Rapid Range Shifts of Species Associated with High Levels of Climate Warming

Abstract

The distributions of many terrestrial organisms are currently shifting in latitude or elevation in response to changing climate.
Using a meta-analysis, we estimated that the distributions of species have recently shifted to higher elevations at a median
rate of 11.0 meters per decade, and to higher latitudes at a median rate of 16.9 kilometers per decade. These rates are approximately
two and three times faster than previously reported. The distances moved by species are greatest in studies showing the highest
levels of warming, with average latitudinal shifts being generally sufficient to track temperature changes. However, individual
species vary greatly in their rates of change, suggesting that the range shift of each species depends on multiple internal
species traits and external drivers of change. Rapid average shifts derive from a wide diversity of responses by individual
species.

Full text

15. A. Woolfe et al., PLoS Biol. 3, e7 (2005).
16. A. Siepel et al., Genome Res. 15, 1034 (2005).
17. L. Z. Holland et al., Genome Res. 18, 1100 (2008).
18. Materials and methods are available as supporting
material on Science Online.
19. K. S. Pollard, M. J. Hubisz, K. R. Rosenbloom, A. Siepel,
Genome Res. 20, 110 (2010).
20. International HapMap Consortium, Nature 449, 851
(2007).
21. G. Robertson et al., Nat. Methods 4, 651 (2007).
22. G. E. Crawford et al., Genome Res. 16, 123 (2006).
23. A. P. Boyle et al., Cell 132, 311 (2008).
24. A. Valouev et al., Nat. Methods 5, 829 (2008).
25. M. Ashburner et al., Nat. Genet. 25, 25 (2000).
26. C. Y. McLean et al., Nat. Biotechnol. 28, 495
(2010).
27. C. J. Bult, Nucleic Acids Res. 36, D724 (2008).
28. P. Wu et al., Int. J. Dev. Biol. 48, 249 (2004).
Acknowledgments: This work was supported by the Howard
Hughes Medical Institute (C.B.L., S.R.S., D.M.K., D.H.),
the NSF (CAREER-0644282 to M.K., DBI-0644111 to
A.S.), the NIH (R01-HG004037 to M.K., P50- HG02568
to D.M.K., U54-HG003067 to K.L-T., 1U01-HG004695
to C.B.L., 5P41-HG002371to B.J.R.), the Sloan
Foundation (M.K.), and the European Science Foundation
(EURYI to K.L-T.).
Supporting Online Material
www.sciencemag.org/cgi/content/full/333/6045/1019/DC1
Materials and Methods
Figs. S1 to S9
Tables S1 to S12
References (29–49)
10 January 2011; accepted 24 June 2011
10.1126/science.1202702
Rapid Range Shifts of Species
Associated with High Levels
of Climate Warming
I-Ching Chen,
1,2
Jane K. Hill,
1
Ralf Ohlemüller,
3
David B. Roy,
4
Chris D. Thomas
1
*
The distributions of many terrestrial organisms are currently shifting in latitude or elevation in response
to changing climate. Using a meta-analysis, we estimated that the distributions of species have
recently shifted to higher elevations at a median rate of 11.0 meters per decade, and to higher latitudes
at a median rate of 16.9 kilometers per decade. These rates are approximately two and three times
faster than previously reported. The distances moved by species are greatest in studies showing the
highest levels of warming, with average latitudinal shifts being generally sufficient to track temperature
changes. However, individual species vary greatly in their rates of change, suggesting that the
range shift of each species depends on multiple internal species traits and external drivers of change.
Rapid average shifts derive from a wide diversity of responses by individual species.
Threats to global biodiversity from climate
change (1-8) make it important to identify
the rates at which species have already
responded to recent warming. There is strong evi-
dence that species have changed the timing of
their life cycles during the year and that this is
linked to annual and longer-term variations in
temperature (9–12). Many species have also
shifted their geographic distributions toward
higher latitudes and elevations (13–17), but this
evidence has previously fallen short of demon-
strating a direct link between temperature change
and range shifts; that is, greater range shifts have
not been demonstrated for regions with the high-
est levels of warming.
We undertook a meta-analysis of available
studies of latitudinal (Europe, North America,
and Chile) and elevational (Europe, North Amer-
ica, Malaysia, and Marion Island) range shifts for
a range of taxonomic groups (18)(tableS1).We
considered N= 23 taxonomic group × geographic
region combinations for latitude, incorporating
764 individual species responses, and N=31
taxonomic group × region combinations for ele-
vation, representing 1367 species responses. For
the purpose of analysis, the mean shift across all
species of a given taxonomic group, in a given
region, was taken to represent a single value (for
example, plants in Switzerland or birds in New
York State; table S1) (18).
