Download Research Note Effects of Warming Conditions in Eastern North

Research Note
Effects of Warming Conditions in Eastern North
American Forests on Red-Backed Salamander
Morphology
JAMES P. GIBBS∗ AND NANCY E. KARRAKER
College of Environmental Science and Forestry, State University of New York, 250 Illick Hall, 1 Forestry Drive, Syracuse,
New York 13210, U.S.A.
Abstract: Several studies have reported climate-associated changes in phenotypically plastic traits of amphibians, yet it remains unknown whether amphibians can manifest an evolutionary response to global climate
change at the rate and magnitude that it is occurring. To assess this issue, we examined temporal change in
the morphology of the red-backed salamander (Plethodon cinereus), a small, abundant woodland salamander
distributed widely in eastern North America with two distinct morphotypes: striped individuals associated
with cooler microclimates and unstriped individuals associated with warmer microclimates. We compiled
morph frequencies for 50,960 individual salamanders from 558 sites as recorded in the published literature
and in unpublished field notes of herpetologists between 1908 and 2004. We observed that striping probability
increased with increasing latitude, longitude, and elevation and decreased ( from 80% to 74% range wide) with
time. The combined forces of regional climate warming and, particularly, forest disturbance have evidently
been sufficient to cause morphological evolution in this amphibian over the last century.
Keywords: amphibians, climate change, evolution, forest disturbance, morphology, Plethodon cinereus
Efectos de las Condiciones de Calentamiento en Bosques Orientales de Norte América sobre la Morfologı́a de
Plethodon cinereus
Resumen: Numerosos estudios han reportado cambios asociados con el clima en las caracterı́sticas
fenotı́picamente plásticas de los anfibios, sin embargo aun no se conoce si los anfibios pueden manifestar
una respuesta evolutiva al cambio climático global en la velocidad y magnitud con que está ocurriendo. Para
evaluar este tema, examinamos el cambio temporal en la morfologı́a de Plethodon cinereus, una salamandra
pequeña, abundante y ampliamente distribuida en bosques de este de Norte América con dos morfotipos
distintos: individuos rayados asociados con microclimas más frescos e individuos sin rayas asociados con microclimas más cálidos. Compilamos las frecuencias de los morfotipos para 50,960 salamandras individuales
de 558 sitios registrados en la literatura publicada y en notas de campo no publicadas de herpetólogos entre
1908 y 2004 Observamos que la probabilidad de individuos con rayas incrementó con el incremento de la
latitud, longitud y elevación y disminuyó (de 80% a 74% en toda el área de distribución) con el tiempo. Las
fuerzas combinadas del calentamiento climático regional y, particularmente, la perturbación de bosques han
sido suficientes evidentemente para causar la evolución morfológica en este anfibio durante el último siglo.
Palabras Clave: anfibios, cambio climático, evolución, morfologı́a, perturbación de bosques, Plethodon cinereus
∗ email
[email protected]
Paper submitted April 7, 2005; revised manuscript accepted September 13, 2005.
913
Conservation Biology Volume 20, No. 3, 913–917
C 2006 Society for Conservation Biology
DOI: 10.1111/j.1523-1739.2006.00375.x
914
Morphological Change in Salamanders
Introduction
Responses of plants and animals to global climate change
are increasingly being evaluated (Root et al. 2003): 62% of
nearly 700 species examined have exhibited shifts in phenological timing and 80% of nearly 450 species showed
changes in range boundaries in the direction predicted
by climate change (Parmesan & Yohe 2003). Amphibians
are of particular interest because environmental temperatures exert strong control over all aspects of their biology. For example, studies at two sites in southern England
showed that two species of frogs and three species of salamanders spawned progressively earlier from 1979 to 1994
and that these patterns were correlated with increasing
winter and spring average temperatures (Beebee 1995).
Timing of calling of four frog species in New York occurred 10–13 days earlier between 1990 and 1999 than
between 1900 and 1912, and this was correlated with increases in mean maximum daily temperatures during the
century (Gibbs & Breisch 2001). In contrast, Blaustein et
al. (2001) concluded from a review of decades-long time
series of calling phenology for four North American anurans that no consistent patterns toward early breeding
had yet emerged.
Whether amphibians can manifest an evolutionary response to climate changes at the magnitude and rate they
are occurring is unknown. The issue is germane because
understanding it can inform our prognosis of the likely status of amphibians in the face of ongoing climate change.
