Integrative Zoology 2013; 8: 162–174
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
doi: 10.1111/1749-4877.12027
ORIGINAL ARTICLE
Climate warming increases biodiversity of small rodents by
favoring rare or less abundant species in a grassland ecosystem
Guangshun JIANG,1,4 Jun LIU,2 Lei XU,4 Guirui YU,3 Honglin HE3 and Zhibin ZHANG4
1
College of Wildlife Resources, Northeast Forestry University, Harbin, China, 2Inner Mongolia Center for Endemic Diseases
Control and Research, Huhehot, China, 3Information Management Group for the Synthesis Center of Chinese Ecosystem Research
Network, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing, China and
4
State Key Laboratory of Integrated Management of Pest Insects and Rodents in Agriculture, Institute of Zoology, Chinese Academy
of Sciences, Beijing, China
Abstract
Our Earth is facing the challenge of accelerating climate change, which imposes a great threat to biodiversity.
Many published studies suggest that climate warming may cause a dramatic decline in biodiversity, especially
in colder and drier regions. In this study, we investigated the effects of temperature, precipitation and a normalized difference vegetation index on biodiversity indices of rodent communities in the current or previous year
for both detrended and nondetrended data in semi-arid grassland of Inner Mongolia during 1982–2006. Our results demonstrate that temperature showed predominantly positive effects on the biodiversity of small rodents;
precipitation showed both positive and negative effects; a normalized difference vegetation index showed positive effects; and cross-correlation function values between rodent abundance and temperature were negatively
correlated with rodent abundance. Our results suggest that recent climate warming increased the biodiversity of
small rodents by providing more benefits to population growth of rare or less abundant species than that of more
abundant species in Inner Mongolia grassland, which does not support the popular view that global warming
would decrease biodiversity in colder and drier regions. We hypothesized that higher temperatures might benefit
rare or less abundant species (with smaller populations and more folivorous diets) by reducing the probability of
local extinction and/or by increasing herbaceous food resources.
Key words: climate change, cross-correlation function, rare species, small mammal biodiversity, time lag
INTRODUCTION
Correspondence: Zhibin Zhang, State Key Laboratory of
Integrated Management of Pest Insects and Rodents in
Agriculture, Institute of Zoology, Chinese Academy of
Sciences, Beijing 100101, China.
Email: [email protected]
162
During the past century, our Earth has been experiencing obvious climate warming, especially in the
Northern Hemisphere (Houghton et al. 2001). This, together with increasing human disturbances, may have
caused a decline in biodiversity, which imposes a great
threat to our living planet (Hillebrand & Matthiessen
2009). It is generally believed that there would be bio-
© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
Climate warming and biodiversity
diversity loss due to global climate change, especially
in high-latitude and high-mountain biomes (McCarthy
et al. 2001; Chapin et al. 2004). Klein et al. (2004) suggest that climate warming may cause dramatic declines
in plant species diversity in high-elevation ecosystems
over short time frames, with higher species losses occurring in drier sites under anthropogenic climate change.
Levinsky et al. (2007) suggest that endemic species will
be negatively affected by future climatic changes, while
widely distributed species will be mildly affected; mammalian species richness is predicted to become dramatically reduced in the Mediterranean region but increased
towards the northeast and at higher elevations. Rull and
Vegas-Vilarrúbia (2006) estimate that roughly one-tenth
to one-third of endemic vascular plant species would
lose their habitats with the 2–4 °C temperature increase
predicted for the region by AD 2100. There could be
some unanticipated effects of global change on diversity, such as the restructuring of small mammal communities, significant loss of richness, and perhaps the rising dominance of native ‘weedy’ species (Blois et al.
2010). Based on the Holocene fossil record, Terry et al.
(2011) suggest that under projections of increased
temperature and decreased precipitation over the next
50 years, granivorous animals should thrive as communities become more dominated by individuals with
a southern geographic affinity. Although many published studies suggest that climate warming will reduce biodiversity, experimental and field evidence is
still lacking. It is necessary to look for evidence on
how biodiversity responds to climate change (Svenning & Condit 2008).
Temperature may affect biodiversity through interactions with precipitation and vegetation. It is generally
believed that higher precipitation plus warmer temperature would produce high primary productivity, supporting higher biodiversity of ecosystems (Polis & Hurd
1996). There are strong correlations between species
richness and altitude, longitude, precipitation, temperature, evapotranspiration and sunlight for several groups
of organisms (Fischer 1960; Currie & Paquin 1987;
Currie 1991; Zhang et al. 2006). These patterns can be
well modelled using spatial climate data (O’Brien 1998;
Zhang et al. 2006). The most commonly observed patterns between biodiversity and climate are based on static spatial modelling analysis and are easily influenced
by covariates between climate and geographic locations.