The latitudinal analysis revealed that spe-
cies have moved away from the Equator at a
median rate of 16.9 km decade
−1
(mean = 17.6
km decade
−1
, SE = 2.9, N= 22 species group ×
region combinations, one-sample ttest versus
zero shift, t= 6.10, P< 0.0001). Weighting each
study by the √(numberofspecies)inthegroup×
region combination gave a mean rate of 16.6 km
decade
−1
. For elevation, there was a median shift
to higher elevations of 11.0 m uphill decade
−1
(mean = 12.2 m decade
−1
, SE = 1.8, N=30spe-
cies groups × regions, one-sample ttest versus
zero shift, t= 7.04, P< 0.0001). Weighting ele-
vation studies by √(number of species) gave a
mean rate of uphill movement of 11.1 m decade
−1
.
A previous meta-analysis (14)ofdistribu-
tion changes analyzed individual species, rather
than the averages of taxonomic groups × regions
that we used, and also included data on latitu-
dinal and elevational shifts in the same analysis
(18). It concluded that ranges had shifted toward
higher latitudes at 6.1 km decade
−1
andtohigh-
er elevations at 6.1 m decade
−1
(14), whereas
the rates of range shift that we found were sig-
nificantly greater [N= 22 species groups × regions,
one-sample ttest versus 6.1 km decade
−1
,t=
3.99, P= 0.0007 for latitude; N= 30 groups ×
regions, one-sample ttest versus 6.1 m decade
−1
,
t= 3.49, P= 0.002 for elevation (18)]. Our
estimated mean rates are approximately three
and two times higher than those in (14), for
1
Department of Biology, University of York, Wentworth Way,
York YO10 5DD, UK.
2
Biodiversity Research Center, Academia
Sinica, 128 Academia Road, Section 2, Nankang Taipei 115,
Taiwan.
3
School of Biological and Biomedical Sciences, and
Institute of Hazard, Risk and Resilience, Durham University,
South Road, Durham DH1 3LE, UK.
4
Centre for Ecology &
Hydrology, Crowmarsh Gifford, Wallingford, Oxfordshire,
OX10 8BB, UK.
*To whom correspondence should be addressed. E-mail:
chris.thomas@york.ac.uk
Fig. 1. Relationship between observed and expected range shifts in response to climate change, for (A)
latitude and (B) elevation. Points represent the mean responses (TSE)ofspeciesinaparticulartax-
onomic group, in a given region. Positive values indicate shifts toward the pole and to higher ele-
vations. Diagonals represent 1:1 lines, where expected and observed responses are equal. Open circles,
birds; open triangles, mammals; solid circles, arthropods; solid inverted triangles, plants; solid square,
herptiles; solid diamond, fish; solid triangle, mollusks.
19 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org
1024
REPORTS
on August 18, 2011www.sciencemag.orgDownloaded from

latitude and elevation respectively, implying much
greater responses of species to climate warming
than previously reported (18). Most of the data
we analyzed are from the temperate zone and
from tropical mountains (table S1), where eco-
systems are at least partly temperature-limited;
different rates of change might be observed in
moisture-limited ecosystems (19).
Published studies have shown nonrandom
latitudinal and elevational changes (1, 7, 13–17)
but have not previously demonstrated a statis-
tical linkage between range shifts and levels
of warming. We found that observed latitudinal
and elevational shifts (the latter more weakly)
have been significantly greater in studies with
higher levels of warming (mean latitudinal shift
versus average temperature increase; N=23spe-
cies groups × regions, Pearson correlation coef-
ficient (r) = 0.59, P= 0.003; mean elevational
shift versus temperature increase; N= 31, r=
0.37, P= 0.042). Temperature gradients differ
across the world, so a given level of warming
leads to different expected range shifts of spe-
cies in different regions (20), assuming that spe-
cies track climate changes. To estimate the
expected shifts, we calculated the distances in
latitude (kilometers) and elevation (meters) that
species in a given region would have been re-
quired to move to track temperature changes
and thus to experience the same average tem-
perature at the end of the recording period as
encountered at the start (18) (table S1). We
found that both observed latitudinal and ele-
vation range shifts were correlated with predicted
distances (Fig. 1A, N= 20 species groups ×
regions, r= 0.65, P= 0.002 for latitude; Fig.
