McClure, Meredith M.1*, Hranac, Carter
R.2, Haase, Catherine G.3,4,
McGinnis, Seth5, Dickson, Brett
G.1,6, Hayman, David T.S.2, McGuire,
Liam P.7,8, Lausen, Cori L.9,
Plowright, Raina K.3, Fuller,
N.7,10, Olson, Sarah H.11
1 Conservation Science Partners, Truckee, USA
2 mEpiLab, Hopkirk Research
Institute, Massey University, Palmerston North, NZL
3 Department of Microbiology and Immunology, Montana
State University, Bozeman, USA
4 Department of Biology, Austin Peay State University,
Clarksville, USA
5 National Center for Atmospheric Research, Boulder,
USA
6 Landscape Conservation Initiative, School of Earth
and Sustainability, Northern Arizona University, Flagstaff, USA
7 Department of Biological Sciences, Texas Tech
University, Lubbock, USA
8 Department of Biology, University of Waterloo,
Waterloo, CAN
9 Wildlife Conservation Society Canada, Kaslo, CAN
10 Texas Parks & Wildlife Department, Austin, USA
11 Wildlife Conservation Society, Health Program,
Bronx, USA
Keywords
Bats, subterranean microclimate, climate change,
hibernation, species distribution model, white-nose
syndrome
Introduction
Climate change and infectious disease emergence are major threats to
biodiversity (Dawson et al. 2011, Fisher et al. 2012). Increasing
temperatures, changes in the amount and timing of precipitation,
increased frequency and severity of extreme conditions, and other
changes in climate conditions (Pachauri et al. 2014) impact species and
communities in a variety of ways. Climate change has already shifted
distributions of a diverse range of species (Parmesan 2006) and is
projected to drive future shifts (Lawler et al. 2009). Some species’
fundamental niches are moving or disappearing altogether (Colwell &
Rangel 2009, Thomas et al. 2004, Thuiller et al. 2006), while others may
expand beyond range limits previously imposed by unsuitable climate
conditions (e.g., Melles et al. 2011, Battisti & Larsson 2015). These
range shifts may in turn drive changes in interspecific co-occurrence
and population dynamics among competitors, predators, and prey (e.g.,
Alexander et al. 2016, Urban et al. 2013) as well as diseases,
parasites, and hosts (Gallana et al. 2013, Adlard et al. 2015, Metcalf
et al. 2017).
All of these climate change impacts may be at play for bats and are
expected to interact with the impacts of white-nose syndrome (WNS). WNS,
caused by a cold-loving fungus (Pseudogymnoascus destructans )
introduced to New York state in 2006, has killed millions of hibernating
bats across eastern and central North America by disrupting hibernation
physiology (Leopardi et al. 2015, Frick et al. 2016). It continues to
spread widely and rapidly from its introduction site, including a 2016
novel introduction to Washington state (USFWS 2020), and is now invading
western North America (herein the West). P. destructans grows on
the skin of hibernating bats and, through a number of physiological
mechanisms, causes them to arouse from their torpid state more
frequently than healthy bats (Frick et al. 2016). These arousals are
energetically expensive (Thomas et al. 1990), causing infected bats to
expend fat stores before the end of winter. Impact severity varies
geographically and among species, and has been linked to
microclimate-dependent fungal growth (Verant et al. 2012, Marroquin et
al. 2017), interspecific and microclimate-dependent differences in host
physiology (Johnson et al. 2014, Moore et al. 2017, McGuire et al. in
review), as well as interspecific differences in hibernation behavior,
including microclimate preferences (Langwig et al. 2012, 2016).
Despite the growing understanding of these mechanisms, WNS impacts on
bats remain difficult to predict, particularly as P. destructansspreads to novel environments supporting diverse species (Harvey et al.
2013). For example, it is common to find large aggregations of
hibernating bats in eastern and central North America, but this is
rarely observed in the West. Instead, western bats tend to hibernate in
widely distributed small groups (Weller et al. 2018, Adams 2003, Bachen
et al. 2018). These differences, along with the rugged, remote
landscapes characterizing much of the West, have made the study of
western bats challenging. Gaps therefore remain in our understanding of
western bat ecology and the potential impacts of various stressors,
including continued WNS spread.
