Climate change effects on soil microarthropod abundance and community structure

Abstract

Long-term ecosystem responses to climate change strongly depend on how the soil subsystem and its inhabitants respond to these perturbations. Using open-top chambers, we studied the response of soil microarthropods to single and combined effects of ambient and elevated atmospheric [CO2], ambient and elevated temperatures and changes in precipitation in constructed old-fields in Tennessee, USA. Microarthropods were assessed five years after treatments were initiated and samples were collected in both November and June. Across treatments, mites and collembola were the most dominant microarthropod groups collected.

Full text

This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright

Author's personal copy
Applied Soil Ecology 47 (2011) 37–44
Contents lists available at ScienceDirect
Applied Soil Ecology
journal homepage: www.elsevier.com/locate/apsoil
Climate change effects on soil microarthropod abundance and community
structure
Paul Kardola,b,c,∗, W. Nicholas Reynoldsa, Richard J. Norbyb, Aimée T. Classena
aThe University of Tennessee, Department of Ecology and Evolutionary Biology, 576 Dabney Hall, Knoxville, TN 37996, United States
bEnvironmental Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831-6422, United States
cDepartment of Forest Ecology and Management, Swedish University of Agricultural Sciences, 901 83 Umeå, Sweden
article info
Article history:
Received 7 July 2010
Received in revised form 20 October 2010
Accepted 1 November 2010
Keywords:
Collembola
Elevated atmospheric CO2
Mites
Old fields
Precipitation
Warming
abstract
Long-term ecosystem responses to climate change strongly depend on how the soil subsystem and its
inhabitants respond to these perturbations. Using open-top chambers, we studied the response of soil
microarthropods to single and combined effects of ambient and elevated atmospheric [CO2], ambient
and elevated temperatures and changes in precipitation in constructed old-fields in Tennessee, USA.
Microarthropods were assessed five years after treatments were initiated and samples were collected in
both November and June. Across treatments, mites and collembola were the most dominant microarthro-
pod groups collected.
We did not detect any treatment effects on microarthropod abundance. In November, but not in June,
microarthropod richness, however, was affected by the climate change treatments. In November, total
microarthropod richness was lower in dry than in wet treatments, and in ambient temperature treat-
ments, richness was higher under elevated [CO2] than under ambient [CO2]. Differential responses of
individual taxa to the climate change treatments resulted in shifts in community composition. In general,
the precipitation and warming treatments explained most of the variation in community composition.
Across treatments, we found that collembola abundance and richness were positively related to soil
moisture content, and that negative relationships between collembola abundance and richness and soil
temperature could be explained by temperature-related shifts in soil moisture content.
Our data demonstrate how simultaneously acting climate change factors can affect the structure of soil
microarthropod communities in old-field ecosystems. Overall, changes in soil moisture content, either
as direct effect of changes in precipitation or as indirect effect of warming or elevated [CO2], had a larger
impact on microarthropod communities than did the direct effects of the warming and elevated [CO2]
treatments. Moisture-induced shifts in soil microarthropod abundance and community composition may
have important impacts on ecosystem functions, such as decomposition, under future climatic change.
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
Atmospheric CO2concentrations ([CO2]) are rising, resulting
in increases in global atmospheric temperatures (2.0–4.5 ◦Cby
the year 2100), and modification of precipitation patterns (IPCC,
2007). Long-term ecosystem responses to atmospheric and climatic
changes (hereafter ‘climate changes’) may largely depend on how
the soil subsystem responds to these perturbations (e.g., Davidson
and Janssens, 2006; Bardgett et al., 2008). While recent studies
have focused on how climate changes can impact soil microbial
communities and the ecosystem processes that they control, such
as litter decomposition and nutrient cycling (e.g., Bardgett et al.,
∗Corresponding author. Present address: Department of Forest Ecology and Man-
agement, Swedish University of Agricultural Sciences, 901 83 Umeå, Sweden.
Tel.: +46 90 786 8398; fax: +46 90 786 8166.
E-mail addresses: Paul.Kardol@seksko.slu.se,p.kardol@gmail.com (P. Kardol).
2008; Castro et al., 2010; Kardol et al., 2010b), effects of climate
changes on soil microarthropods received less attention (Hågvar
and Klanderud, 2009). Soil microarthropods play an important role
in the functioning of the decomposer food web by, for exam-
ple, exerting top-down control of primary (bacteria, fungi) and
secondary (nematodes, protozoa) decomposers (e.g., Petersen and
Luxton, 1982; Seastedt, 1984; Beare et al., 1992; Cole et al., 2004;
Filser, 2002; Sackett et al., 2010). Soil microarthropods also affect
decomposition processes directly through fragmentation of litter
and through fecal production (e.g., Seastedt, 1984; Sackett et al.,
2010). Hence, a better understanding of effects of climate changes
on the abundance and community structure of soil microarthro-
pods can aid predictions of how soil ecosystems may function under
future climatic conditions.
Climate changes can influence soil microarthropod community
abundance and composition directly by altering soil microclimate
and indirectly by altering resource availability and the composi-
tion of the soil food web. Warming and changes in precipitation
0929-1393/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.apsoil.2010.11.001

Author's personal copy
38 P. Kardol et al. / Applied Soil Ecology 47 (2011) 37–44
amounts, for example, can directly alter soil temperature and
moisture, factors that strongly influence microarthropod repro-
duction and development rates (e.g., van Straalen, 1994; Uvarov,
2003). In fact, soil microarthropods are extremely responsive
to changes in soil moisture, a pattern seen in numerous stud-
ies across diverse ecosystems (e.g., Frampton et al., 2000; Pflug
and Wolters, 2001; Lindberg et al., 2002; Tsiafouli et al., 2005;
Moron-Rios et al., 2010). Unlike soil moisture, warming impacts
on microarthropods have been context dependent and abundance
responses varied across experiments (e.g., Coulson et al., 1996;
Huhta and Hänninen, 2001; Hågvar and Klanderud, 2009). Work by
Sjursen et al. (2005) suggested that warming may indirectly alter
soil microarthropod communities by causing a shift in the abun-
dance and composition of soil organisms upon which they prey.