1B, N= 30 groups × regions, r= 0.39, P=
0.035 for elevation), so our analyses directly
link terrestrial range shifts to regional and study
differences in the warming experienced.
Despite reports that many species lag behind
climate change (21–23), nearly as many studies
of observed latitudinal changes fall above as
below the observed = expected line in Fig. 1A
(9 points above, 11 below; c
2
= 0.20, 1 df, P=
0.65), suggesting that mean latitudinal shifts are
not consistently lagging behind the climate. The
lag in elevation response (Fig. 1B; 2 points above
the 1:1 line, 28 below; c
2
= 22.53, 1 df, P<
0.001) is equally surprising because the required
distances to track climate are much shorter than
for latitudinal shifts (20). Real and apparent ele-
vation lags may arise if suitable new conditions
at higher elevations occur only in locations that
cannot be reached easily (for example, on other
mountain peaks), or they may reflect the topo-
graphic and microclimatic complexity of moun-
tainous terrain [for example, cooler locations
may be on poleward-facing slopes rather than
higher (24)]; the need for finer-resolution analy-
ses (25); and additional topographic, climatic, ge-
ological, and ecological constraints [for example,
causing declines in cloud forest species (26–28)].
Taxonomic differences are not consistent pre-
dictors of recent response rates. For example,
birds seem to have responded least in terms of
elevational shifts but had a slightly greater than
expected latitudinal shift (Fig. 1). Much greater
variation is associated with differences among
species within a taxonomic group than between
taxonomic groups (Fig. 2 and table S2). For lat-
itudinal studies, on average 22% (average of
N= 23 species groups × regions) of the species
actually shifted in the opposite direction to that
expected. Similarly, 25% of species shifted down-
hill rather than to higher elevations (average of
N= 29 species groups × regions). Thus, despite
an overall significant shift toward higher lati-
tudes and elevations, which is greatest where
the climate has warmed the most, and despite
around three-quarters of species shifting pole-
ward and to higher elevations, we found that
species have exhibited a high diversity of range
shifts in recent decades.
At least three processes are likely to generate
the high diversity of range shifts among species:
time delays in species’responses, individualistic
physiological constraints, and alternative and in-
teracting drivers of change. Species may lag be-
hind climate change if they are habitat specialists
or immobile species that cannot colonize across
fragmented landscapes (17,21–23), or if they
possess other traits associated with low extinc-
tion or colonization rates (29). Species may also
show individualistic physiological responses to
different aspects of the climate, such as different
sensitivities to maximum and minimum temper-
atures at critical times of their life cycles. These
sensitivities will combine with variable wait times
for different novel climatic extremes to take
place (30). Species are also affected to dif-
ferent extents by nonclimatic factors and by
multispecies interactions, which themselves de-
pend on a diversity of environmental drivers
(21,28). For example, a species might retreat
toward the Equator at its poleward margin if it
contracts with habitat loss faster than it expands
through climate warming; whereas the poleward
range margin of a species that thrives in novel ag-
ricultural landscapes may spread at a rate exceed-
ing that expected, were warming the sole driver.
We found that rates of latitudinal and eleva-
tional shifts are substantially greater than reported
Fig. 2. Observed latitudinal shifts of the northern range boundaries of species within four exemplar
taxonomic groups, studied over 25 years in Britain. (A) Spiders (85 species), (B)groundbeetles
(59 species), (C) butterflies (29 species), and (D) grasshoppers and allies (22 species). Positive
latitudinal shifts indicate movement toward the north (pole); negative values indicate shifts toward the
south (Equator). The solid line shows zero shift, the short-dashed line indicates the median observed
shift, and the long-dashed line indicates the predicted range shift.
www.sciencemag.org SCIENCE VOL 333 19 AUGUST 2011 1025
REPORTS
on August 18, 2011www.sciencemag.orgDownloaded from

in a previous meta-analysis, and increase with
the level of warming. We conclude that average
rates of latitudinal distribution change match
those expected on the basis of average temper-
ature change, but that variation is so great within
taxonomic groups that more detailed physio-
logical, ecological and environmental data are
required to provide specific prognoses for indi-
vidual species.
References and Notes
1. J. A. Pounds, M. P. L. Fogden, J. H. Campbell, Nature
398, 611 (1999).
2. S. E. Williams, E. E. Bolitho, S. Fox, Proc. Biol. Sci. 270,
1887 (2003).