Climate change presents an additional layer of uncertainty regarding WNS
impacts on bats. Bat hibernation physiology and behavior, as well asP. destructans physiology, are closely linked to climate
conditions. Hibernaculum temperature and humidity, along with winter
duration, dictate healthy hibernating bats’ success in surviving winter
on their fat stores (Thomas et al., 1990, Speakman & Thomas, 2003);
temperature and humidity also determine fungal growth rates (Verant et
al. 2012, Marroquin et al. 2017). WNS survivorship largely depends on
whether fat stores can sustain bats through winter given increased
arousal frequencies and associated energy costs resulting from P.
destructans infection (Langwig et al. 2012, 2016, Hayman et al. 2016,
Haase et al. 2019). A warming climate may shift bats’ winter
distributions to track shifts in preferred hibernaculum conditions. In
some hibernacula, higher temperatures may increase fungal loads by
expanding availability of suitable growth conditions, while other
hibernacula may experience the opposite trend. These warming
temperatures may simultaneously alter bats’ winter energy expenditures,
to their benefit or detriment. Meanwhile, shorter winters could help to
reduce mortality resulting from infected bats expending fat stores too
quickly.
We modeled current winter distributions of five bat species using a
hybrid correlative-mechanistic approach (McClure et al. accepted; Fig.
1). We correlated observed winter occurrence of our focal species with
landscape attributes expected to influence hibernaculum selection (e.g.,
topography, vegetation cover, water availability). As an additional
predictor, we integrated a spatially explicit estimate of hibernation
survivorship derived from a mechanistic bioenergetic model (Haase et al.
2019, Hranac et al. accepted). The bioenergetic model uses the
hypothesized energetic requirements of bats during hibernation to
dynamically model energy expenditure for the duration of a predicted
winter under specified hibernaculum conditions. The model was
parameterized for each of our focal species using field measurements of
key aspects of hibernation physiology, and was run under current climate
conditions, including model-based estimates of mean winter ambient
temperatures experienced in hibernacula (McClure et al. 2020) and winter
duration at a given location (Hranac et al. accepted).
Here, we apply this bioenergetic model given the presence of P.
destructans : we include parameters capturing P. destructans’climate-dependent growth as well as its climate-dependent effects on
host physiology, under both current climate conditions and scenarios of
future climate change (Fig. 1). We then update our species distribution
models (McClure et al. accepted) with the resulting survivorship
estimates to predict changes in the distribution of suitable winter
hibernacula under these projected future conditions. To our knowledge,
there has been no attempt to model changing distributions of winter
hibernacula in response to WNS exposure or climate change, let alone
both. Our objective is to understand and predict the individual and
joint effects of these two imminent stressors on North American bat
populations. Our goal is to support researchers and managers in
anticipating and planning for future impacts to bats. We expect this
work will support managers in identifying species and geographies that
are expected to be most affected by WNS, identifying populations for
which WNS impacts may be either exacerbated or mitigated by climate
change, and allocating monitoring and management resources accordingly.
Methods
We sought to estimate the change in five focal bat species’ probability
of occurrence (estimated under current conditions in McClure et al.
accepted) given two future scenarios: a) exposure to P.
destructans, and b) exposure to P. destructans and climate
change. These species, including Corynorhinus townsendii, Myotis
californicus, M. lucifugus, M. velifer, and Perimyotis subflavus, were
selected based on data availability and representation of diverse
distributions and habitat requirements among hibernating bats. To
estimate bats’ probability of occurrence given exposure to P.
destructans , we ran the spatial bioenergetic model described in Hranac
et al. (accepted; also see Haase et al. 2019) to project winter
survivorship from parameters capturing the influence of the hibernaculum
environment (temperature and humidity) on fungal growth and the
resulting impact of the fungus on bat hibernation physiology. To
estimate bats’ probability of occurrence given the additional impacts of
climate change, we ran the bioenergetic model with the P.
destructans growth parameters above as well as projected future climate
parameters (winter duration and ‘best available’ temperatures,
identified as the subterranean temperature closest to the species’
preferred temperature as identified from published literature that was
projected to be available in a given location; Fig. 1). The bioenergetic
model, P. destructans growth parameters, and spatial application
of the model are described fully in Haase et al. (2019) and Hranac et
al. (accepted) and summarized in Appendix 1. We therefore focus here on
describing integration of future climate scenarios into the bioenergetic
model and subsequently SDMs for our five focal species.