In addition, temperature and other climate factors may indirectly
influence soil microarthropod communities through changes in
plant physiology or community structure which can alter resource
availability and microhabitat conditions (e.g., Cotrufo and Ineson,
1995; Kardol et al., 2010b).
The limited number of studies on response of soil microarthro-
pods to climate change have often focused on single-factor effects
of elevated [CO2] (e.g., Jones et al., 1998; Lussenhop et al., 1998;
Hansen et al., 2001), warming (e.g., Coulson et al., 1996; Bokhorst
et al., 2008), or changes in precipitation patterns (e.g., O’Lear and
Blair, 1999; Tsiafouli et al., 2005). Climate changes, however, will
not happen in isolation of one another. For example, elevated
[CO2] may ameliorate negative effects of soil drying through reduc-
ing plant stomatal conductance and transpiration, while increased
evapotranspiration resulting from higher temperatures may exac-
erbate effects of soil drying (e.g., Dermody et al., 2007). On the
other hand, drying may result in more extreme soil temperatures
(e.g., Tsiafouli et al., 2005). Interactions among climate change vari-
ables, thus may alter soil microarthropod communities in ways that
are not always predictable from the impact of individual climate
change factors (Couteaux and Bolger, 2000; Loranger et al., 2004).
For example, Harte et al. (1996) found that warming increased
microarthropod abundance and biomass under wet conditions, but
not under dry conditions.
We took advantage of a long-term, multi-factor climate
change experiment in constructed old-field ecosystems (estab-
lished in 2002) to investigate the single and interactive effects
of atmospheric [CO2], air temperature, and precipitation on soil
microarthropod abundance and community structure. We mea-
sured treatment responses of soil microarthropods at two different
sampling points, one at the end of the growing season and one at
the end of the experiment (November 2007 and June 2008, respec-
tively). Each of the climate change factors has the potential to affect
soil microarthropods; however, because microarthropods are sen-
sitive to soil drying, we predicted that the dry treatment would
have the strongest effect on microarthropod abundance and com-
munity structure. Because [CO2] and temperature can both alter
soil moisture (directly or indirectly via changes in plant physiology
and plant community structure), we also predicted there would
be interactive effects of precipitation, [CO2], and temperature on
microarthropod communities.
2. Materials and methods
2.1. Experimental design
The Old field Community, Climatic and Atmospheric Manipula-
tion (OCCAM) experiment was established at Oak Ridge National
Environmental Research Park (35◦54N; 84◦21W) in Oak Ridge,
Tennessee, USA. The site was used for agriculture until 1943 and
then was left fallow until 1964 when a managed fescue field was
established. Soils are classified as Captina silt loam – fine-silty,
siliceous, mesic typic fragiudult, well drained, and slightly acidic
(Edwards and Norby, 1999). Whole-soil N (1.62gNkg
−1) and C
(18.3gCkg
−1), determined prior to the start of the experiment,
were not affected by the climate change treatments and did not
change over time (Garten et al., 2009). Experimental plots were
established in 2002 and were planted with seven plant species
common to old fields in the southeastern United States—Plantago
lanceolata L., a herbaceous forb; Andropogon virginicus L., a cespitose
C4bunchgrass; Festuca pratense L. syn F. elatior L.,aC
3bunch-
grass; Dactylis glomerata L.,aC
3bunchgrass; Trifolium pratense
L., a herbaceous legume; Solidago canadensis, a herbaceous forb;
and Lespedeza cuneata,aN
2-fixing sub-shrub. The response of the
plant community over time is described in Engel et al. (2009) and
in Kardol et al. (2010a). The experimental design is described in
detail elsewhere (Wan et al., 2007; Dermody et al., 2007; Garten
et al., 2008). Briefly, in 2003, [CO2], temperature, and precipi-
tation treatments were applied using open-top chambers (4 m
diameter) arranged in a randomized complete block, split-plot
design (n= 3). Temperature and [CO2] concentrations were regu-
lated continuously starting in April 2003 with heating and cooling
units and CO2additions as described in Norby et al. (1997). Rain-
out shelters were constructed to eliminate natural precipitation.
Whole chambers were treated with ambient or elevated [CO2]
(ambient + 300 ppm), and ambient or elevated temperatures (ambi-
ent+3◦C). Each chamber was split into wet and dry sub-plots,
established with differential irrigation. Wet and dry plots were
irrigated with weekly additions of 2 mm (dry) and 25 mm (wet)
rainwater, which was collected and stored on site.
Soil volumetric water content (VWC) integrated over 0–15 cm
depth was measured weekly within each sub-plot by time
domain reflectometry (TDR100, Campbell Scientific) as described
by Dermody et al. (2007). Averaged from October 2005 (two years
prior to the date we collected our first set of soil samples; see para-
graph 2.2) to June 2008 (the time we collected our second set of
soil samples), within each plot, the dry sub-plots had significantly
lower (3.9%) soil moisture than the wet sub-plots (Fig. 1A). Further,
elevated temperature generally reduced soil moisture, while ele-
vated [CO2] generally increased soil moisture (see also Dermody et
al., 2007). Changes in soil moisture in response to [CO2] and tem-
perature were small relative to the changes in soil moisture due
to the precipitation treatment. Soil temperature (0–10 cm) within
each sub-plot was recorded every hour using a data logger sys-
tem (CR10X, Campbell Scientific). From October 2005 to June 2008,
soil temperature was, on average, 1.4 ◦C higher in chambers with
elevated temperature than in chambers with ambient tempera-
ture (ANOVA: F1,6 = 59.62, P<0.001; Fig. 1B); no other treatments
affected soil temperature.