Studies on relationships between biodiversity and climate based on long term time series data that remove
or limit the co-varying effects of geographic factors are
rare.
© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
In the semi-arid grassland of Inner Mongolia, Pei
et al. (2009) report that the annual average temperature has increased all over the region during the past
40 years, increasing by 0.5 °C per decade from 1964–
1983 and by 0.9 °C per decade from 1984–2003. Jiang
et al. (2011) found that climate (El Niño–Southern Oscillation, precipitation and temperature) and vegetation
(normalized difference vegetation index [NDVI]) significantly affected population abundances of many rodent
species in this region from 1982–2006. In this study,
we investigated the impacts of climate and vegetation
on the biodiversity of these small rodents based on long
term data. Based on time series data for Inner Mongolia (cold and dry region), we aimed to test the following
hypothesis: (i) climate warming may decrease the biodiversity of small rodents and (ii) climate warming may
benefit more rare or less abundant species than more
abundant species in their population growth.
MATERIALS AND METHODS
Study area
The Inner Mongolia Autonomous Region is located
in northern China (37°24′–53°23′N, 97°12′–126°04′E)
(see Fig. 1) and has a cold and dry climate. The annual average temperature range is −5 to 9 °C, with an average of 3.8 °C, and the precipitation is 150–500 mm,
with an average of 326 mm. The winter ambient temperature has steadily increased by 0.5 to 0.9 °C per decade
over the past 40 years (Zhai & Ren 1997). We analyzed
the long term time series of rodent populations collected from 21 sites located in 21 counties of central Inner
Mongolia, covering an area of 257 900 km2. The study
sites and plant species of the grassland are described in
detail by Jiang et al. (2011).
Rodent abundance and species diversity
Since 1982, the abundance of rodent populations has
been surveyed twice a year by the Inner Mongolia Center for Endemic Diseases Control and Research (Table 1;
for details, see Jiang et al. 2011). Over the past 25 years,
27 small rodent species have been captured. Among the
27 species, the predominant species were Brandt’s vole
[Lasiopodomys brandtii (Radde, 1861)], the mid-day
jird [Meriones meridianus (Pallas, 1773)], the striped
dwarf hamster [Cricetulus barabensis (Pallas, 1773)],
the Mongolian five-toed jerboa [Allactaga sibirica (Forster, 1778)], the desert hamster [Phodopus roborovskii
(Satunin, 1903)] and the Mongolian jird [Meriones un-
163
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
G. Jiang et al.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
Climate data
Legend
Survey location
Provineial boundary
Inner Mongolia county boundary
0
125
250
500 km
The monthly temperature and precipitation data covering all trapping sites during 1980−2006 had a spatial
resolution of 10 × 10 km (http://www.cerndata.ac.cn/).
We calculated the yearly and seasonal (Spring, Mar–
May; Summer, Jun–Aug; Autumn, Sep–Nov; Winter,
Dec–Feb) temperature and precipitation of each location
based on the monthly data by using the spatial analysis
tool (Zonal Statistics) in ArcGIS (ESRI 1996).
The monthly data, as the amount or greenness of vegetation (Tucker et al. 1991), with the size of NDVI pixel
being 64 km2 for the period of 1980−2006, was obtained
from the Environmental and Ecological Science Data
Center for West China, National Natural Science Foundation of China (http://westdc.westgis.ac.cn). We calculated the yearly NDVI of each location based on the
monthly NDVI by using the spatial analysis tool (Zonal
Statistics) in ArcGIS (ESRI 1996).
Statistical analysis
Figure 1 Survey locations in Inner Mongolia, China.
guiculatus (Milne-Edwards, 1867)], in terms of average
abundance and/or distribution locations. Their life history characteristics are presented in Table 1.
Three biodiversity indices (i.e. the richness, the Shannon and the evenness indices) of rodent species in each
of 21 locations were calculated by referring to Spellerberg and Fedor (2003) and Krebs (1989). We calculated
the cross-correlation coefficients (CCFs) between biodiversity indices and climate or vegetation, following
Shumway and Stoffer (2006). The rodent abundances of
16 species with enough long and continuous time series
(more than 5 successive years) were calculated, following Jiang et al. (2011). The average abundance of each
species was calculated by averaging the values over the
entire study period. The CCFs between climate or vegetation and rodent abundances of the same 16 species
were also obtained from Jiang et al. (2011).
164
We used both detrended and nondetrended data for
statistical analysis; the former mainly contains annual variations, while the latter contains both annual and
decadal variations. For the detrended data, the biodiversity index, climate and vegetation data were detrended using linear regression models with generalized least
squares (Venables & Ripley 2002). We focused on data
analysis of the effects of climate or vegetation on biodiversity of small rodents in the current year and 1-year
time lag (the previous year).