We first projected daily temperatures at midcentury (2050) under a range
of possible climate futures at high spatial resolution (1 km), which
were then used to derive our climate parameters of interest. Global
circulation models (GCMs) represent the energy budget of the earth
system and the impact of external factors such as solar input and
greenhouse gas emissions, simulating global patternsand processes across
the earth’s major climate system components (atmosphere,ocean, sea ice,
and land surface) to project future climate attributes (e.g.,
temperature,-recipitation) under possible future scenarios of carbon and
other heat-trapping gas concentrations (Kiel & Ramanathan 2006).
regional climate models (RCMs) dynamically (i.e., mechanistically)
downscale coarse GCM projections by resolving processes that occur at
finer resolutions than GCM grid sizes (≥100 km) within a more limited
geographic sco-e (Kotamarthi et al. 2016). They account for the effects
of local complexity, e.g., topography and coastlines, and simulate
hydrologic processes at scales more relevant to decision-making (25-50
km). However, these outputs are still too coarse for many applications.
GCM and RCM projections can be further statistically downscaled using a
variety of approaches. Although many methods exist and vary considerably
in their complexity, they all fundamentally aim to account for
differences between model simulations applied to historical periods and
observed climate attributes during those periods, then apply those
statistical adjustments to future projections (Kotamarthi et al. 2016).
The NA-CORDEX Program data archive (Mearns et al. 2017), hosted by the
National Center for Atmospheric Research , contains output from RCMs run
over a domain covering most of
North America using boundary conditions from multiple CMIP5 GCMs
(Appendix 2, Fig. A1). These projections span a range of possible
climate futures in terms of greenhouse gas concentration scenarios and
projected severity of future change, as well as performance in capturing
regionally important drivers and processes.
The NA-CORDEX data archive includes outputs from two RCMs that offer 25
km spatial resolution and span the complete range (2.4 - 4.6°C) of GCM
equilibrium climate sensitivity (ECS), an emergent property of GCMs that
serves as a metric of relative severity of projected change. These are
the RegCM4 model (Giorgi et al. 2012) and the WRF model (Skamarock et
al. 2008) (Fig. 2). These models differ in their underlying sub-models
and -processes (see https://na-cordex.org/rcm-characteristics),
which may mean that each best represents the meteorological phenomena
driving future climate change in different subregions of North America.
Kotamarthi et al. (2016) suggest that it is critical to understand the
phenomena that are most relevant to climate impacts of interest when
selecting the most appropriate downscaling tool. In the Mountain West,
complex terrain is the primary driver of climate, with midlatitude
cyclones, katabatic winds, monsoons, and associated air-mass
thunderstorms being the most prominent resulting phenomena. The maritime
climate along the Pacific coast also produces midlatitude cyclones, as
well as orographic lifting and atmospheric rivers (Kotamarthi et al.
2016).
For each of the above RCMs, we selected downscaled outputs run on
boundary conditions from two GCMs - GFDL-ESM2M (ECS = 2.4°C) and
HadGEM2-ES (ECS = 4.6°C) - to span the range of available models’
climate sensitivity (Appendix 2, Fig. A1). This approach is in keeping
with the recommendation from Kotamarthi et al. (2016) to use output from
multiple GCMs with different physical parameterizations to cover a
broader range of model uncertainty. Thus, in total, we consider four
possible climate futures (2 RCMs x 2 GCMs).
We used versions of these outputs that were bias-corrected using a
multivariate quantile mapping method (MBCn; Cannon 2018) with Daymet
temperatures as the observed dataset (Thornton et al. 2019). Because the
dynamically downscaled RCMs were still considerably coarser (25 km) than
our desired spatial resolution (1 km), we further statistically
downscaled them by spatially interpolating the data to 1 km and applying
an adiabatic lapse rate correction based on elevation (Wallace & Hobbs
2006).
To estimate survivorship under future conditions, we first derived
30-year means centered on the year 2050 for mean annual surface
temperature (MAST) and duration of the frost-free period for each of the
four climate scenarios. We then used projected MAST and a model linking
surface and subterranean temperatures (McClure et al. 2020) to estimate
the best available hibernaculum temperature likely to be available
(i.e., the temperature closest to the mean ambient temperature at which
each species has been observed during hibernation in the published
literature) in any given location for a given species (see Appendix 1
for details). Similarly, projected frost-free period was used to
estimate hibernation-specific winter duration (i.e., time between
immergence and emergence from hibernacula) as described in Hranac et al.