2.2. Soil sampling and microarthropod extractions
In November 2007 and June 2008 (hereafter referred to as
November and June), three soil cores (0–15 cm depth, 2 cm diam-
eter) were collected from each sub-plot. Microarthropods were
extracted from the soil samples using modified high-gradient
Tullgren funnels (Crossley and Blair, 1991) for 72 h at room temper-
ature. Individuals were collected into vials containing 70% ethanol
for preservation and storage. Except for Prostigmata, mites and
collembola were identified to species or morphospecies accord-
ing to Christiansen and Bellinger (1980–1981),Balogh and Balogh
(1992), and Niedbala (2002). Other microarthropods were identi-
fied to higher-level taxonomic groups. Juveniles were not included
in the analyses. The three samples from each sub-plot were aver-
aged to one; in our analyses, we only used one data point for each
sub-plot.

Author's personal copy
P. Kardol et al. / Applied Soil Ecology 47 (2011) 37–44 39
Fig. 1. Mean soil moisture expressed as volumetric water content (VWC) (0–15 cm) (A), and mean soil temperature (0–10 cm) (B) from October 2005 to June 2008 for
precipitation, [CO2], and temperature treatments. Earlier data on soil moisture in this experiment are published in Dermody et al. (2007). Each point represents the bi-
monthly (two week) average for a treatment plot. Dry treatments are shown in the left panels; wet treatments are shown in the right panels. AT = ambient temperature;
ET = elevated temperature; AC = ambient [CO2]; EC = elevated [CO2]; D = dry; W = wet.
2.3. Statistical analyses
Across treatments, microarthropod abundance and richness
was significantly higher in June than in November (Fig. 2;
Supplementary material, Table S1). We therefore decided to run
further statistical analyses separately for the two sampling times.
The direct and interactive effects of precipitation, [CO2], and tem-
perature on abundance and taxon richness (total microarthropods,
and separately for mites and collembola) were tested using a three-
way, split-plot analysis of variance (ANOVA) (PROC MIXED, SAS
Institute, Cary, NC, USA). Precipitation, [CO2], temperature, and
their interactions were considered fixed effects, and blocks and the
interactions between block, [CO2], and temperature were included
as random factors. The Kenward–Rogers method was used to esti-
mate the denominator degrees of freedom for tests of individual
fixed effects, as appropriate for data sets with small sample sizes.
For abundance, data were log-transformed to meet assumptions of
normality and homogeneity of variance.
Treatment effects on microarthropod community composition
were analyzed using CANOCO, version 4.5 (ter Braak and ˇ
Smilauer,
2002). For the November data, the largest gradient length of
detrended correspondence analysis was 3.8 standard deviation
units, while for the June data, the largest gradient length was 2.9.
Taken together, redundancy analysis (RDA) (i.e., a linear ordina-
tion method) was therefore considered most appropriate (Lepˇ
s and
ˇ
Smilauer, 2003). RDA analyses were carried out using abundance
data, and we included the climate treatments and their interactions
as explanatory variables. In all RDA analyses, block was included
as covariable. Significance of effects was tested using Monte Carlo
permutation tests (999 permutations). To exclude variation among
blocks from the statistical tests, samples were permuted within
blocks. Marginal effects (i.e., the independent effect of each vari-
able) were tested by manual selection of each individual variable.
The relationship between soil moisture/temperature and
microarthropod abundance and richness (total, and separately
for mites and collembola) was determined by linear regression
analysis. Given the generally slow response of microarthropod pop-
ulations to changes in environmental conditions (life cycles may
vary from a few weeks to over two years; Walter and Proctor, 1999)
and given the potential variability among groups of microarthro-
pods in the time needed to adjust to altered soil moisture and
temperature levels, we tested relationship between soil mois-
ture/temperature and microarthropod abundance and richness
using: (1) average soil moisture and temperature data at the day
closest to sampling, (2) average soil moisture and temperature one
month prior to the sampling day, (3) average soil moisture and
temperature one year prior to the sampling day, and (4) average
soil moisture and temperature two years prior to the sampling day.
Average soil moisture and temperature data provide an integrated
estimate of how soil microclimate might influence the abundance
and diversity of microarthropods. Results from regression analyses
did not differ qualitatively among the estimates of soil moisture and
temperature; therefore we arbitrarily chose to only present result
of analyses with average soil moisture and temperature one month
prior to sampling. Soil moisture content was inversely related to
soil temperature (p< 0.01). Therefore, we used partial least square
regressions to test if significant relationships between soil temper-
ature and microarthropod abundance and richness hold when soil
moisture was corrected for. Regression analyses were run in JMP
8.0.2.2 (SAS Institute, Cary, NC, USA).

Author's personal copy
40 P. Kardol et al. / Applied Soil Ecology 47 (2011) 37–44
0
100
200
300
400
Mites
Collembola
0
100
200
300
400
TaTe
[CO 2]a
TaTe
[CO 2]e
Dry
TaTe
[CO 2]a
TaTe
[CO2]e
Wet
Other
TaTe
[CO2]a
TaTe
[CO2]e
Dry
TaTe
[CO2]a
TaTe
[CO2]e
Wet
JuneBA November
Abundance(102 ind. m-2)
Abundance(102 ind. m-2)
0
2
4
6
8
10
0
2
4
6
8
10
TaTe
[CO 2]a
TaTe
[CO 2]e
Dry
TaTe
[CO 2]a
TaTe
[CO 2]e
Wet
TaTe
[CO 2]a
TaTe
[CO 2]e
Dry
TaTe
[CO 2]a
TaTe
[CO 2]e
Wet
Taxonomic richness
Taxonomic richness
JuneDC November
Fig. 2. Microarthropod abundance (A and B) and taxonomic richness (C and D) under [CO2], temperature (T), and precipitation treatments in November 2007 and June 2008.
Data are mean ±s.e. Abbreviations: Ta=ambient temperature; Te= elevated temperature; [CO2]a= ambient [CO2]; [CO2]e= elevated [CO2].