The CCFs of the seasonal yearly climate data and the
yearly NDVI, and each biodiversity index in both the
current year and the previous year were calculated for
each location. We calculated the mean CCF coefficients
by using bootstrap confidence intervals (CI) (based on
10 000 replications) to examine if the CCF coefficients
overlap zero, following Efron and Tibshirani (1993) (see
also Jiang et al. 2011). We calculated the mean CCF and
95% CI of each rodent species diversity index based on
CCFs by of all locations (counties) of the relevant species.
The significant in-phase (positive) association between 2 time series was defined if both the mean CCF
and its 95% CI were larger than zero; the out-of-phase
(negative) association was defined if both the mean CCF
and its 95% CI were smaller than zero, following Lillegard et al. (2005) and Cheal et al. (2007) (see also Jiang
et al. 2011).
The CCFs between rodent abundance and climate or
vegetation of 16 species represent the impact of climate
© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
Mid-day jird
Meriones meridianus (Pallas, 1773)
Mongolian five-toed jerboa
Allactaga sibirica (Forster, 1778)
Striped dwarf hamster
Cricetulus barabensis (Pallas, 1773)
Large Japanese field mouse
Apodemus speciosus (Temminck, 1844)
Daurian pika
Ochotona dauurica (Pallas, 1776)
Desert hamster
Phodopus roborovskii (Satunin, 1903)
Northern three-toed jerboa
Dipus sagitta (Pallas, 1773)
Mongolian jird
Non-hibernation
Meriones unguiculatus (Milne-Edwards, 1867)
Dzhungarian hamster
Phodopus sungorus (Pallas, 1773)
Thick-tailed jerboa
Stylodipus telum (Lichtenstein, 1823)
Eversman's hamster
Allocricetulus eversmanni (Brandt, 1859)
Daurian suslik
Spermophilus dauricus Brandt, 1843
Gray dwarf hamster
Cricetulus migratorius (Pallas, 1773)
Narrow-skulled vole
Microtus gregalis (Pallas, 1779)
2*
3*
4*
5*
6*
© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
7*
8*
9*
10*
11*
12*
13*
14*
15
Non-hibernation
Non-hibernation
Hibernation
Non-hibernation
Hibernation
Non-hibernation
Hibernation
Non-hibernation
Non-hibernation
Non-hibernation
Non-hibernation
Hibernation
Non-hibernation
Non-hibernation
Brandt’s vole
Lasiopodomys brandtii (Radde, 1861)
1*
Life history characteristics
Species name
Species
code
Social
Solitary
Solitary
Solitary
Solitary
Solitary
Social
Solitary
Solitary
Social
Social
Solitary
Social
Solitary
Social
Diurnal
Nocturnal
Crepuscular
and diurnality
Nocturnal
Nocturnal
Nocturnal
Crepuscular
and diurnality
Nocturnal
Nocturnal
Crepuscular
and diurnality
Nocturnal
Nocturnal
Nocturnal
Nocturnal
Crepuscular
and diurnality
Folivorous
Granivorous
Folivorous
Granivorous
Omnivorous
Granivorous
Granivorous
Omnivorous
Granivorous
Folivorous
Granivorous
Ganivorous
Omnivorous
Granivorous
Folivorous
1
7
4
12
4
13
14
12
15
1
1
21
20
10
4
Number of
locations
0.148 ± 0.136
0.179 ± 0.037
0.198 ± 0.056
0.286 ± 0.038
0.300 ± 0.046
0.418 ± 0.050
0.498 ± 0.115
0.518 ± 0.043
0.531 ± 0.054
0.558 ± 0.301
0.862 ± 0.221
0.961 ± 0.087
0.973 ± 0.048
1.043 ± 0.189
1.705 ± 0.613
Average
abundance ± SE
Table 1 The life history characteristics, population abundance (individuals/100 trap nights) and number of distribution locations of 27 rodent species in Inner Mongolia during 1982–2006. Life history characteristics are mainly taken from Zhao (1981) and Luo et al. (2000). *Indicates the species for which enough long and continuous
time series on rodent abundance was obtained to be used to analyze the effects of rodent abundance on cross-correlation coefficients between rodent abundance and climate or vegetation in Figs 5 and 6
Climate warming and biodiversity
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
165
166
Species name
Five-toed pygmy jerboa
Cardiocranius paradoxus Satunin, 1903
Maximowiczi's voles
Microtus maximowiczii (Schrenk, 1859)
Lesser long-tailed hamster
Cricetulus longicaudatus (Milne-Edwards,
1867)
Norway rat
Rattus norvegicus (Berkenhout, 1769)
Steppe pika
Ochotona pusilla (Pallas, 1769)
Red-cheeked ground squirrel
Spermophilus erythrogenys Brandt, 1841
Greater long-tailed hamster
Tscherskia triton (de Winton, 1899)
Great gerbil
Rhombomys opimus (Lichtenstein, 1823)
Yellow steppe lemming
Eolagurus luteus (Eversmann, 1840)
Grey red-backed vole
Clethrionomys rufocanus (Sundevall, 1846)
House mouse
Mus musculus Linnaeus, 1758
Striped field mouse
Apodemus agrarius (Pallas, 1771) Species