(accepted; also see Appendix 1). We then ran the bioenergetic
survivorship model for each of our five focal species under each future
scenario using these projected future climate parameters.
Projections of future winter survivorship under each scenario were then
used as predictors in species-specific SDMs that were previously derived
under current conditions. These SDMs are fully described in McClure et
al. (accepted), but briefly, we brought model-based, spatially-explicit
estimates of winter survivorship together with landscape attributes
hypothesized to influence hibernaculum selection (e.g., topography,
precipitation, presence of karst and mines) as predictors of relative
probability of occurrence throughout the states and territories
encompassing each species’ known range (National Atlas of the United
States 2011). We used boosted regression trees (Elith et al. 2008) to
link these predictors to our response data, which consisted of species
occurrence records compiled from multiple sources (e.g., online
databases of museum records and vetted observations, Natural Heritage
Programs, our own field studies). The influence of each predictor on
final predictive models for each species are summarized in Fig. 2. We
then applied the final model for each species to predictor values in
each 1-km cell to predict and map relative probability of occurrence.
Here, we essentially updated these models by replacing survivorship
estimates under current conditions with projected survivorship under
future scenarios. We then estimated and mapped the change in occurrence
probability between current conditions and each future scenario as the
difference in estimated occurrence probability for each raster cell.
Results &
Discussion
Mean projected climate parameters (MAST and frost-free period) among the
four climate scenarios assessed are mapped in Fig. 3, along with the
inter-scenario range and the mean projected change in each parameter
from current conditions. Spatial patterns in the mean parameter values
reflect latitudinal, topographic, and coastal influences on temperature
and frost-free period, as expected. We observed high agreement among
climate scenarios (i.e., low inter-scenario range) for projected MAST,
with increasing disagreement at very high latitudes. Disagreement among
climate scenarios in length of the frost-free period was higher in some
areas and more sporadic than that seen in MAST projections, which may
reflect a stronger influence of topography. Projected change in MAST
increased with latitude and with elevation, while projected change in
frost-free period was more spatially variable, with the largest
increases in the Appalachian region and localized portions of the West
coast.
Projected changes in probability of occurrence for each of five focal
species under future scenarios are mapped in Figs. 4-6 and Figs. A2-A3
(Appendix 2). We focus on projections from SDMs in which the
survivorship predictor accounted for at least 5% of the boosted
regression tree model fit under current conditions (McClure et al.
accepted; Fig. 2), which included models for M. californicus, M.
lucifugus, and P. subflavus. Projections from SDMs to which
survivorship contributed less than 5% (C. townsendii, M.
velifer) are expected to be less useful because little clear
relationship between known species occurrences and survivorship emerged.
Generally, probability of occurrence was projected to decline following
exposure to P. destructans (with the exception of C.
townsendii, Appendix 2, Fig. A2). However, projected occurrence
probability increased for most species in most places when climate
change was also considered. The greatest projected declines withP. destructans exposure were typically in areas with the highest
occurrence probability under current conditions (i.e., the areas
currently expected to be most suitable for a given species). Spatial
patterns in change in occurrence probability after considering climate
impacts were more variable. For M. californicus , we projected
moderate declines in occurrence probability in British Columbia, but a
strong increase in other high occurrence probability portions of the
range (Fig. 4). For M. lucifugus, we projected decreases in the
severity of declines, but climate change had little impact on areas
already expected to remain stable or experience increased occurrence
probability (Fig. 5). In contrast, we observed thresholding behavior inP. subflavus such that projected rangewide declines underP. destructans exposure were replaced by a marked increase in
occurrence probability in the southeast given climate change (Fig. 6).
This threshold appears to follow and is thus probably driven by spatial
patterns in the frost-free period (Fig. 3). We do not interpret
projected changes under each future scenario for C. townsendii orM. velifer because the low contribution of winter survivorship
estimates to SDM fits appear to result in unreliable and
counterintuitive behavior of models for these species (Appendix 2, Figs.
