus, as measured by the difference between surface and matrix temperatures, was typically most pronounced on the hottest days. Although pika activity was most frequent in early morning, late afternoon, and at night, pikas were also active during the hottest part of the day, presumably because of their ability to behaviorally thermoregulate by retreating into the cooler talus matrix. Data on populations of pikas which inhabit marginal sites can help us understand how pikas and other montane animals might respond in a world of climate change so that we may more effectively plan for their conservation.
Article
Background Contemporary climate change is affecting nearly all biomes, causing shifts in animal distributions, phenology, and persistence. Favorable microclimates may buffer organisms against rapid changes in climate, thereby allowing time for populations to adapt. The degree to which microclimates facilitate the local persistence of climate-sensitive species, however, is largely an open question. We addressed the importance of microrefuges in mammalian thermal specialists, using the American pika (Ochotona princeps) as a model organism. Pikas are sensitive to ambient temperatures, and are active year-round in the alpine where conditions are highly variable. We tested four hypotheses about the relationship between microrefuges and pika occurrence: 1) Local-habitat Hypothesis (local-habitat conditions are paramount, regardless of microrefuge); 2) Surface-temperature Hypothesis (surrounding temperatures, unmoderated by microrefuge, best predict occurrence); 3) Interstitial-temperature Hypothesis (temperatures within microrefuges best predict occurrence), and 4) Microrefuge Hypothesis (the degree to which microrefuges moderate the surrounding temperature facilitates occurrence, regardless of other habitat characteristics). We examined pika occurrence at 146 sites across an elevational gradient. We quantified pika presence, physiographic habitat characteristics and forage availability at each site, and deployed paired temperature loggers at a subset of sites to measure surface and subterranean temperatures. ResultsWe found strong support for the Microrefuge Hypothesis. Pikas were more likely to occur at sites where the subsurface environment substantially moderated surface temperatures, especially during the warm season. Microrefugium was the strongest predictor of pika occurrence, independent of other critical habitat characteristics, such as forage availability. Conclusions By modulating surface temperatures, microrefuges may strongly influence where temperature-limited animals persist in rapidly warming environments. As climate change continues to manifest, efforts to understand the changing dynamics of animal-habitat relationships will be enhanced by considering the quality of microrefuges.