3. C. D. Thomas et al., Nature 427, 145 (2004).
4. A. Fischlin et al., in Climate Change 2007: Impacts,
Adaptation and Vulnerability. Contribution of Working
Group II to the Fourth Assessment Report of the
Intergovernmental Panel on Climate Change, M. L. Parry,
O. F. Canziani, J. P. Palutikof, P. J. van der Linden,
C. E. Hanson, Eds. (Cambridge Univ. Press, Cambridge,
2007), pp. 211–272.
5. K. E. Carpenter et al., Science 321, 560 (2008).
6. C. H. Sekercioglu, S. H. Schneider, J. P. Fay, S. R. Loarie,
Conserv. Biol. 22, 140 (2008).
7. C. J. Raxworthy et al., Glob. Change Biol. 14, 1703
(2008).
8. B. Sinervo et al., Science 328, 894 (2010).
9. A. Menzel et al., Glob. Change Biol. 12, 1969 (2006).
10. C. Rosenzweig et al., Nature 453, 353 (2008).
11. T. L. Root, D. P. MacMynowski, M. D. Mastrandrea,
S. H. Schneider, Proc. Natl. Acad. Sci. U.S.A. 102, 7465
(2005).
12. S. J. Thackeray et al., Glob. Change Biol. 16, 3304
(2010).
13. C. Parmesan et al., Nature 399, 579 (1999).
14. C. Parmesan, G. Yohe, Nature 421, 37 (2003).
15. R. Hickling, D. B. Roy, J. K. Hill, R. Fox, C. D. Thomas,
Glob. Change Biol. 12, 450 (2006).
16. C. Rosenzweig et al., in Climate Change 2007:
Impacts, Adaptation and Vulnerability. Contribution
of Working Group II to the Fourth Assessment Report
of the Intergovernmental Panel on Climate Change,
M.L.Parry,O.F.Canziani,J.P.Palut.,P.J.vanderLinden,
C. E. Hanson, Eds. (Cambridge University Press, Cambridge,
2007), pp. 79–131.
17. C. D. Thomas, Divers. Distrib. 16, 488 (2010).
18. Materials and methods are available as supporting
material on Science Online.
19. W. Foden et al., Divers. Distrib. 13, 645 (2007).
20. S. R. Loarie et al., Nature 462, 1052 (2009).
21. M. S. Warren et al., Nature 414, 65 (2001).
22. R. Menéndez et al., Proc. Biol. Sci. 273, 1465 (2006).
23. R. Nathan et al., Ecol. Lett. 14, 211 (2011).
24. A. J. Suggitt et al., Oikos 120, 1 (2011).
25. C. D. Thomas, A. M. A. Franco, J. K. Hill, Trends Ecol. Evol.
21, 415 (2006).
26. J. A. Pounds et al., Nature 439, 161 (2006).
27. I.-C. Chen et al., Glob. Ecol. Biogeogr. 20, 34 (2011).
28. G. Forero-Medina, L. Joppa, S. L. Pimm, Conserv. Biol.
25, 163 (2011).
29. A. L. Angert et al., Ecol. Lett. 14, 677 (2011).
30. D. R. Easterling et al., Science 289, 2068 (2000).
Acknowledgments: We thank A. Bergamini, R. Hickling,
R. Wilson, and B. Zuckerberg for data; H.-J. Shiu for
statistical assistance; S.-F. Shen, the Ministry of Education
in Taiwan, a UK Overseas Research Scholarship Award,
and the Natural Environment Research Council for
support; and anonymous referees for comments on the
manuscript. We are particularly grateful to the many
thousands of volunteers responsible for collecting most of
the original records of species. All data sources are listed
in the supporting online material.
Supporting Online Material
www.sciencemag.org/cgi/content/full/333/6045/1024/DC1
Materials and Methods
Tables S1 and S2
References (31–51)
1 April 2011; accepted 6 July 2011
10.1126/science.1206432
Aneuploidy Drives Genomic
Instability in Yeast
Jason M. Sheltzer,
1
Heidi M. Blank,
1
Sarah J. Pfau,
1
Yoshie Tange,
2
Benson M. George,
1
Timothy J. Humpton,
1
Ilana L. Brito,
3
Yasushi Hiraoka,
2,4
Osami Niwa,
5
Angelika Amon
1
*
Aneuploidy decreases cellular fitness, yet it is also associated with cancer, a disease of enhanced
proliferative capacity. To investigate one mechanism by which aneuploidy could contribute to
tumorigenesis, we examined the effects of aneuploidy on genomic stability. We analyzed 13 budding
yeast strains that carry extra copies of single chromosomes and found that all aneuploid strains
exhibited one or more forms of genomic instability. Most strains displayed increased chromosome loss
and mitotic recombination, as well as defective DNA damage repair. Aneuploid fission yeast strains
also exhibited defects in mitotic recombination. Aneuploidy-induced genomic instability could facilitate
the development of genetic alterations that drive malignant growth in cancer.