code
16*
17
18
19
20
21
22
23
24
25
26*
27
Non-hibernation
Non-hibernation
Non-hibernation
Non-hibernation
Non-hibernation
Non-hibernation
Hibernation
Non-hibernation
Non-hibernation
Non-hibernation
Non-hibernation
Hibernation
Life history characteristics
Solitary
Social
Social
Solitary
Social
Solitary
Solitary
Social
Social
Solitary
Social
Solitary
Nocturnal
Nocturnal
Nocturnal
Diurnal
Diurnal
Crepuscular
and diurnality
Diurnal
Diurnal
Nocturnal
Nocturnal
Nocturnal
Nocturnal
Folivorous
Omnivorous
Folivorous
Folivorous
Folivorous
Granivorous
Folivorous
Folivorous
Omnivorous
Granivorous
Omnivorous
Granivorous
1
10
1
3
1
4
2
4
8
5
4
3
Number of
locations
0.000 ± <0.001
0.000 ± 0.014
0.005 ± 0.005
0.004 ± 0.005
0.005 ± 0.005
0.006 ± 0.006
0.007 ± 0.006
0.008 ± 0.003
0.020 ± 0.008
0.024 ± 0.015
0.024 ± 0.008
0.070 ± 0.017
Average
abundance ± SE
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
Table 1 Continued
G. Jiang et al.
© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
Climate warming and biodiversity
and vegetation on rodent abundance (Jiang et al. 2011).
The strength of these associations may be related to rodent abundance of different species, or it may be different between rare or less abundant and more abundant
species. We used the Pearson correlation analysis to test
the significant correlations between these CCFs and rodent abundance of the 16 species. All statistical analyses
were conducted by using the time series analysis (TSA)
packages in the R software (R Development Core Team
2006).
RESULTS
Effects of temperature
The current year, yearly and spring temperatures
showed positive effects on richness for nondetrended
data (Fig. 2a), and the yearly and winter temperatures
showed positive effects on the Shannon index for both
detrended and nondetrended data (Fig. 2b); spring temperature showed negative effects on the evenness index
for detrended data (Fig. 2c); and summer temperature
showed positive effects on the evenness index for nondetrended data (Fig. 2c).
In the previous year (1-year time lag), the yearly and
winter temperature showed positive effects on 3 biodiversity indices for both detrended and nondetrended
data (Fig. 2a–c); spring temperature showed positive effects on the richness and evenness indices for both detrended and nondetrended data (Fig. 2a,c); autumn temperature showed negative effects on the richness index
for detrended data (Fig. 2a) and positive effects on the
Shannon index for both detrended and nondetrended
data (Fig. 2b).
Effect of precipitation
In the current year, spring precipitation showed positive effects on 3 biodiversity indices for both detrended and nondetrended data (Fig. 3a–c); summer precipitation showed negative effects on the Shannon index for
detrended and nondetrended data (Fig. 3b); autumn precipitation showed positive effects on the richness index
for detrended and nondetrended data (Fig. 3a); and winter precipitation showed negative effects on both richness and evenness indices for both detrended and nondetrended data (Fig. 3a,c), and positive effects on the
Shannon index for detrended data (Fig. 3b).
In the previous year, autumn precipitation showed
positive effects on the richness index for nondetrended data (Fig. 3a) and summer precipitation showed neg-
© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
Figure 2 Effects of temperature on biodiversity of small rodents as measured by the means of cross-correlation coefficient
value with 95% confidence interval for detrended and nondetrended data; *P < 0.05; (a) richness index, (b) Shannon index
and (c) evenness index. In a–c, the left figures are for detrended and the right figures are for nondetrended data; the top figures are for the current year and the lower figures are for the
previous year (1-year time lag).
167
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
G. Jiang et al.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
ative effects on the evenness index for both detrended
and nondetrended data (Fig. 3c).
Effect of normalized difference vegetation index
In the current year, yearly NDVI showed positive effects on the evenness index for nondetrended data (Fig.
4c). In the previous year, yearly NDVI showed positive
effects on the Shannon index for both detrended and
nondetrended data (Fig. 4b), and showed positive effects on the evenness index for nondetrended data (Fig.
4c).