We therefore evaluated it using the eastern red-backed
salamander (Plethodon cinereus) as a model species. P.
cinereus is abundant in the litter layer of mature forests
of eastern North America and has two primary morphotypes: a completely dark morph ( black or unstriped)
and one with a dorsal red or occasionally yellow stripe
(striped). A third, completely red morph (erythristic) occurs sporadically (e.g., Lotter & Scott 1977). The two
primary morphs interbreed freely and the polymorphism
has a genetic basis (Rabb 1955; Highton 1975). Studies
of spatial variation in the morph frequencies clearly indicate that the unstriped morph is more closely associated with warmer, drier climates than the striped morph
(e.g., Burger 1935; Test 1952; Williams et al. 1968; Lotter
& Scott 1977). Moreover, maintenance of the polymorphism by temperature-related selection on the colorationassociated character is suggested by three lines of evidence: (1) age-specific morph frequencies indicate that
unstriped morphs experience higher mortality in colder
than warmer sites (Lotter & Scott 1977; Moreno 1989),
(2) unstriped morphs retreat from the surface earlier than
striped morphs in the fall season (Lotter & Scott 1977;
Moreno 1989), and (3) unstriped individuals have lower
standard metabolic rates than striped ones at 15◦ C (the
temperature at which morph behaviors are also most divergent), which likely permits unstriped individuals to be
more active at warmer, drier sites (Moreno 1989).
Conservation Biology
Volume 20, No. 3, June 2006
Gibbs & Karraker
Extremely high local densities of P. cinereus (up to 2.8
individuals/m2 [Mathis 1991]) and relatively unambiguous coloration patterns permit rapid, reliable estimation
of local morph frequencies. Extensive documentation of
morph frequencies have been made since the rise of scientific herpetology in North America around 1900, during which time many hundreds of reports, each typically
based on large samples of individuals in any given local
population, have accumulated in the literature. Given the
available data, we sought to evaluate the stability of the
morph frequencies of P. cinereus in the context of ongoing changes in climate across its range. Specifically, we
compiled morph identities for >50,000 individual salamanders reported between 1908 and 2004 and examined whether a temporal trend was evident in P. cinereus
morph frequencies.
Methods
We attempted to locate every published field estimate of
P. cinereus morph ratios ( based on a sample size of ≥10
[median sample size/report = 31, mean = 91.2]) from
throughout the species’ range. We examined all citations
of the species in Cambridge Scientific Abstracts and Biblioline’s Wildlife Worldwide and requested via three major
herpetology-oriented computer “listservs” contributions
of unpublished data from herpetologists active in plethodontid salamander research. Much of the older literature
(pre-1950) was identified by inspecting the bibliographies of early theses and treatises on salamanders (e.g.,
Dunn 1926; Bishop 1941; Thurow 1955). For each sitelevel report of morph frequency we recorded study location, latitude, longitude, elevation, number of striped and
unstriped individuals (erythristic individuals were rare
[<1%] and were pooled with striped), and sampling period (from which operative year of study was calculated
as the average of the initial and final year of the study).
When not provided directly from a published report or
the contributors, site latitude, longitude, and elevation
were determined from U.S. Geological Survey topographical maps (1:24,000 scale) or Canadian National Topographic System maps (1:50,000 scale). Reports were excluded if they were based on museum specimens, which
are typically drawn from widely scattered areas. Information sources, too voluminous to be included here in an
appendix, are available from the authors.
To evaluate historical changes in P. cinereus morphology, we controlled for primary drivers of spatial variation
in ambient temperature to focus on the potential contribution of time to variation in morph frequencies. To do
so, we used probit analysis to relate salamander morph
frequency at a given site to elevation, latitude, longitude,
and operative year of study. We applied a logit (log odds)
transformation to the response proportion following Collett (1991) with PROC PROBIT of the Statistical Analysis
Gibbs & Karraker
Morphological Change in Salamanders
915
Table 1. Probit analysis of contribution of latitude, longitude, elevation, and year of study to probability
of a P. cinereus being striped, based on an analysis of 50,960 individual salamanders drawn from 558
populations across the species’ range between 1908 and 2004.
Parameter
Estimate
Intercept
Latitude
Longitude
Elevation
Year
−2.041
0.221
0.0103
0.0013
−0.0029
95% confidence limits
−2.119
0.220
0.0101
0.0013
−0.0030
System (version 8, SAS Institute, Cary, North Carolina).