3. Results
3.1. Microarthropod abundance and richness
Across samples, the microarthropods collected represented 34
unique taxa (Supplementary material, Table S2). Low taxonomic
richness could be due to disturbances such as plot construction and
plant manipulation at the beginning of the experiment. Mites were
the most common group found across all the samples, followed by
collembola. Other groups (Thysanoptera, Diplura, Symphyla, and
Protura) were rare and not observed in all samples; therefore, they
were not individually analyzed. Total microarthropod abundance
and richness were generally higher in June (close to the peak of
the growing season) than in November (at the end of the growing
season).
We were unable to detect a difference in total microarthro-
pod abundance, abundance of mites, and abundance of collembola
across our treatments on any date (Fig. 2;Table 1). Collembola
abundance was generally lower in dry than in wet treatments, but
within-treatment variation was high. In contrast to the abundance
data, our treatments significantly altered total microarthropod
richness (Fig. 2;Table 1). In November, total microarthropod
richness was lower in dry treatments than in wet treatments.
Additionally, there was a significant interaction between the [CO2]
and the temperature treatments: microarthropod richness was
higher in treatments with elevated [CO2] than in treatments with
ambient [CO2], but only under ambient temperature (Fig. 2).
In June, microarthropod richness was not affected by the cli-
mate change treatments. In both seasons, richness patterns of
mites and collembola generally mirrored the patterns of the total
microarthropod community; however, no significant treatment
effects were detected for either of these groups (Table 1).
3.2. Microarthropod community composition
The climate change treatments altered the compositional struc-
ture of the microarthropod community (Fig. 3;Supplementary
material, Table S3). The full RDA model, including precipitation,

Author's personal copy
P. Kardol et al. / Applied Soil Ecology 47 (2011) 37–44 41
0.10.1-
-1.0 1.0
Oribatida mspp.1
Oribatida msp.2
Epilohmanniasp.
Nothrus sp.
Prostigmata
Galumna sp.
Euphthiracarus humeralis
Folsomia candida
Isotoma sp.
Onychurius sp.
Japigidae
Wet
Dry
Elevated T
Ambient T
Ambient CO2
Elevated CO2
0.10.1-
0.10.1-
Oribatida msp.1
Epilohmannia sp.
Nothrus sp.
Prostigmata
Uropodidae msp.1
Oribatida msp.3
Euphthiracarus humeralis
Isotoma sp.
Thysanoptera
Japigidae
Wet
Dry
Elevated T
Ambient T
Ambient CO2
Elevated CO2
Sphagnoppia biflagelata
1st RDA axis, eigenvalue = 0.15
2nd RDA axis, eigenvalue = 0.07
2nd RDA axis, eigenvalue = 0.06
1st RDA axis, eigenvalue = 0.15
November June
Fig. 3. ‘Species’-treatment plots resulting from redundancy analysis showing composition of the microarthropod community in November (left) in June (right). Treatments
include precipitation, [CO2], temperature, and their interactions. For clarity, treatment interactions are not shown. Eigenvalues along the axes indicate the amount of explained
variability in community composition. Only the best fitting taxa are shown (species fit >9% for November, and >7% for June).
[CO2], and temperature as explanatory factors, explained 32%
and 33% of the variation in community composition in Novem-
ber and June, respectively. In November, a significant amount
of the variation in the mite community composition (8.2%) was
explained by the temperature treatment, as shown on the first
RDA axis (Fig. 3). The remaining treatments did not explain a
significant amount of the variation. However, most taxa were pos-
itively associated with wet treatments, and most strongly under
ambient [CO2] (as indicated by the marginally significant inter-
action between water and [CO2]; Supplementary material, Table
S3). In June, variation in microarthropod community composition
was significantly explained by precipitation (12.8%), tempera-
ture (10.7%), and the interaction between [CO2] and temperature
(15.5%) (Supplementary material, Table S3). Most notably, the
oribatid mite Epilohmannia sp. was strongly associated with dry
treatments, and with elevated temperature.
3.3. Relationships between soil microclimate and microarthropod
abundance and richness
In November, across treatments, there was a positive relation-
ship between soil moisture and total microarthropod richness, but
not between soil moisture and total microarthropod abundance
(Table 2). There were no relationships between soil temperature
and total microarthropod abundance and richness. There were
positive relationships between both collembola abundance and
collembola richness and soil moisture content. For collembola,
there were negative relationships between soil temperature and
abundance, and between soil temperature and richness, however,
these relationships did not hold when we controlled for soil mois-
ture (Table 2). For mites, there was no significant relationship
between abundance and richness and soil moisture content.
In June, there were no relationships between soil moisture
content and any of the measured microarthropod parameters. In
contrast, there was a positive relationship between soil tempera-
ture and abundance of mites; this relationship held even when soil
moisture was controlled for (Table 2).
4. Discussion
There is increasing recognition that climate changes can affect
soil organisms and the functions they provide (e.g., Schröter
et al., 2004; Hågvar and Klanderud, 2009; Kardol et al., 2010b;
Lindroth, 2010). Our results demonstrate that single and combined
effects of elevated [CO2], warming, and change in precipitation
regime can shape the abundance and community structure of soil
microarthropods, which are important regulators of ecosystem
processes. Further, we found that climate change effects on soil
microarthropods varied by sampling date, and importantly, that
the major taxonomic groups of soil microarthropods, i.e., collem-
bola and mites, differed in their response to our climate change
treatments.
We found weak support for our hypothesis that the precipitation
treatment would have the strongest impact on the abundance and
community structure of soil microarthropods. Total microarthro-
pod abundance was not affected by the precipitation treatment,
and this could be largely attributed to the lack of response of the
mites, the most abundant group of microarthropods found in our
system. While positive relationships between mite abundance and
soil moisture have been established across a range of ecosystems
(e.g., Lindberg et al., 2002; Badejo and Akinwole, 2006; Chikoski
et al., 2006; Classen et al., 2006), mites might be adapted to strong
seasonal fluctuations in soil moisture content typical in old-field
ecosystems in the southeastern USA. Hence, the relatively small dif-
ference in soil moisture content between wet and dry treatments
may have been insufficient to cause major shifts in mite abun-
dance. Harte et al. (1996) suggest that summer minimum water
content may be more critical for mite abundance than the across-
year average. While summer moisture levels in our experiment may
have been critically low (5–10%), the summer minima did not differ
substantially between wet and dry treatments (Fig. 1A).