Effect of rodent abundance
In the current year, CCFs between yearly, spring or
winter temperature and rodent abundance are significantly and negatively correlated with the rodent abundance for both detrended and nondetrended data (Figs
5a and 6). The CCFs between precipitation or NDVI
and rodent abundance time series are not significantly
correlated with rodent abundance (Fig. 5b,c).
DISCUSSION
Effect of temperature
Figure 3 Effects of precipitation on biodiversity of small rodents as measured by the means of cross-correlation coefficient
value with 95% confidence interval for detrended and nondetrended data; *P < 0.05; (a) richness index, (b) Shannon index
and (c) evenness index. In a–c, the left figures are for detrended and the right figures are for nondetrended data; the top figures are for the current year and the lower figures are for the
previous year (1-year time lag).
168
Many studies project the species biodiversity loss
or range shift resulting from global warming (e.g. Peters & Lovejoy 1992; Hughes 2000; McLaughlin et al.
2002). However, our results demonstrate that temperature shows predominantly positive effects (26 of 28
significant correlations are positive) on biodiversity of
small rodents in the semi-arid grassland of Inner Mongolia grassland, not supporting the prevailing view. An
increase in temperature may benefit plant growth in
spring and extend the growing season, and thus supporting coexistence of additional species. Chmielewski and Rotzer (2001) showed that an early spring warming by 1 °C caused an advance in the beginning of plant
growing season of 7 days, the extension of growing season was mainly the result of an earlier onset of spring,
and an increase of mean annual air temperature by 1 °C
led to an extension of 5 days. Increase in temperature is
more obvious in winter in Inner Mongolia (Zhai & Ren
1997). Winter temperature increase may benefit overwintering survival and onset of spring breeding of rodents (Schmidt–Nielsen 1975).
In this study, the positive effect of temperature on
biodiversity of small rodents may be mainly caused by
the stronger positive effect of temperature on rare or
less abundant species. We have 3 hypotheses for ex-
© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
Climate warming and biodiversity
Figure 4 Effects of normalized difference vegetation index
on biodiversity of small rodents as measured by the means of
cross-correlation coefficient value with 95% confidence interval for detrended and nondetrended data; *P < 0.05; (a) richness index, (b) Shannon index and (c) evenness index. In a–
c, the left figures are for detrended and the right figures are for
nondetrended data; the top figures are for the current year and
the lower figures are for the previous year (1-year time lag).
© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
Figure 5 Pearson correlation coefficients between average
abundance of rodents (number of individuals/100 trap nights)
of 16 species and cross-correlation coefficients (between rodent
abundance and climate or vegetation) for detrended or nondetrended data in the current year and the previous year (n = 16).
*P < 0.05. (a) temperature, (b) precipitation and (c) normalized
difference vegetation index. In Fig. 5a–c, the left figures are for
detrended and the right figures are for nondetrended data; the
top figures are for the current year and the lower figures are for
the previous year (1-year time lag).
169
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
G. Jiang et al.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
Figure 6 Linear relationship between average abundance of rodents (number of individuals/100 trap nights) of 16 species and crosscorrelation coefficients (between rodent abundance and yearly, spring or winter temperature) for detrended or nondetrended data in
the current year (n =16). The data is from Fig. 5a in the current year.
plaining this observation. First, cold temperature may
increase the probability of local extinction of rare or less
abundant species. The winter temperature is very low
in Inner Mongolia, and the winter mortality of small rodents is very high. As shown in Table 1, among the first
13 more abundant species, only 5 species were found in
170
fewer than 5 locations, while in the last 14 less abundant
species, 10 species were found in fewer than 5 locations.
Thus, rare or less abundant species with fewer locations
may have more stochastic local extinction under harsh
cold climatic conditions. Second, cold temperature may
favor granivorous rodent species by inhibiting folivo-
© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
Climate warming and biodiversity
rous rodent species due to poor grass vegetation. Our
previous studies have demonstrated that high temperature tended to increase grass vegetation in this region
(Jiang et al. 2011), which may favor folivorous rodent
species. Indeed, as shown in Table 1, among the first
13 more abundant species, only 3 species were folivorous, while in the least abundant 14 species, 8 were folivorous. Third, cold temperature may increase overwintering mortality of rare or less abundant species. Large
group size may help rodents to survive the cold winter by mutual warming. Thus, cold temperature may impose more negative effects on social rodent species that
are rare or less abundant than those of the more abundant species. The other 2 life history traits (i.e. hibernating or diurnality) seem not to be helpful in explaining
the observed correlation between CCFs and population
abundance of rodents because they are similar for both
less and more abundant species (Table 1).