Elevation, latitude, and longitude were included as controls because in a linear regression model these three variables together explain virtually all (97%) of the spatial
variation in annual average temperature by site over the
last 100 years at 269 climate stations within the species’
range (equation: mean annual temperature (◦ C) = 50.51
− 1.07∗ latitude − 0.050∗ longitude − 0.0013∗ elevation;
adjusted R2 = 0.967; F = 2090.82; p < 0.001). Historical temperature data were obtained from the U.S. Historical Climate Network and the Meteorological Service
of Canada for all sites that occurred within the range
of P. cinereus (Petranka 1998) that had ≥80% complete
monthly records of average temperature for 1900–1999.
We used these same climate data to estimate whether
mean annual temperatures had increased within the range
of P. cinereus. To do so, we performed a linear regression on mean annual temperature against year for 1900–
1999 to estimate station-level trends, weighted site trend
estimates by the area of the Thiessen polygon (see DeMers 2004) that each station occupied within the species’
range (ArcView, version 3.2, Redlands, California), and
then compared mean trend across sites with a two-tailed
z test against the null hypothesis that mean trend = 0 (Zar
1984).
−1.964
0.221
0.0105
0.0013
−0.0029
χ2
p
2660
837,790
11,326
1,024,366
26,931
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
ical significance of the temporal trend, holding all model
parameters constant at their mean values while varying
year from 1900 to 2000 indicated that the predicted probability of an individual P. cinereus being striped declined
across the species’ range by 6%—from 80% to 74%—over
the last century (Fig. 1).
Discussion
Our results suggest that morph frequencies of P. cinereus
have shifted over the last century across the species’
range. Although some of the warming of P. cinereus habitats may have a regional and even global component
(Hansen et al. 2001), we speculate that site-level warming
due to forest disturbance (e.g., Whitney 1994) has played
a far greater role. More specifically, forests throughout
the range of P. cinereus have been exploited heavily over
the last three centuries. In the “north region” as defined
by the U.S. Forest Service (which includes much of P.
cinereus’ range) aggregate forest cover declined to 47%
Results
Mean trend in annual temperature at the 269 historical
climate stations was 0.00675◦ C/year (± 0.000525 1SE)
and significantly different (t = 12.866, df = 268, p <
0.001) from zero, suggesting a 0.7◦ C warming has occurred within the range of P. cinereus over the last century. Probit analysis of the site-level morph frequencies of
P. cinereus (n = 558 sites and 50,960 individual salamanders) indicated that striping probability increased with
latitude, longitude (in an easterly direction), and elevation, and decreased with time (Table 1, Fig. 1). To evaluate possible skewing caused by unequal sample sizes
among reports, each of the major studies (≥1000 individuals) was iteratively dropped from the analysis. In all
reanalyses, parameter estimates varied but parameter significance and signs of estimates did not. In terms of biolog-
Figure 1. Relationship between probability of a P.
cinereus being striped—P(striped)—and latitude,
longitude, elevation, and year of study. Magnitude of
each variable is scaled 0–1 relative to its possible
range of values (see key).
Conservation Biology
Volume 20, No. 3, June 2006
916
Morphological Change in Salamanders
of its original extent between 1630 and 1909 and then recovered to 57% by 1997 (USDA 2001). Moreover, the forest remaining is structurally quite different. For example,
in many eastern states remaining old-growth forests with
ecological conditions comparable to what P. cinereus
evolved under represent a tiny fraction of original forests
(e.g., New York [0.36%, Ketchledge 1965], Pennsylvania
[0.05%; Smith 1989], Indiana [<0.01%; Parker 1989] and
Illinois [0.03%; Parker 1989]).
Microhabitats of old-growth forest differ dramatically
from secondary forest. They are structurally more complex, and at the forest floor, where P. cinereus lives, they
are darker, cooler, and moister (e.g., Matlack 1993; McGee
et al. 1999). Increases in average soil temperatures associated with forest disturbance can be dramatic (e.g., increases from undisturbed forests to partial cuts of 3◦ C and
to clearcuts of 6◦ C; Jemison 1934). Pfingsten and Walker
(1978) reported that frequency of the unstriped morph
of P. cinereus tended to increase significantly more at
forested sites disturbed between repeated surveys than at
those that remain undisturbed. Given the strong selection
intensity (i.e., substantial changes in soil temperatures
from forest disturbance) and duration and extent of selection (forest disturbance widespread over 1–2 centuries
throughout species’ range versus the generation time of
Plethodon species [ca. 10 years for P. jordani, Hairston
1983]), the magnitude of morphological change observed
(ca. 6% decline in striping frequency range wide) is not
unexpected.