While mite abundance was not responsive to our precipitation
treatments, the abundance of collembola was. Collembola abun-
dance tended to be lower in dry than in wet treatments and it
was positively related to soil moisture in the November collection.
However, in June, there was no relationship between soil moisture
content and collembola abundance. This discrepancy between the

Author's personal copy
42 P. Kardol et al. / Applied Soil Ecology 47 (2011) 37–44
Table 1
Result from ANOVA testing the effects of climate change treatments on abundance and richness of total microarthropods, mites, and collembola in November 2007 and June 2008. df = denominator degrees of freedom, as estimated
by the Kenward–Rogers method, which determines df for each variable dependent on the covariance structure related to the random effects. Bold values indicate p≤0.05.
Microarthropods (total) Mites Collembola
November June November June November June
df Fpdf Fpdf Fpdf Fpdf FpDf Fp
Abundance
Precipitation 16 2.40 0.14 8 4.06 0.08 14 0.36 0.56 16 2.40 0.14 8 4.26 0.07 14 0.43 0.52
[CO2] 16 0.00 0.96 8 0.35 0.57 14 0.01 0.91 16 0.00 0.96 8 0.32 0.59 14 0.04 0.85
Temperature 16 0.23 0.64 8 2.31 0.17 14 1.57 0.23 16 0.23 0.64 8 2.44 0.16 14 1.61 0.23
Precipitation ×[CO2] 16 0.06 0.81 8 0.23 0.64 14 0.02 0.89 16 0.06 0.81 8 0.19 0.68 14 0.12 0.73
Precipitation ×temperature 16 0.00 0.97 8 2.93 0.13 14 0.55 0.47 16 0.00 0.97 8 3.05 0.11 14 0.55 0.47
[CO2]×temperature 16 0.60 0.45 8 0.13 0.73 14 0.89 0.36 16 0.60 0.45 8 0.11 0.75 14 1.00 0.33
Precipitation ×[CO2]×temperature 16 1.38 0.26 8 1.16 0.31 14 2.57 0.13 16 1.38 0.26 8 0.95 0.36 14 2.71 0.12
Richness
Precipitation 16 9.88 0.01 8 2.22 0.17 16 3.03 0.10 14 3.91 0.07 8 4.08 0.08 6 1.48 0.26
[CO2] 16 1.31 0.27 6 0.02 0.88 16 0.12 0.73 14 0.16 0.70 8 0.32 0.59 6 0.07 0.80
Temperature 16 1.31 0.27 6 0.78 0.41 16 0.12 0.73 14 1.41 0.26 8 2.89 0.13 6 0.01 0.93
Precipitation ×[CO2] 16 1.31 0.27 8 0.33 0.58 16 0.48 0.50 14 0.63 0.44 8 0.08 0.78 6 0.03 0.87
Precipitation ×temperature 16 0.00 1.00 8 2.22 0.17 16 1.94 0.18 14 0.00 1.00 8 4.08 0.08 6 0.27 0.62
[CO2]×temperature 16 6.61 0.02 6 0.02 0.88 16 4.36 0.05 14 0.63 0.44 8 0.32 0.59 6 0.38 0.56
Precipitation ×[CO2]×temperature 16 2.04 0.17 8 4.75 0.06 16 3.03 0.10 14 3.91 0.07 8 0.75 0.41 6 0.03 0.87
Table 2
Relationship between soil moisture content (average one month prior to soil sampling) and soil microarthropod abundance and richness across climate change treatments. Data are shown for November 2007 and June 2008.
Soil moisture Soil temperature
November June November June
R2Sign F1,22 pR
2Sign F1,22 pR
2Sign F1,22 pR
2Sign F1,22 p
Abundance
Total microarthropods 0.02 −0.44 0.52 0.02 −0.52 0.48 0.01 + 0.17 0.69 0.14 + 3.63 0.07
Mites 0.02 −0.44 0.51 0.09 −2.06 0.16 0.07 + 1.56 0.22 0.35 + 12.06 <0.01
Collembola 0.37 + 13.13 <0.01 0.01 + 0.18 0.67 0.20 −5.53 0.03a0.00 −0.01 0.94
Richness
Total microarthropods 0.32 + 10.17 <0.01 0.01 −0.23 0.63 0.09 −2.30 0.14 0.06 + 1.32 0.26
Mites 0.02 + 0.51 0.48 0.00 + 0.06 0.81 0.00 −0.06 0.82 0.06 + 1.51 0.23
Collembola 0.38 + 13.55 <0.01 0.01 −0.14 0.71 0.19 −5.09 0.03a0.01 + 0.15 0.70
Bold values indicate p≤0.05.
aWhen soil moisture was controlled for in partial regression analysis, the relationship was no longer significant.

Author's personal copy
P. Kardol et al. / Applied Soil Ecology 47 (2011) 37–44 43
November and June patterns is somewhat counterintuitive as water
stress in the dry treatments was higher in summer than in winter.
However, collembola generally have long life cycles, and recov-
ery after a disturbance or drought event can take several months
(Lindberg and Bengtsson, 2005). We suspect collembola abundance
at the time of sampling reflected the integrated response of the
community over several months prior to sampling when soil mois-
ture conditions were higher (Fig. 1A).
In contrast to our expectations, we did not find significant
effects of warming and elevated [CO2] on abundance of soil
microarthropods, and these treatments did not alter the responses
to the precipitation treatment. Regression analyses indicated sig-
nificant relationships between soil microarthropod abundance
and soil temperature, thus we were surprised there was not a
detectable difference between the warming treatments. There was
a large amount of within-treatment variation in microarthropod
responses to warming and in the response of soil warming to the
treatments; this variation may explain the discrepancy between
our statistical tests.