Effect of precipitation
In the static model studies based on spatial data, biodiversity is often positively associated with precipitation
(e.g. Fischer 1960; Currie & Paquin 1987; Currie 1991;
O’Brien 1998; Zhang et al. 2006). However, in our
studies, we found that precipitation showed both positive (9 of 17 significant correlations are positive) and
negative (8 of 17 significant correlations are negative)
associations with biodiversity indices. However, we
found that summer or winter precipitation showed significant negative effects, while spring and autumn precipitation showed positive effects. This suggests that the
amount of green vegetation available to and eaten by rodents influences their reproduction (Reichman & van de
Graaff 1975). In addition, reproductive response of rodents has been frequently associated with increased water intake, increased general fitness or the presence of
some estrogenic substance in the greenery (Reichman
& van de Graaff 1975). In our study region, reproduction of small rodents is positively associated with spring
precipitation, which is essential for grass growth (Xie et
al. 2012). Abundant spring precipitation may increase
populations of and have more benefits for rare folivorous species. If summer precipitation is very abundant,
high precipitation may impose negative effects on small
rodents in summer by flooding their burrows and nests
(see Brown & Ernest 2002; Zhang et al. 2003; Jiang et
al. 2011), and thereby increase the extinction probability of rare species with smaller population size or fewer locations. This may help to explain why summer precipitation reveals a negative effect on the biodiversity
of small rodents. High winter precipitation may have
© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
a similar effect of summer precipitation when in early
spring snow melts and then flood burrows of small rodents. It is not clear why autumn precipitation showed
a more positive effect on rare or less abundant species.
High autumn precipitation may be beneficial to folivorous
rodent species by extending the growing seasons of grass.
Effect of vegetation
It is well known that vegetative stratification providing food and cover from predators is an important factor of habitats to support small mammal communities
(Yahner 1983; Fitzgibbon 1997). Some previous studies indicate that vegetation biomass was not related to
small mammal biodiversity (e.g. Sullivan et al. 2006).
In this study, 4 significant positive effects of NDVI (i.e.
biomass of vegetation indicator) on biodiversity indices
were detected, which support our hypothesis that good
vegetation favors species that are rare or less abundant
with more folivorous diets. Previous studies have indicated that in more disturbed grassland, weeds are abundant and favor granivorous rodent species, whereas in
non-disturbed or less-disturbed grassland, herbaceous
species are dominant and, thus, favor folivorous species
(Fan et al. 1999; Zhong et al. 1999; Zhang et al. 2003).
Effect of detrending
Detrended and nondetrended data represent mainly annual and decadal variations of biodiversity of small
rodents in this study. In some cases, both detrended and
nondetrended data showed similar significant associations, but in many cases significant associations were
observed either in detrended data or in nondetrended data. For example, yearly and spring temperature in
the current year showed positive significantly effects
on the richness for detrended data but not for detrended data (Fig. 2a). Spring precipitation in the current
year showed significant effects on richness for data. The
NDVI in both current and previous years showed significant positive effects on evenness for nondetrended data,
but not for detrended data (Fig. 4c). These results suggest that temperature, precipitation or NDVI can have
significant effects on the biodiversity of small rodents
on both annual and decadal scales.
Effect of time lag
Our previous studies indicate that temperature, precipitation and NDVI had time lag effects on rodent
abundances in the study region (Jiang et al. 2011). Thus,
time lags may have an impact on the biodiversity of
small rodents. In this study, in many cases, we found
171
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
G. Jiang et al.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
that significant effects of temperature, precipitation or
NDVI occurred either in the current year or in the previous year. For example, we found that yearly, spring and
winter temperature in the previous year showed significant positive effects, whereas autumn temperature in
the previous year showed significant negative effects on
richness for detrended data, but not in the current year
(Fig. 2a). Spring and autumn precipitation in the current
year showed significant positive effects on richness and
winter precipitation showed significant negative effects
on richness for detrended data, but not in the previous
year. NDVI in the previous year showed significant positive effects on the Shannon index for both detrended
and nondetrended data, but not in the current year (Fig.
4b). These diverse observations can be explained by different effects of temperature, precipitation or NDVI on
rodent species. Jiang et al. (2011) report that both temperature and precipitation in current year show positive
effects on rodent population; the NDVI shows a negative effect on rodent abundance in the current year, but a
positive effect in the previous year.
Effect of diversity indices
Species richness, diversity and evenness are all measures of relative biodiversity (complexity) in communities. They are not, however, interchangeable, and it is
helpful to distinguish among these terms (Spellerberg &
Fedor 2003). Species richness is a measure of variety of
species, and is used to refer to the number of species in a
given area or in a given sample. The Shannon–Wiener diversity index is based on percentage composition by species. With increasing sample size, the index approaches
a constant value (Peet 1975) and gives more weight to
rare species (Krebs 1989). Evenness is a measure of the
relative dominance of different species within a community (Molinari 1989), and may be a more sensitive indicator of community change than richness or diversity
alone (Wittebolle et al. 2009). Wilsey et al. (2005) argue that local evenness may be more sensitive to global
change than local richness.