A2-A3).
It may be important to consider patterns in projected changes in
occurrence probability not just across the known range of each species,
but also more specifically at known hibernacula. We summarized projected
changes in relative probability of occurrence at the points of winter
capture or observation that informed development of species distribution
models (Table 1). For M. californicus and P. subflavus,the vast majority of winter locations are projected to exhibit decreased
occurrence probability with exposure to WNS (92.6 and 98.9% of
locations, respectively), but climate change scenarios reduce these
figures to 43.2 and 65.3%, respectively, on average. Thus, although
climate change is projected to significantly mitigate the impacts of WNS
on these species, approximately half of known hibernaculum locations may
still experience declines in occurrence. In the case of M.
lucifugus, WNS exposure is projected to result in decreased occurrence
probability at 41.2% of winter locations, and climate change is
anticipated to have little effect on this pattern (projected declines at
43.1% of locations, on average).
All four climate scenarios showed close agreement regarding future
changes in occurrence probability. This agreement may be driven by one
or more factors. First, derived estimates of MAST and frost-free period
may not be sensitive to differences among scenarios in projected daily
temperatures. This appears to be more likely for MAST than for
frost-free period (Fig. 3) and is not surprising given that calculation
of the frost-free period is threshold dependent (i.e., definition of the
frost-free period is dependent on the first and last day of the year on
which a precise threshold temperature is reached). Second, the
subterranean temperature model and/or winter duration model may not be
sensitive to MAST and frost-free period parameters, respectively (see
Fig. 1). This is unlikely in the case of the subterranean temperature
model, given that MAST is the model’s strongest predictor (McClure et
al. 2020). It is also unlikely in the case of the winter duration model
given that inclusion of frost-free period as a predictor improved the
model by 25.39 AIC units (Hranac et al. accepted). Third, the
survivorship model may not be sensitive to variation in the best
available temperature estimate derived from the subterranean temperature
model and/or our estimate of winter duration. We suggest that derivation
of the ‘best available’ temperature for a given species at a given
location from the subterranean temperature model likely absorbs the
majority of the variability among climate scenarios (see Appendix 1,
Hranac et al. accepted). Finally, for some species, SDMs may not be
sensitive to variation in winter survivorship estimates. SDM sensitivity
to survivorship is expected to be directly related to the contribution
of the survivorship predictor to the boosted regression tree model for a
given species (see McClure et al. accepted).
Although all climate scenarios produced very similar projections of
future change in occurrence probability, differences were apparent in
some places for most species. For M. californicus, differences
were most apparent along the Pacific coast near the California-Oregon
border and around the state of Oklahoma (Fig. 4). For P.
subflavus, the location of the threshold between increasing and
decreasing occurrence probability fluctuated across the Appalachian
region among scenarios (Fig. 6). Model disagreement was also evident in
Oklahoma for C. townsendii and M. velifer, as well as the
Columbia Plateau of eastern Washington and the Sierra Nevada range of
California, respectively (Appendix 2, Figs. A2-A3).
We suggest that our predictions of species distributions in the presence
of P. destructans and future climate conditions can help managers
to better anticipate the species- and place-specific impacts of these
stressors, individually and synergistically, across North America. Our
results may inform on-the-ground monitoring, which will be important for
efforts to track trends in bat distribution and abundance, such as the
North American Bat Monitoring Program (Loeb et al. 2015). Our results
may help to inform placement of passive acoustic detectors for
monitoring as P. destructans continues to spread and the climate
continues to warm. For example, monitoring of bat populations could be
targeted in areas where our projections suggest that suitable
hibernation conditions are likely to be lost and that occurrence
probability is likely to decline (vulnerable hibernacula). Conversely,
monitoring as well as protection efforts could target hibernacula that
are likely to be retained (potential refugia). Our predictions may also
enable assessment of the distribution of at-risk and stable hibernacula
across federal, state, and private lands to guide engagement strategies
for conservation. Additionally, they may help managers to prepare for
possible range expansions into or contractions from their jurisdictions
under future climate conditions.