We consider it unlikely that the temporal trends in
morph frequencies resulted from the haphazard nature
of sampling inherent in a study such as this. Two of the
variables included in the analysis for biological reasons—
latitude and longitude—also largely control for the nonrandom nature of the sampling pattern at the geographic
scale at which the study was conducted. Nevertheless,
complementary approaches could be used to evaluate the
changes in P. cinereus morphology reported here. One
approach is to “mine” museum collections. The striping pattern remains distinct in preserved specimens and
many major museums possess extensive holdings of this
species. Many museum specimens were collected in
“bulk” (e.g., Lotter 1975), which reduces the probability
that collections would be skewed toward atypical individuals collected as curiosities, and collections extend to
the early 1800s. A second approach would be to resurvey field populations at sites where large numbers of P.
cinereus have been collected previously. For example,
resurveys of Lotter and Scott’s (1977) 50 sampling sites
(at each of which 100 individuals were sampled) span
a wide range of climatic conditions and would provide a
statistically powerful approach to monitoring change owing to the paired nature of the sampling design and the
decades-long monitoring window.
Our results have implications for the ongoing debate on
global amphibian declines (e.g., Stuart et al. 2004), which
Conservation Biology
Volume 20, No. 3, June 2006
Gibbs & Karraker
have been attributed to a variety of causes. Whether amphibians are being affected at large spatial scales by ongoing climate change is an area that needs much research
(Carey & Alexander 2003). A particular challenge has
been identifying data sets of sufficient spatial and temporal scale to address the issue (Blaustein et al. 2001). Our
data set for P. cinereus suggests that the species may be
undergoing large-scale, long-term morphological change
possibly as a result of ongoing shifts in ambient temperatures across its range. Because changes in soil temperature do not appear to strongly affect overall abundance
of P. cinereus (Wyman & Hawksley-Lescault 1987; Young
& Yahner 2003), the main implication of our findings is
that P. cinereus may serve as an important indicator of
changes that other amphibians are experiencing that occupy similar geographical range but whose reproduction
and hence persistence are more closely associated with
changes in ambient temperature, such as the mink frog
(Rana septentrionalis) (Hedeen 1986).
Acknowledgments
We are particularly grateful to the many herpetologists
who kindly contributed morph frequencies from their unpublished field records: J. Crawford, S. Droege, R. Jung, A.
Kuhns, D. Lantz, V. Meretsky, M. Meyer, J. Nagel, M. Oldham, E. Roberts, K. Roe, F. W. Scheuler, S. Spohr, C. Strojny,
M. Takahashi, L. Weir, and M. Whitney. M. Buff assisted
with manipulating Canadian digital elevation models.
Literature Cited
Beebee, T. J. C. 1995. Amphibian breeding and climate. Nature 374:219–
220.
Bishop, S. C. 1941. The salamanders of New York. New York State Museum Bulletin 324.
Blaustein, A. R., L. K. Belden, Olson, D. H., D. M. Green, T. L. Root,
and J. M. Kiesecker. 2001. Amphibian breeding and climate change.
Conservation Biology 15:1804–1809.
Burger, J. W. 1935. Plethodon cinereus (Green) in eastern Pennsylvania
and New Jersey. The American Naturalist 69:578–586.
Carey, C., and M. A. Alexander. 2003. Climate change and amphibian
declines: is there a link? Diversity and Distributions 9:111–121.
Collett, D. 1991. Modelling binary data. Chapman and Hall, London.
DeMers, M. N. 2004. Fundamentals of geographic information systems.
3rd edition. J. Wiley, Hoboken, New Jersey.
Dunn, E. R. 1926. The salamanders of the family Plethodontidae. Smith
College, Northampton, Massachusetts.
Gibbs, J. P., and A. R. Breisch. 2001. Climate warming and calling phenology of frogs near Ithaca, New York, 1900–1999. Conservation
Biology 15:1175–1178.
Hairston, N. G. 1983. Growth, survival, and reproduction of Plethodon
jordani: trade-offs between selective pressures. Copeia 1983:1024–
1035.
Hansen, J. E., R. Ruedy, M. Sato, M. Imhoff, W. Lawrence, D. Easterling,
T. Peterson, and T. Karl. 2001. A closer look at United States and
global surface temperature change. Journal of Geophysical Research
106:23947–23963.
Gibbs & Karraker
Hedeen, S. E. 1986. The southern geographic limit of the mink frog,
Rana septentrionalis. Copeia 1986:239–244.
Highton, R. 1975. Geographic variation in genetic dominance of the
color morphs of the red-backed salamander, Plethodon cinereus.
Genetics 80:363–374.
Jemison, G. M. 1934. The significance of the effect of stand density upon
weather beneath the canopy. Journal of Forestry 32:446–451.
Ketchledge, E. H. 1965. Changes in the forests of New York. The [New
York] Conservationist 19:29–34.
Lotter, C. F. 1975. Multiple parameter differentiation in a polymorphic
salamander. Ph.D. thesis. University of Connecticut, Storrs.
Lotter, F., and N. J. Scott Jr. 1977. Correlation between climate and distribution of the color morphs of the salamander Plethodon cinereus.
Copeia 1977:681–690.
Mathis, A. 1991. Territories of male and female terrestrial salamanders: costs, benefits, and intersexual spatial associations. Oecologia
86:433–440.
Matlack, G. R. 1993. Microenvironment variation within and among
forest edge sites in the eastern United States. Biological Conservation
66:185–194.
McGee, G. G., D. J. Leopold, and R. D. Nyland. 1999. Structural characteristics of old-growth, maturing, and managed northern hardwood
stands: implications for forest ecosystem management. Ecological
Applications 9:1316–1329.
Moreno, G. 1989. Behavioral and physiological differentiation between
the color morphs of the salamander, Plethodon cinereus. Journal of
Herpetology 23:335–341.
Parker, G. R. 1989. Old-growth forests of the central hardwood region.
Natural Areas Journal 9:5–11.
Parmesan C., and G. Yohe. 2003. A globally coherent fingerprint
of climate change impacts across natural systems. Nature 421:
37–42.
Petranka, J. W. 1998. Salamanders of the United States and Canada. Smithsonian Institution Press, Washington, D.C.
Pfingsten, R. A., and C. F. Walker. 1978. Some nearly all black popula-
Morphological Change in Salamanders
917
tions of Plethodon cinereus (Amphibia, Urodela, Plethodontidae) in
northern Ohio. Journal of Herpetology 12:163–167.
Rabb, G. B. 1955. Observations on the identity of the salamander
Plethodon huldae. Copeia 1955:261–262.
Root, T. L., J. T. Price, K. R. Hall, S. H. Schneider, C. Rosenzweig, and
J. A. Pounds. 2003. Fingerprints of global warming on wild animals
and plants. Nature 421:57–60.
Smith, T. L. 1989. An overview of old-growth forests in Pennsylvania.
Natural Areas Journal 9:40–44.
Stuart, S. N., J. S. Chanson, N. A. Cox, B. E. Young, A. S. L. Rodrigues, D.
L. Fischman, and R. W. Waller. 2004. Status and trends of amphibian
declines and extinctions worldwide. Science 306:1783–1786.
Test, F. H. 1952. Spread of the black phase of the red-backed salamander
in Michigan. Evolution 6:197–203.
Thurow, G. R. 1955. Taxonomic and ecological studies on the zig-zag
(Plethodon dorsalis) and the red-backed salamander (Plethodon
cinereus). Ph.D. thesis. Indiana University, Bloomington.
USDA (U.S. Department of Agriculture). 2001. U.S. forest facts and historical trends. USDA Forest Service, Washington, D.C. Available from
http://fia.fs.fed.us/library/ForestFacts.pdf (accessed May 2005).
Whitney, G. G. 1994. From coastal wilderness to fruited plain: a history
of environmental change in temperate North America from 1500 to
the present. Cambridge University Press, Cambridge, United Kingdom.
Williams, E. E., R. Highton, and D. M. Cooper. 1968. Breakdown of polymorphism of the red-backed salamander on Long Island. Evolution
22:76–86.
Wyman, R. L., and D. S. Hawksley-Lescault. 1987. Soil activity affects
distribution, behavior, and physiology of the salamander Plethodon
cinereus. Ecology 68:1819–1827.
Young, G. I., and R. H. Yahner. 2003. Distribution of, and microhabitat
use by, woodland salamanders along forest-farmland edges. Canadian
Field-Naturalist 117:19–24.
Zar, J. H. 1984. Biostatistical analysis. 2nd edition. Prentice Hall, New
Jersey.
Conservation Biology
Volume 20, No. 3, June 2006