Whole-chromosome aneuploidy—or a
karyotype that is not a multiple of the
haploid complement—is found in great-
er than 90% of human tumors and may contrib-
ute to cancer development (1,2). It has been
suggested that aneuploidy increases genomic
instability, which could accelerate the acquisition
of growth-promoting genetic alterations (1,3).
However, whereas aneuploidy is a result of ge-
nomic instability, there is at present limited evi-
dence as to whether genomic instability can be a
consequence of aneuploidy itself. To test this
possibility directly, we assayed chromosome seg-
regation fidelity in 13 haploid strains of Saccha-
romyces cerevisiae that carry additional copies
of single yeast chromosomes (4). These aneu-
ploid strains (henceforth disomes) display im-
paired proliferation and sensitivity to conditions
that interfere with protein homeostasis (4,5).
We measured the segregation fidelity of a yeast
artificial chromosome (YAC) containing human
DNA and found that the rate of chromosome
missegregation was increased in 9 out of 13 di-
somic strains relative to a euploid control (Fig.
1A). The increase ranged from 1.7-fold to 3.3-
fold, comparable to the fold increase observed
in strains lacking the kinetochore components
Chl4 or Mcm21. Consistent with chromosome
segregation defects, 8 out of 13 disomic strains
displayed impaired proliferation on plates con-
taining the microtubule poison benomyl, includ-
ing a majority of the strains that had increased
rates of YAC loss (Fig. 1B).
Chromosome missegregation can result from
defects in chromosome attachment to the mitotic
spindle or from problems in DNA replication or
repair. Defects in any of these processes delay
mitosis by stabilizing the anaphase inhibitor
Pds1 (securin) (6). Five out of five disomes (di-
somes V, VIII, XI, XV, and XVI) exhibited de-
layed degradation of Pds1 relative to wild type
after release from a pheromone-induced G
1
arrest
(Fig. 1C and fig. S1). Defective chromosome bi-
orientation delays anaphase through the mitotic
checkpoint component Mad2 (6). Deletion of
MAD2 had no effect on Pds1 persistence in four
disomes, but eliminated this persistence in disome
V cells (fig. S1). Disome V also delayed Pds1 deg-
radation after release from a mitotic arrest in-
duced by the microtubule poison nocodazole,
which demonstrated that this strain exhibits a bi-
orientation defect. Disome XVI, which displayed
Mad2-independent stabilization of Pds1, recov-
ered from nocodazole with wild-type kinetics (fig.
S2). Thus, Pds1 persistence results predominant-
ly from Mad2-independent defects in genome
replication and/or repair (see below).
We next investigated whether aneuploidy
could affect the rate of forward mutation. Di-
somes V, VIII, X, and XIV displayed an in-
creased mutation rate at two independent loci,
whereas disome IV displayed an increased
mutation rate at CAN1 but not at URA3 (Fig.
2A). The fold increase ranged from 2.2-fold to
7.1-fold, less than the 9.5-fold and 12-fold in-
creases observed in a recombination-deficient
rad51Dmutant and a mismatch repair–deficient
msh2Dmutant, respectively. Additionally, in an as-
say for microsatellite instability, we found that di-
somes VIII and XVI displayed increased instability
in a poly(GT) tract (fig. S3), which demonstrated
that aneuploidy can enhance both simple se-
quence instability and forward mutagenesis.
To define the mechanism underlying the
increased mutation rate in aneuploid cells, we
1
David H. Koch Institute for Integrative Cancer Research and
Howard Hughes Medical Institute (HHMI), Massachusetts In-
stitute of Technology, Cambridge, MA 02139, USA.
2
Graduate
School of Frontier Biosciences, Osaka University 1-3 Yamadaoka,
Suita 565-0871, Japan.
3
Department of Ecology, Evolution and
Environmental Biology, Columbia University, New York, NY
10027, USA.
4
Kobe Advanced ICT Research Center, National Insti-
tute of Information and Communications Technology 588-2 Iwaoka,
Iwaoka-cho, Nishi-ku, Kobe 651-2492, Japan.
5
The Rockefeller
University, 1230 York Avenue, New York, NY 10065, USA.
*To whom correspondence should be addressed. E-mail:
angelika@mit.edu
19 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org1026
REPORTS
on August 18, 2011www.sciencemag.orgDownloaded from