With very few exceptions, our results indicated that
all 3 diversity indices showed similar responses to environmental variables. In this study, we found that the autumn temperature in the previous year showed significant negative effects on the richness index for detrended
data but positive effects on the Shannon index (Fig. 2a
and b). Winter precipitation in the current year showed
significant negative effects on both richness and evenness for detrended data, but positive effects on the Shannon index (Fig. 2a–c). Our results also indicated that
172
evenness was more consistent with richness and Shannon diversity indices, while the latter 2 indices showed
slight differences. However, our hypotheses now require
further testing on a wider geographic scale. Specific hypotheses for future tests are necessary, and these could be
at the individual species level or the community level.
ACKNOWLEDGMENTS
We thank the IPN (SIP 20080127, SIP 20090442),
COECYT-SINALOA (CECyT-SIN 2009) and FOMIX
CONACYT-SINALOA (FOMIX-SIN-2008-C01-99712)
for financial support.
REFERENCES
Blois JL, McGuire JL, Hadly EA (2010). Small mammal diversity loss in response to Late Pleistocene climatic change. Nature 465, 771–4.
Brown JH, Ernest SK (2002). Rain and rodents: complex dynamics of desert consumers. Bioscience 52,
979–87.
Chapin FS, Callaghan TV, Bergeron Y et al. (2004).
Global change and the boreal forest: thresholds, shifting states or gradual change? AMBIO 33, 361–5.
Cheal AJ, Delean S, Sweatman H, Thompson AA (2007).
Spatial synchrony in coral reef fish populations and
the influence of climate. Ecology 88, 158–69.
Chmielewski FM, Rotzer T (2001). Response of tree
phenology to climate change across Europe. Agricultural and Forest Meterology 108, 101–12.
Currie DJ (1991). Energy and large-scale patterns of animal and plant–species richness. The American Naturalist 137, 27–49.
Currie DJ, Paquin V (1987). Large-scale biogeographic patterns of species richness of trees. Nature 329,
326–7.
Efron B, Tibshirani RJ (1993). An Introduction to the
Bootstrap. Chapman and Hall, London, UK.
ESRI, Environmental System Research Institute (1996).
Using ArcView GIS. ESRI, Redlands.
Fan N, Zhou W, Wei W, Wang Q, Jiang Y (1999). Rodent pest management in the Qinghai–Tibet Alpine
Meadow Ecosystem. In: Singleton CR, Hinds LA,
Leirs H, Zhang Z, eds. Ecologically-based Management of Rodents. Australian Centre for International
Agricultural Research, Canberra, pp. 285–304.
© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
Climate warming and biodiversity
Fischer AG (1960). Latitudinal variations in organic diversity. Evolution 14, 64–81.
Fitzgibbon CD (1997). Small mammals in farm woodlands: the effects of habitat, isolation and surrounding land use patterns. Journal of Applied Ecology 34,
530–9.
Hillebrand H, Matthiessen B (2009). Biodiversity in a
complex world: consolidation and progress in functional biodiversity research. Ecology Letters 12,
1405–19.
Houghton JT, Ding Y, Griggs DJ et al. (2001). Climate
Change 2001: The Scientific Basis. Cambridge University Press, Cambridge.
Hughes L (2000). Biological consequences of global warming: is the signal already apparent? Trends in
Ecology & Evolution 15, 56–61.
Jiang G, Zhao T, Liu J et al. (2011). Effects of ENSOlinked climate and vegetation on population dynamics of sympatric rodent species in semi-arid grasslands of Inner Mongolia, China. Canadian Journal of
Zoology 89, 678–91.
Klein JA, Harte J, Zhao X (2004). Experimental warming causes large and rapid species loss, dampened by
simulated grazing, on the Tibetan Plateau. Ecology
Letters 7, 1170–9.
Krebs CJ (1989). Ecological Methodology. Harper and
Row, New York.
Levinsky I, Skov F, Svenning Jens-Christian, Rahbek
C (2007). Potential impacts of climate change on
the distributions and diversity patterns of European
mammals. Biodiversity and Conservation 16, 3803–
16.
Lillegard M, Engen S, Saether BE (2005). Bootstrap
method for estimating spatial synchrony in fluctuating populations. Oikos 109, 342–50.
Luo Z, Chen W, Gao W (2000). Fauna Sinica, Mammalia Vol. 6 Rodentia. Part III: Cricetidae. Science
Press, Beijing, pp. 26–329 (In Chinese).
McCarthy JJ, Canziani OF, Leary NA, Dokken DJ,
White KS (2001). Climate Change 2001: Impacts,
Adaptation, and Vulnerability. Cambridge University
Press, Cambridge.