Across treatments, the abundance of collembola was negatively
related to soil temperature. This relationship could be explained by
temperature-related shifts in soil moisture content. Indeed, while
temperature may directly affect survival and reproductive ability
of collembola (e.g., van Straalen, 1994), indirect effects through
changes in soil moisture may be of higher importance (e.g., Sinclair
and Stevens, 2006). For mites, we found a positive relationship
between soil temperature and abundance. This relationship was
found only in June, and could not be explained by related changes
in soil moisture content. We lack data on seasonal mite popula-
tion dynamics, but it could be that positive effects of increased soil
temperature during the preceding winter may have increased mite
abundance.
Reported effects of atmospheric [CO2] on soil microarthro-
pods include both negative and positive responses (see reviews:
Couteaux and Bolger, 2000; Lindroth, 2010). The concentration
of CO2in the soil is inherently high, thus we did not predict
that atmospheric [CO2] would directly impact on the abundance
of soil microarthropods (e.g., van Veen et al., 1991). However,
in our experiment, elevated [CO2], by regulating soil moisture
availability, can reduce the effect of warming and drought on
soil moisture content (Dermody et al., 2007). Therefore, posi-
tive indirect effects of elevated [CO2] on soil microarthropods
through enhanced water availability, and thus, interaction effects
of precipitation ×[CO2] were expected. Changes in soil moisture
in response to [CO2], however were small relative to the precipita-
tion treatment (Dermody et al., 2007), and treatment effects on soil
microarthropod abundance via changes in soil moisture were likely
minor.
Individual soil microarthropod taxa differed in their responses
to our climate change treatments, resulting in altered commu-
nity composition and richness. Not surprisingly, in November, the
highest taxon richness in our study was found in elevated [CO2]
and elevated temperature treatments. These treatments had the
highest soil moisture content (Fig. 1a; Dermody et al., 2007), and
there was a positive correlation between microarthropod richness
and soil moisture content. However, the richness patterns in our
study must be interpreted with caution. In June, the precipita-
tion treatment had no effect on community richness. Recovery
of soil microarthropod communities after severe disturbance pre-
dominantly depend on colonization from external species pools
(Kardol et al., 2009); however, our experimental design, i.e., open-
top chambers, would have constrained recovery through external
dispersal. Moreover, in November, abundance and richness of
microarthropods were positively related (R= 0.57); hence, we can-
not exclude the possibility that low community richness in dry
treatments was a result of under sampling.
Indirect effects of climate changes on soil microarthropods via
changes in plant productivity and community composition might
play an important role in net microarthropod responses to climate
changes (e.g., Jones et al., 1998; Loranger et al., 2004). In our exper-
iment, aboveground biomass production was generally higher in
wet than in dry treatments, and was enhanced by elevated [CO2]
and elevated temperature (Kardol et al., 2010a). Moreover, indi-
vidual plant species differed in their responses resulting in cover
dominance shifts (Engel et al., 2009; Kardol et al., 2010b). For exam-
ple, the cover of L. cuneata was much higher in wet than in dry
treatments (November: 53% versus 18%; June 51% versus 14%). Pre-
vious work in this system showed how shifts in proportional cover
of the dominant plant species, in response to the climate change
treatments, interacted with direct effects of climate change to shift
soil enzyme profiles as well as the taxonomic and functional com-
position of soil nematode communities (Kardol et al., 2010b). It is
likely that changes in plant community composition contributed to
the net climate change effects on soil microarthropods by altering
both the quality and quantity of resources that plants take up from
and return to the soil and by altering abiotic and structural soil con-
ditions (e.g., Hasegawa, 2001; Lindberg et al., 2002; Wardle et al.,
2006).
Soil biological responses to climate changes are complex, and
have seldom been considered in a realistic, multifactor frame-
work. In this light, our results provide new insights into how
simultaneously acting climate change factors may affect an eco-
logically important group of soil organisms, microarthropods. Most
importantly, after long-term applications of our treatments (which
allowed for multiple generation of the microarthropods studied),
we found that changes in soil moisture, either as direct effect of
changes in precipitation regime or as an indirect effect of warming
or elevated [CO2], had the largest effect on soil microarthropods,
and soil moisture appears to be more important for microarthropod
dynamics than the direct effects of warming and elevated [CO2].
However, the observed changes were often subtle, and we empha-
size that generalizations about the effects of climate changes on
soil microarthropods must be made with caution. Published stud-
ies provide conflicting data on soil microarthropod responses to
climate changes, with the possibility that responses are specific to
particular plant species, communities, or ecosystems. This strongly
accentuates the need of more explicit consideration of the environ-
mental and biological context in studies of climate change on soil
microarthropods.
Acknowledgements
We thank J.A. Weltzin who was integral in establishing and
designing this experiment. We thank E. Bernard for advice and
use of equipment, and L. Souza for help with the statistical anal-
yses. Research was sponsored by the U.S. Department of Energy,
Office of Science, Biological and Environmental Research Program,
Grant No. DE-FG02-02ER63366, and work was conducted in collab-
oration with Oak Ridge National Laboratory, which is managed by
UT Battelle, LLC, for the U.S. Department of Energy under Contract
DE-AC05-00OR22725.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.apsoil.2010.11.001.
References
Badejo, M.A., Akinwole, P.O., 2006. Microenvironmental preferences of oribatid
mite species on the floor of a tropical rainforest. Exp. Appl. Acarol. 40, 145–
156.

Author's personal copy
44 P. Kardol et al. / Applied Soil Ecology 47 (2011) 37–44
Balogh, J., Balogh, P., 1992. The Oribatid Mites Genera of the World. The Hungarian
Natural Museum Press, Budapest.
Bardgett, R.D., Freeman, C., Ostle, N.J., 2008. Microbial contributions to climate
change through carbon cycle feedbacks. ISME J. 2, 805–814.
Beare, M.H., Parmelee, R.W., Hendrix, P.F., Cheng, W.X., Coleman, D.C., Crossley,
D.A., 1992. Microbial and faunal interactions and effects on litter nitrogen and
decomposition in agroecosystems. Ecol. Monogr. 62, 569–591.
Bokhorst, S., Huiskes, A., Convey, P., van Bodegom, P.M., Aerts, R., 2008. Climate
change effects on soil arthropod communities from the Falkland Islands and the
Maritime Antarctic. Soil Biol. Biochem. 40, 1547–1556.
Castro, H.F., Classen, A.T., Austin, E.E., Norby, R.J., Schadt, C.W., 2010. Soil microbial
community responses to multiple experimental climate change drivers. Appl.
Environ. Microb. 76, 999–1007.
Chikoski, J.M., Ferguson, S.H., Meyer, L., 2006. Effects of water addition on soil arthro-
pods and soil characteristics in a precipitation-limited environment. Acta Oecol.
Int. J. Ecol. 30, 203–211.
Christiansen, K., Bellinger, P., 1980–1981. The Collembola of North America North
of the Rio Grande, Grinnell College, Iowa.
Classen, A.T., DeMarco, J., Hart, S.C., Whitham, T.G., Cobb, N.S., Koch, G.W., 2006.
Impacts of herbivorous insects on decomposer communities during the early
stages of primary succession in a semi-arid woodland. Soil Biol. Biochem. 38,
972–982.
Cole, L., Dromph, K.M., Boaglio, V., Bardgett, R.D., 2004. Effect of density and species
richness of soil mesofauna on nutrient mineralisation and plant growth. Biol.
Fertil. Soils 39, 337–343.
Cotrufo, M.E., Ineson, P., 1995. Effects of enhanced atmospheric CO2and nutrient
supply on the quality and subsequent decomposition of fine roots of Betula
pendula Roth and Picea sitchensis (Bong). Carr. Plant Soil 170, 267–277.
Coulson, S.J., Hodkinson, I.D., Webb, N.R., Block, W., Bale, J.S., Strathdee, A.T., Wor-
land, M.R., Wooley, C., 1996. Effects of experimental temperature elevation on
high-arctic soil microarthropod populations. Polar Biol. 16, 147–153.
Couteaux, M.M., Bolger, T., 2000. Interactions between atmospheric CO2enrichment
and soil fauna. Plant Soil 224, 123–134.
Crossley, D.A., Blair, J.M., 1991. A high-efficiency, low-technology Tullgren-type
extractor for soil microarthropods. Agric. Ecosyst. Environ. 34, 187–192.
Davidson, E.A., Janssens, I.A., 2006. Temperature sensitivity of soil carbon decompo-
sition and feedbacks to climate change. Nature 440, 165–173.
Dermody, O., Weltzin, J.F., Engel, E.C., Allen, P., Norby, R.J., 2007. How do elevated
[CO2], warming, and reduced precipitation interact to affect soil moisture and
LAI in an old field ecosystem? Plant Soil 301, 255–266.
Edwards, N.T., Norby, R.J., 1999. Below-ground respiratory responses of sugar maple
and red maple saplings to atmospheric CO2enrichment and elevated air tem-
perature. Plant Soil 206, 85–97.
Engel, E.C., Weltzin, J.F., Norby, R.J., Classen, A.T., 2009. Responses of an old-field plant
community to interacting factors of elevated CO2, warming, and soil moisture.
J. Plant Ecol. 2, 1–11.
Filser, J., 2002. The role of Collembola in carbon and nitrogen cycling in soil. Pedo-
biology 46, 234–245.
Frampton, G.K., van den Brink, P.J., Gould, P.J.L., 2000. Effects of spring precipita-
tion on a temperate arable collembolan community analysed using principal
response curves. Appl. Soil Ecol. 14, 231–248.
Garten, C.T., Classen, A.T., Norby, R.J., 2009. Soil moisture surpasses elevated CO2
and temperature as a control on soil carbon dynamics in a multi-factor climate
change experiment. Plant Soil 319, 85–94.
Garten, C.T., Classen, A.T., Norby, R.J., Brice, D.J., Weltzin, J.F., Souza, L., 2008. Role
of N2-fixation in constructed old-field communities under different regimes of
[CO2], temperature, and water availability. Ecosystems 11, 125–137.
Hågvar, S., Klanderud, K., 2009. Effect of simulated environmental change on alpine
soil arthropods. Global Change Biol. 15, 2972–2980.
Hansen, R.A., Williams, R.S., Degenhardt, D.C., Lincoln, D.E., 2001. Non-litter effects
of elevated CO2on forest floor microarthropod abundances. Plant Soil 236,
139–144.
Harte, J., Rawa, A., Price, V., 1996. Effects of manipulated soil microclimate on meso-
faunal biomass and diversity. Soil Biol. Biochem. 28, 313–322.
Hasegawa, M., 2001. The relationship between the organic matter composition of a
forest floor and the structure of a soil arthropod community. Eur. J. Soil Biol. 37,
281–284.
Huhta, V., Hänninen, S.M., 2001. Effects of temperature and moisture fluctuations
on an experimental soil microarthropod community. Pedobiology 45, 279–286.
IPCC [Intergovernmental Panel on Climate Change], 2007. Climate Change 2007: The
Physical Science Basis. Cambridge University Press, Cambridge, UK.
Jones, T.H., Thompson, L.J., Lawton, J.H., Bezemer, T.M., Bardgett, R.D., Blackburn,
T.M., Bruce, K.D., Cannon, P.F., Hall, G.S., Hartley, S.E., Howson, G., Jones, C.G.,
Kampichler, C., Kandeler, E., Ritchie, D.A., 1998. Impacts of rising atmospheric
carbon dioxide on model terrestrial ecosystems. Science 280, 441–443.
Kardol, P., Newton, J.S., Bezemer, T.M., van der Putten, W.H., 2009. Contrasting diver-
sity patterns of soil mites and nematodes in secondary succession. Acta Oecol.
35, 603–609.
Kardol, P., Campany, C.E., Souza, L., Norby, R., Weltzin, J.F., Classen, A.T., 2010a.
Climate change effects on plant biomass alter dominance patterns and com-
munity evenness in an experimental oldfield ecosystem. Global Change Biol. 16,
2676–2687.
Kardol, P., Cregger, M.A., Campany, C.E., Classen, A.T., 2010b. Soil ecosystem func-
tioning under climate change: plant species and community effects. Ecology 91,
767–781.
Lepˇ
s, J., ˇ
Smilauer, P., 2003. Multivariate analysis of ecological data using CANOCO.
Cambridge University Press, Cambridge, United Kingdom.
Lindberg, N., Bengtsson, J., 2005. Population responses of oribatid mites and collem-
bolans after drought. Appl. Soil Ecol. 28, 163–174.
Lindberg, N., Bengtsson, J., Persson, T., 2002. Effects of experimental irrigation and
drought on the composition and diversity of soil fauna in a coniferous stand. J.
Appl. Ecol. 39, 924–936.
Lindroth, R.L., 2010. Impacts of elevated atmospheric CO2and O3on forests: phy-
tochemistry, trophic interactions, and ecosystem dynamics. J. Chem. Ecol. 36,
2–21.
Loranger, G.I., Pregitzer, K.S., King, J.S., 2004. Elevated CO2and O3t concentrations
differentially affect selected groups of the fauna in temperate forest soils. Soil
Biol. Biochem. 36, 1521–1524.
Lussenhop, J., Treonis, A., Curtis, P.S., Teeri, J.A., Vogel, C.S., 1998. Response of soil
biota to elevated atmospheric CO2in poplar model systems. Oecologia 113,
247–251.
Moron-Rios, A., Rodriguez, M.A., Perez-Camacho, L., Rebollo, S., 2010. Effects of
seasonal grazing and precipitation regime on the soil macroinvertebrates of a
Mediterranean old-field. Eur. J. Soil Biol. 46, 91–96.
Niedbala, W., 2002. Ptyctimous Mites (Acari, Oribatida) of the Nearctic Region.
Monographs of the Upper Silisian Museam, No. 4, Bytom.
Norby, R.J., Edwards, N.T., Riggs, J.S., Abner, C.H., Wullschleger, S.D., Gunderson, C.A.,
1997. Temperature-controlled open-top chambers for global change research.
Global Change Biol. 3, 259–267.
O’Lear, H.A., Blair, J.M., 1999. Responses of soil microarthropods to changes in soil
water availability in tallgrass prairie. Biol. Fertil. Soils 29, 207–217.
Petersen, H., Luxton, M., 1982. A comparative analysis of soil fauna populations and
their role in decomposition processes. Oikos 39, 287–388.
Pflug, A., Wolters, V., 2001. Influence of drought and litter age on Collembola com-
munities. Eur. J. Soil Biol. 37, 305–308.
Sackett, T.E., Classen, A.T., Sanders, N.J., 2010. Linking soil food web structure
to above- and below-ground ecosystem processes: a meta-analysis. Oikos,
doi:10.1111/j.1600-0706.2010.18728.x.
Schröter, D., Brussaard, L., De Deyn, G., Poveda, K., Brown, V.K., Berg, M.P., War-
dle, D.A., Moore, J., Wall, D.H., 2004. Trophic interactions in a changing world:
modeling aboveground–belowground interactions. Basic Appl. Ecol. 5, 515–528.
Seastedt, T.R., 1984. The role of microarthropods in decomposition and mineraliza-
tion processes. Ann. Rev. Entomol. 29, 25–46.
Sinclair, B.J., Stevens, M.I., 2006. Terrestrial microarthropods of Victoria Land and
Queen Maud Mountains Antarctica: implications of climate change. Soil Biol.
Biochem. 38, 3158–3170.
Sjursen, H., Michelsen, A., Jonasson, S., 2005. Effects of long-term soil warming
and fertilisation on microarthropod abundances in three sub-arctic ecosystems.
Appl. Soil Ecol. 30, 148–161.
ter Braak, C., ˇ
Smilauer, P., 2002. CANOCO Reference Manual and CanoDraw for Win-
dows User’s Guide: Software for Canonical Community Ordination (Version 4.5).
Microcomputer Power, Ithaca, NY, USA.
Tsiafouli, M.A., Kallimanis, A.S., Katana, E., Stamou, G.P., Sgardelis, S.P., 2005.
Responses of soil microarthropods to experimental short-term manipulations
of soil moisture. Appl. Soil Ecol. 29, 17–26.
Uvarov, A.V., 2003. Effects of diurnal temperature fluctuations on population
responses of forest floor mites. Pedobiology 47, 331–339.
van Straalen, N.M., 1994. Adaptive significance of temperature responses in Collem-
bola. Acta Zool. Fennica 195, 135–142.
van Veen, J.A., Liljeroth, E., Lekkerkerk, L.J.A., Van de Geijn, S.C., 1991. Carbon fluxes
in plant-soil systems at elevated atmospheric CO2levels. Ecol. Appl. 1, 175–
181.
Wan, S., Norby, R.J., Ledford, J., Weltzin, J.F., 2007. Responses of soil respiration to
elevated CO2, air warming, and changing soil water availability in a model old-
field grassland. Global Change Biol. 13, 2411–2424.
Walter, D.E., Proctor, H.C., 1999. Mites. Ecology, Evolution and Behaviour. CABI Pub-
lishing, Wallingford, Oxon, UK.
Wardle, D.A., Yeates, G.W., Barker, G.M., Bonner, K.I., 2006. The influence of plant
litter diversity on decomposer abundance and diversity. Soil Biol. Biochem. 38,
1052–1062.