Our findings suggest that by mid-century, changing temperatures may
offer a ‘rescue’ effect for many bat populations from the deleterious
effects of P. destructans. However, given the pace of P.
destructans’ spread from the East and its recent detection in New
Mexico and Montana (USFWS 2020), this rescue effect may arrive too late
for many hibernacula. Furthermore, a warming climate is not predicted to
shield all species in all areas (e.g., M. californicus in British
Columbia, M. lucifugus in mountainous regions, P.
subflavus in the northeastern United States), and climate change may
have other deleterious impacts on bats that are beyond the scope of our
models (e.g., increasing aridity, driving declines in insect
populations). It is therefore important that managers continue to strive
for effective proactive conservation strategies to combat the
devastating impacts of P. destructans as the fungus continues to
spread. Even in the absence of a ‘cure’ for WNS, conservation and
management actions that minimize other sources of mortality may allow
bat populations to persist long enough for conditions to improve.
Acknowledgements
This project has been funded in part with Federal funds from the
Department of Defense Environmental Research and Development Program
(SERDP), under Contract Number W912HQ-16-C-0015. DTSH is funded by Royal
Society Te Aparangi, grant number MAU1701. We are grateful to Linda
Mearns for input on appropriate selection and application of projected
climate data. We thank Nathan Justice, Eric Stofferahn, and Tony Chang
for valuable technical support. Finally, we are indebted to the many
individuals and organizations who generously provided raw data that made
possible the studies supporting this paper, including subterranean
microclimate data, bat species occurrence locations, hibernaculum
immergence and emergence observations, and bat physiology records (see
Haase et al. 2019, McClure et al. 2020, Hranac et al. accepted, and
McClure et al. accepted for further details).
Required disclosure: Any opinions, findings, and conclusions or
recommendations expressed in this publication are those of the author(s)
and do not necessarily reflect the views of the Government.
Literature
Cited
Adlard, R.D., Miller, T.L. and Smit, N.J., 2015. The butterfly effect:
parasite diversity, environment, and emerging disease in aquatic
wildlife. Trends in Parasitology, 31(4), pp.160-166.
Alexander, J.M., Diez, J.M., Hart, S.P. and Levine, J.M., 2016. When
climate reshuffles competitors: a call for experimental macroecology.
Trends in Ecology & Evolution, 31(11), pp.831-841.
Battisti, A. and Larsson, S., 2015. Climate change and insect pest
distribution range. Climate Change and Insect Pests. CABI, Wallingford,
pp.1-15.
Cannon, A.J., 2018. Multivariate quantile mapping bias correction: an
N-dimensional probability density function transform for climate model
simulations of multiple variables. Climate dynamics, 50(1-2), pp.31-49.
Colwell, R.K. and Rangel, T.F., 2009. Hutchinson’s duality: the once and
future niche. Proceedings of the National Academy of Sciences,
106(Supplement 2), pp.19651-19658.
Dawson, T.P., Jackson, S.T., House, J.I., Prentice, I.C. and Mace, G.M.,
2011. Beyond predictions: biodiversity conservation in a changing
climate. Science, 332(6025), pp.53-58.
Fisher, M.C., Henk, D.A., Briggs, C.J., Brownstein, J.S., Madoff, L.C.,
McCraw, S.L. and Gurr, S.J., 2012. Emerging fungal threats to animal,
plant and ecosystem health. Nature, 484(7393), pp.186-194.
Frick, W.F., Puechmaille, S.J. and Willis, C.K., 2016. White-nose
syndrome in bats. In Bats in the Anthropocene: Conservation of bats in a
changing world (pp. 245-262). Springer, Cham.
Gallana, M., Ryser-Degiorgis, M.P., Wahli, T. and Segner, H., 2013.
Climate change and infectious diseases of wildlife: altered interactions
between pathogens, vectors and hosts. Current Zoology, 59(3),
pp.427-437.
Giorgi, F., E. Coppola, F. Solmon, L. Mariotti, M.B. Sylla, X. Bi, N.
Elguindi, G.T. Diro, V. Nair, G. Giuliani, U.U. Turuncoglu. 2012.
RegCM4: model description and preliminary tests over multiple CORDEX
domains. Climate Research 52: 7-29.
Haase, C.G., Fuller, N.W., Hranac, C.R., Hayman, D.T., McGuire, L.P.,
Norquay, K.J., Silas, K.A., Willis, C.K., Plowright, R.K. and Olson,
S.H., 2019. Incorporating evaporative water loss into bioenergetic
models of hibernation to test for relative influence of host and
pathogen traits on white-nose syndrome. PloS one, 14(10), p.e0222311.
Harvey, M.J., Altenbach, J.S., & Best, T.L. (2013). Bats of the United
States and Canada. Johns Hopkins University Press.
Johnson, J.S., Reeder, D.M., McMichael III, J.W., Meierhofer, M.B.,
Stern, D.W., Lumadue, S.S., Sigler, L.E., Winters, H.D., Vodzak, M.E.,
Kurta, A. and Kath, J.A., 2014. Host, pathogen, and environmental
characteristics predict white-nose syndrome mortality in captive little
brown myotis (Myotis lucifugus). PLoS One, 9(11), p.e112502.
Kiehl, J.T. and Ramanathan, V. eds., 2006. Frontiers of climate
modeling. Cambridge University Press.
Kotamarthi, R., L. Mearns, K. Hayhoe, C. Castro, D. Wuebbles. 2016. Use
of climate information for decision-making and impacts research: state
of our understanding. Prepared for the Department of Defense, Strategic
Environmental Research and Development Program. 55pp.
Langwig, K.E., Frick, W.F., Hoyt, J.R., Parise, K.L., Drees, K.P., Kunz,
T.H., Foster, J.T. and Kilpatrick, A.M., 2016. Drivers of variation in
species impacts for a multi-host fungal disease of bats. Philosophical
Transactions of the Royal Society B: Biological Sciences, 371(1709),
p.20150456.
Langwig, K.E., Frick, W.F., Bried, J.T., Hicks, A.C., Kunz, T.H. and
Marm Kilpatrick, A., 2012. Sociality, density‐dependence and
microclimates determine the persistence of populations suffering from a
novel fungal disease, white‐nose syndrome. Ecology letters, 15(9),
pp.1050-1057.
Lawler, J.J., Shafer, S.L., White, D., Kareiva, P., Maurer, E.P.,
Blaustein, A.R. and Bartlein, P.J., 2009. Projected climate‐induced
faunal change in the Western Hemisphere. Ecology, 90(3), pp.588-597.
Leopardi, S., Blake, D. and Puechmaille, S.J., 2015. White-nose syndrome
fungus introduced from Europe to North America. Current Biology, 25(6),
pp.R217-R219.
Loarie, S.R., Duffy, P.B., Hamilton, H., Asner, G.P., Field, C.B. and
Ackerly, D.D., 2009. The velocity of climate change. Nature, 462(7276),
pp.1052-1055.
Marroquin, C.M., Lavine, J.O. and Windstam, S.T., 2017. Effect of
humidity on development of Pseudogymnoascus destructans, the causal
agent of bat white-nose syndrome. Northeastern Naturalist, 24(1),
pp.54-64.
McClure, M.L., Crowley, D., Haase, C.G., McGuire, L.P., Fuller, N.W.,
Hayman, D.T.S., Lausen, C.L., Plowright, R.K. Dickson, B.G. and Olson
S.H. 2020. Linking surface and subterranean climate: implications for
the study of hibernating bats and other cave dwellers. Ecosphere
11(10):e03274.
Mearns, L.O., et al. 2017. The NA-CORDEX dataset, version 1.0. NCAR
Climate Data Gateway, Boulder CO. https://doi.org/10.5065/D6SJ1JCH
Melles, S.J., Fortin, M.J., Lindsay, K. and Badzinski, D., 2011.
Expanding northward: influence of climate change, forest connectivity,
and population processes on a threatened species’ range shift. Global
Change Biology, 17(1), pp.17-31.
Metcalf, C.J.E., Walter, K.S., Wesolowski, A., Buckee, C.O.,
Shevliakova, E., Tatem, A.J., Boos, W.R., Weinberger, D.M. and Pitzer,
V.E., 2017. Identifying climate drivers of infectious disease dynamics:
recent advances and challenges ahead. Proceedings of the Royal Society
B: Biological Sciences, 284(1860), p.20170901.
Mitchell, T.D., P.D. Jones. 2005. An improved method of constructing a
database of monthly climate observations and associated high-resolution
grids. International Journal of Climatology 25:693–712.
Moore, M.S., Field, K.A., Behr, M.J., Turner, G.G., Furze, M.E., Stern,
D.W., Allegra, P.R., Bouboulis, S.A., Musante, C.D., Vodzak, M.E. and
Biron, M.E., 2018. Energy conserving thermoregulatory patterns and lower
disease severity in a bat resistant to the impacts of white-nose
syndrome. Journal of Comparative Physiology B, 188(1), pp.163-176.
Opdam, P. and Wascher, D., 2004. Climate change meets habitat
fragmentation: linking landscape and biogeographical scale levels in
research and conservation. Biological conservation, 117(3), pp.285-297.
Pachauri, R.K., Allen, M.R., Barros, V.R., Broome, J., Cramer, W.,
Christ, R., Church, J.A., Clarke, L., Dahe, Q., Dasgupta, P. and Dubash,
N.K., 2014. Climate change 2014: synthesis report. Contribution of
Working Groups I, II and III to the fifth assessment report of the
Intergovernmental Panel on Climate Change. IPCC.
Parmesan, C., 2006. Ecological and evolutionary responses to recent
climate change. Annu. Rev. Ecol. Evol. Syst., 37, pp.637-669.
Reeder, D.M., Frank, C.L., Turner, G.G., Meteyer, C.U., Kurta, A.,
Britzke, E.R., Vodzak, M.E., Darling, S.R., Stihler, C.W., Hicks, A.C.
and Jacob, R., 2012. Frequent arousal from hibernation linked to
severity of infection and mortality in bats with white-nose syndrome.
PLoS One, 7(6), p.e38920.
Skamarock, W.C., J.B. Klemp, J. Dudhia, D.O. Gill, D.M. Barker, M.G.
Duda, X. Huang, W. Wang, J.G. Powers. 2008. A Description of the
Advanced Research WRF Version 3. NCAR Technical Note NCAR/TN–475+STR.
125pp.
Thomas, D.W., Dorais, M. and Bergeron, J.M., 1990. Winter energy budgets
and cost of arousals for hibernating little brown bats, Myotis
lucifugus. Journal of mammalogy, 71(3), pp.475-479.
Thomas, C.D., Cameron, A., Green, R.E., Bakkenes, M., Beaumont, L.J.,
Collingham, Y.C., Erasmus, B.F., De Siqueira, M.F., Grainger, A.,
Hannah, L. and Hughes, L., 2004. Extinction risk from climate change.
Nature, 427(6970), pp.145-148.
Thornton, P.E., Thornton, M.M., Mayer, B.W., Wei, Y., Devarakonda, R.,
Vose, R.S., and Cook, R.B., 2019. Daymet: Daily Surface Weather Data on
a 1-km Grid for North America, Version 3. ORNL DAAC, Oak Ridge,
Tennessee, USA.
Thuiller, W., Lavorel, S., Araújo, M.B., Sykes, M.T. and Prentice, I.C.,
2005. Climate change threats to plant diversity in Europe. Proceedings
of the National Academy of Sciences, 102(23), pp.8245-8250.
Travis, J.M.J., 2003. Climate change and habitat destruction: a deadly
anthropogenic cocktail. Proceedings of the Royal Society of London.
Series B: Biological Sciences, 270(1514), pp.467-473.
Urban, M.C., Zarnetske, P.L. and Skelly, D.K., 2013. Moving forward:
dispersal and species interactions determine biotic responses to climate
change. Annals of the New York Academy of Sciences, 1297(1), pp.44-60.
Verant, M.L., Boyles, J.G., Waldrep Jr, W., Wibbelt, G. and Blehert,
D.S., 2012. Temperature-dependent growth of Geomyces destructans, the
fungus that causes bat white-nose syndrome. PloS one, 7(9), p.e46280.
Wallace, J.M. and Hobbs, P.V., 2006. Atmospheric science: an
introductory survey (Vol. 92). Elsevier.
Tables
Table 1. Percent of points of
winter capture/observation for each of five focal species that are
projected to exhibit decreased relative probability of occurrence
following exposure to WNS and climate change. Future climate scenarios
were driven by each combination of two global circulation models (GCMs):
GFDL-ESM2M (Scenarios 1, 2) and HadGEM2-ES (Scenarios 3, 4) and two
dynamically-downscaled regional climate models (RCMs): RegCM4 (Scenarios
1, 3) and WRF (Scenarios 2, 4).