McLaughlin JF, Hellmann JJ, Boggs CL, Ehrlich PR
(2002). Climate
�����������������������������������������
change hastens population extinctions. PNAS 99, 6070–4.
Molinari J (1989). A calibrated index for the measurement of evenness. Oikos 56, 319–26.
© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
O’Brien EM (1998). Water-energy dynamics, climate,
and prediction of woody plant species richness: an
interim general model. Journal of Biogeography 25,
379–98.
Peet PK (1975). Relative diversity indices. Ecology 56,
496–8.
Pei H, Cannon A, Whitfield P, Hao L (2009). Pentad average temperature changes of Inner Mongolia during
recent 40 years. Journal of Applied Meteorological
Science 20, 443–50 (In Chinese).
Peters RL, Lovejoy TE (1992). Global Warming and Biological Diversity. Yale University Press, New Haven, CT.
Polis GA, Hurd SD (1996). Linking marine and terrestrial food webs: allochthonous input from the ocean
supports high secondary productivity on small islands
and coastal land communities. The American Naturalist 147, 396–423.
R Development Core Team (2006). R: a language and
environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. [Cited
10 Jan 2008.] Available from URL: http://www.Rproject.org
Reichman OJ, van de Graaff KM (1975). Association
between ingestion of green vegetation and desert rodent reproduction. Journal of Mammalogy 56, 503–6.
Rull V, Vegas-Vilarrúbia T (2006). Unexpected biodiversity loss under global warming in the neotropical
Guayana Highlands: a preliminary appraisal. Global
Change Biology 12, 1–9.
Schmidt-Nielson K (1975). Animal Physiology: Adaptation and Environment. Cambridge University Press,
London.
Shumway RH, Stoffer DS (2006). Time Series Analysis and Its Applications With R Examples, 2nd edn.
Springer Science Business Media, New York, USA.
Spellerberg Ian F, Fedor PJ (2003). A tribute to Claude
Shannon (1916–2001) and a plea for more rigorous use of species richness, species diversity and the
‘Shannon–Wiener’ Index. Global Ecology and Biogeography 12, 177–9.
Sullivan TP, Sullivan DS (2006). Plant and small mammal diversity in orchard versus non-crop habitats. Agriculture, Ecosystems and Environment 116, 235–43.
Svenning JC, Condit R (2008). Biodiversity in a warmer
world. Science 322, 206–7.
Terry RC, Li C, Hadly EA (2011). Predicting smallmammal responses to climatic warming: autecolo-
173
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
G. Jiang et al.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
gy, geographic range and the Holocene fossil record.
Global Change Biology 17, 3019–34.
Tucker CJ, Dregne HE, Newcomb WW (1991). Expansion and contraction of the Sahara Desert between
1980 and 1990. Science 253, 299–301.
Venables WN, Ripley BD (2002). Modern Applied Statistics with S, 4th edn. Springer, New York, NY, USA.
Wilsey BJ, Chalcraft DR, Bowles CM, Willig MR
(2005). Relationships among indices suggest that
richness is an incomplete surrogate for grassland biodiversity. Ecology 86, 1178–84.
Wittebolle L, Marzorati M, Clement L et al. (2009). Initial community evenness favours functionality under
selective stress. Nature 458, 623–6.
Xie X, Wen Y, Niu H, Shi D, Zhang Z (2012). Re-feeding
evokes reproductive overcompensation of food-restricted Brandt’s voles. Physiology & Behavior 105,
653–60.
Yahner RH (1983). Small mammals in farmstead shelterbelts: habitat correlates of seasonal abundance and
174
community structure. Journal of Wildlife Management 47, 74–84.
Zhai P, Ren F (1997). On changes of China’s maximum
and minimum temperatures in the recent 40 years.
Acta Meterorologica Sinica 55, 418–29.
Zhang Z, Pech R, Davis S, Shi D, Wan X, Zhong W
(2003). Extrinsic and intrinsic factors determine the
eruptive dynamics of Brandt’s voles Microtus brandti
in Inner Mongolia, China. Oikos 100, 299–310.
Zhang Z, Xie Y, Wu Y (2006). Human
�����������������������
disturbance, climate and biodiversity determine biological invasion
at a regional scale. Integrative Zoology 1, 130–8.
Zhao K (1981). Rodents in Inner Mongolia. Inner Mongolia People’s Publishing House, Huhot (In Chinese).
Zhong W, Wang M, Wan X (1999). Ecological management of Brandt’s vole (Microtus brandti) in Inner
Mongolia, China. In: Singleton CR, Hinds LA, Leirs
H, Zhang Z, eds. Ecologically-based Management of
Rodents. Australian Centre for International Agricultural Research, Canberra, pp. 199–214.
© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS