Download Differential response of ants to nutrient addition in a tropical Brown

Soil Biology & Biochemistry 46 (2012) 10e17
Contents lists available at SciVerse ScienceDirect
Soil Biology & Biochemistry
journal homepage: www.elsevier.com/locate/soilbio
Differential response of ants to nutrient addition in a tropical Brown Food Web
Justine Jacquemin a, b, *, Mark Maraun c, Yves Roisin b, Maurice Leponce a
a
Section of Biological Evaluation, Royal Belgian Institute of Natural Sciences, Rue Vautier 29, B-1000 Brussels, Belgium
Evolutionary Biology & Ecology, Université Libre de Bruxelles, Belgium
c
J.F. Blumenbach Institute of Zoology and Anthropology, Animal Ecology, Georg August University of Göttingen, Germany
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 26 July 2011
Received in revised form
18 October 2011
Accepted 9 November 2011
Available online 24 November 2011
In tropical ecosystems, access to nitrogen (N) and phosphorus (P) limits the decomposition rate of leaflitter. Leaf-litter ants are abundant in this microhabitat and present a wide variety of diets. Our aim was
to study the response of various ant trophic groups to an increased nutrient availability which boosts the
decomposition of their habitat and selectively affects the abundance of their prey.
A 6-month nutrient addition experiment (CN, CNP) was performed in a tropical montane forest of the
Ecuadorian Andes. The density of ants, of other predators (e.g. arachnids, beetles) and of their potential
prey (mesofauna, ranging from 0.1 to 2 mm) was measured in treatments and control plots.
The litter volume in fertilized plots decreased significantly. Collembola and total mesofauna density
were enhanced by the CNP addition. Ants responded differentially according to their trophic group:
despite increased prey availability, predatory species in general and collembolan hunters in particular
were negatively affected by both treatments. Other ant trophic group densities did not change. By
contrast, the density of Dermaptera increased with the treatments. A complementary isotopic approach
allowed us to trace carbon fluxes through the food web.
Our results suggest that the nutrient input enhanced the litter decomposition rate, leading to
reduction of habitat size. They also suggest that predatory ants in tropical leaf-litter food webs are
limited by habitat size rather than by prey availability, and that these ants are more affected by habitat
loss than their prey, other ant trophic groups and other macrofauna taxa.
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
Trophic structure
Stable isotopes
Carbon
Nitrogen
Phosphorus
Tropical forest
Bottom-up
1. Introduction
In tropical Brown Food Webs (BFW) (Kaspari, 2004), leaf-litter
decomposing microbes (fungi and bacteria) are grazed by microbivores, such as collembolans and oribatid mites (Moore et al.,
1988), which in turn are consumed by several predator taxa:
mesostigmatid mites (Uropodina and Gamasina), Arachnida
(spiders, pseudoscorpions, opiliones), predatory beetles (e.g.
Staphylinidae) and ants (Schaefer, 1990; Schneider and Maraun,
2009). Leaf-litter constitutes thus both the habitat and the food
reservoir of leaf-litter ants, and it has been shown that an experimental increase of food resource leads to higher decomposition
rate, and, subsequently, to the destruction of the habitat for ants
(Shik and Kaspari, 2010). However, there is a large diversity of
dietary habits among litter ants, from fungus growers to specialized
trap-jaw predators, honeydew tenders or omnivorous genera such
* Corresponding author. Section of Biological Evaluation, Royal Belgian Institute
of Natural Sciences, Rue Vautier 29, B-1000 Brussels, Belgium. Tel.: þ32 2 627 43 57.
E-mail address: [email protected] (J. Jacquemin).
0038-0717/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.soilbio.2011.11.007
as Pheidole and Solenopsis (Blüthgen and Feldhaar, 2010), and it is
not known how the different ant trophic groups respond to an
increase of food resource availability, and whether they are more
limited by habitat or by food availability. In this study, we performed a 6-month nutrient addition experiment to evaluate the
importance of bottom-up forces in the BFW and to study the
response of several taxa of the BFW, including ants according to
their trophic habits.
The decomposition of leaf-litter (i.e. the rate of detritus
conversion into microbial biomass) is limited by access to nutrients.
Soil and litter microorganisms are generally assumed to be limited
by the availability of carbon (C) and nitrogen (N) (Gallardo and
Schlesinger, 1994; Demoling et al., 2007). There is also evidence
that phosphorus (P) limits leaf-litter decomposition in tropical
forests (Hobbie and Vitousek, 2000; Kaspari et al., 2008a) for two
reasons: weathered tropical soils are phosphorus-deficient
(Vitousek and Sanford, 1986), and fast-growing microbes with
high titers of P-rich ribosomes are very demanding for P (Elser et al.,
1996). The effect cascades upwards as an increased proportion of
phosphorus in decomposing litter increases the abundance of
microbivores (Kaspari et al., 2008a; Kaspari and Yanoviak, 2009).
J. Jacquemin et al. / Soil Biology & Biochemistry 46 (2012) 10e17
Enriquez et al. (1993) have highlighted a positive relationship
between plant decomposition rates and N and P concentrations in
litter. Furthermore, according to the Structural Elements Hypothesis (Sterner and Elser, 2002), the nitrogen content of litter limits
the growth and abundance of silk-spinning invertebrates (spiders,
mesostigmatid mites and pseudoscorpions), as the demand for this
element is very high for these taxa (Xu and Lewis, 1990). Kaspari
et al. (2008a) have shown that the availability of potassium,
sodium and other micronutrients also play a role in cellulose and
leaf-litter decomposition rates.
Experimental additions of nutrients have demonstrated that
microbial activity is nutrient limited, both in temperate and tropical
forests (Scheu and Schaefer, 1998; Krashevska et al., 2010). Additions of C (as glucose), N and P lead to both higher microbial
biomass and respiration. Some of these experiments aim at
increasing the availability of resources (microorganisms) for the
microbivores and their predators and to observe changes in their
density in order to evaluate the strength of “bottom-up” forces
(Maraun et al., 2001; Milton and Kaspari, 2007). However,
increased resource availability may have diverse effects on microbivores. Glucose application had positive effects on collembolan
growth when enhancing microbial activity (Kaneda and Kaneko,
2004), but none of the animal groups studied by Scheu and
Schaefer (1998) showed a response parallel to that of the microorganisms. Maraun et al. (2001) reported that the responses of
microorganisms, mesofauna and macrofauna to nutrient additions
differed one from another, suggesting differential limitation of soil
and litter biota.
Moreover, the absence of numerical response to resource
supplementation is not necessarily due to the absence of either
trophic linkage or bottom-up control (Oelbermann et al., 2008).
Indeed, any increase in microbial activity leads to the destruction of
the litter habitat as detritus is converted into microbial biomass and
CO2 (Kaspari et al., 2008a; Kaspari and Yanoviak, 2009). This
supports the Ecosystem Size Hypothesis (Post et al., 2000) predicting that larger predators will be disproportionately limited by
habitat relative to their prey. In the particular case of leaf-litter ants,
it has been shown that higher resource availability enhances
microbial activity and finally leads to the destruction of habitat,
becoming thus limiting for ants (Shik and Kaspari, 2010).
Three resource addition experiments conducted in the tropics
(McGlynn, 2006; McGlynn et al., 2010; Shik and Kaspari, 2010) have
been performed by adding dead insects to the plots, but not all the
ants are predators. Few BFW studies include ants (but see Milton
and Kaspari, 2007; Shik and Kaspari, 2010), and neither nutrient
addition experiments nor BFW studies have ever distinguished
trophic groups among litter ants. In this regard, stable isotopes are
a useful tool for determining trophic habits of animals, since the
combined measurement of N and C stable isotope ratios in animal
tissue provides a time integrated and detailed information on their
diet and trophic position. The abundance of nitrogen stable isotope
(d15N) is used to estimate the trophic position of each taxon since
the d15N of a consumer is typically enriched by 3.4& relative to its
diet (Post, 2002; Maraun et al., 2008, 2011).
In this study, our objectives were (a) to determine, through an
isotopic approach (d15N signature), the number of trophic levels
that are occupied by leaf-litter ants in a tropical BFW, and to
determine their trophic positions relative to other taxa of the litter
fauna; (b) by performing nutrient addition experiments, to evaluate
the effect of increased food availability on the density of various
actors of the BFW, at different levels: grazing mesofauna, predatory
mesofauna, ants and non-ant macrofauna; (c) to study how food
flows from microbes to predators through the BFW, by tracing
carbon fluxes using an isotopically distinct carbon source (cane
sugar) in the nutrient addition; (d) to study the response of the ants
11
according to their trophic group and to determine which factor is
more limiting for leaf-litter ants: prey availability or volume of
microhabitat. We tested the Ecosystem Size Hypothesis (Post et al.,
2000) predicting that predators will be disproportionately limited
by habitat relative to their prey. We verified whether leaf-litter ants
respond differentially to nutrient addition treatments, with predatory ants being particularly affected by habitat loss.
2. Materials and methods
2.1. Study site
The study was conducted in Copalinga (4.0912 S, 78.9607 W),
a privately owned reserve on the eastern slope of the Ecuadorian
Andes, 1000 m a.s.l. This secondary forest is next to Podocarpus
National Park. High precipitations occur from February to June,
while August to December is drier (average annual rainfall:
2000 mm 387 SD; average annual temperature: 22.3 C 0.9 SD;
C. Vits, pers. comm., period: 2003e2011).
2.2. Nutrient addition
We delineated twenty-four 2 2 m plots: 8 control, 8 CNenriched and 8 CNP-enriched plots. We added the carbon and
nutrients as CN and CNP combinations, as it has been shown by
Krashevska et al. (2010) that, in tropical forests, saprophytic fungi
are mainly limited by C and N, and bacteria by P (with a maximum
microbial biomass reached in CNP treatments). Carbon and nutrients were added every two months (overall 3 applications)
(C ¼ 380 g m2; N ¼ 7 g m2; P ¼ 0.6 g m2 per application). The
total amount added was equivalent to 5 times the annual input of
these elements by litterfall (Sandmann, 2007; Krashevska et al.,
2010). The carbon source was cane sugar (saccharose from a C4
plant), the nitrogen source was ammonium nitrate (NH4NO3) and
the phosphorus was provided as sodium phosphate (Na2HPO4).
During each application, the plots were sprinkled with nutrients
and 4 L of rain water to initiate dissolution. Plots sprinkled with rain
water only served as control.
2.3. Sampling design
Six months after the first nutrient application, we extracted the
litter fauna by using the two most adapted methods according to
the size of the organisms studied.
The leaf-litter mesofauna (0.1e2 mm, Swift et al., 1979) was
extracted from soil cores. Two soil cores were collected inside each
plot (50 cm from the border, one east, the other west) with a soil
corer (ø 5.3 cm), only the upper 5 cm of organic layer was used for
mesofauna extraction. The mobile soil mesofauna (Collembola,
Oribatida, Gamasina, Uropodina) was extracted by heat using
a modified high gradient extractor (Macfadyen, 1961) for 4 days.
The specimens from the two samples per plot were pooled (n ¼ 8
per treatment).
The macrofauna (ants and non-ants) (>2 mm, Swift et al., 1979)
were extracted from the leaf-litter using mini-Winkler extractors
during 48 h. In each plot, only the central 2 m2 were collected in
order to avoid border effects (n ¼ 8 per treatment).
All specimens were conserved in a saturated NaCl solution and
stored at 18 C until identification and further processing. Salted
water preserves N and C isotopic ratios (unlike ethanol, which
modifies the 13C/12C ratio) (Ponsard and Amlou, 1999; Tillberg et al.,
2006).
The effect of nutrient addition on the leaf-litter fauna density
was studied on the meso- and macrofauna (ants and non-ants).
12
J. Jacquemin et al. / Soil Biology & Biochemistry 46 (2012) 10e17
2.4. Isotopic analysis
We determined the trophic position of leaf-litter taxa by
measuring their d15N and d13C signatures. The limits between the
different trophic levels were calculated in relation to the d15N
signature of a baseline, Graffenrieda emarginata (Ruiz & Pav.),
Melastomataceae, one of the most frequent trees on the nutrientpoor organic soil in Southern Ecuador (Haug et al., 2004).
We selected the 10 most abundant taxa of the mesofauna
(representing 75.9% of the collected individuals) for isotopic analysis and 20 species of leaf-litter ants (Table 1). The 20 ant species
were selected on the basis of three criteria: 1) their high abundance, only species present in at least three plots of a same treatment were selected (to allow replicated measurement); 2) their
presence in at least three control plots (to allow comparisons with
the treated plots); 3) their belonging to distinct trophic groups
(honeydew tenders/nectar eaters, generalist predators or omnivores, specialized predators, fungus or yeast eaters) (Davidson
et al., 2003; Davidson, 2005; Ryder Wilkie et al., 2010) to have
the largest possible range of isotopic signatures.
Whole ant worker bodies were analyzed, without removing the
abdomen (Davidson, 2005). The possible bias induced by this
procedure (Tillberg et al., 2006) was measured on a test species,
Octostruma balzani. We measured the d15N and d13C signatures of
the whole body of 5 pooled individuals (3 replicates from different
plots) and compared it to specimens with the abdomen removed
(head, thorax and legs of ten individuals pooled, n ¼ 3 replicates).
There was no significant difference between the isotopic signatures
of specimens processed with or without abdomen (ManneWhitney
U, p ¼ 0.827 for d15N and d13C signatures).
Between 1 and 31 ant workers, and between 1 and 121 mesofauna individuals were pooled into tin capsules to obtain
Table 1
List of the mesofauna taxa and ant species collected in control, CN- and CNPenriched plots, selected for isotopic analyses.
Taxon
Mesofauna
Collembola
Oribatida
Mesostigmata
Trophic Group
Small species
Entomobryoidea
Galumnidae
Haplozetidae
Scheloribatidae
Crotoniidae type 2
Gamasina
Veigaiidae
Ascoidea (Cheiroseius)
Uropodina Uropodidae type 1
Polyaspididae
Ants
Hypoponera sp.01
Nylanderia sp.01
Pheidole sp.04
Pheidole sp.07
Solenopsis sp.03
Apterostigma sp.01
Myrmicocrypta cf. tuberculata
Cyphomyrmex cf. rimosus
Acropyga fuhrmanni
Brachymyrmex sp.02
Crematogaster nigropilosa
Dolichoderus sp.01
Basiceros disciger
Brachymyrmex sp.03
Octostruma cf. balzani
Pachycondyla sp.02
Acanthognathus brevicornis
Strumigenys sp.01
Strumigenys sp.04
Strumigenys sp.07
Decomposer
Decomposer
Decomposer
Decomposer
Decomposer
Decomposer
Predator
Predator
Predator
Predator
a sufficient amount of material for both C and N stable isotope ratio
analysis. Then the samples were dried at 60 C for 24 h, weighed
and stored in a desiccator until analysis (n ¼ 2 or 3 replicates with
individuals from distinct plots). Samples were analyzed with an
elemental analyzer (NA 1500, Carlo Erba, Milan, Italy) coupled to
a mass spectrometer (Finnigan MAT 251, Bremen, Germany). The
abundance of heavy stable isotopes (d13C and d15N) is calculated as
follows: dX(&) ¼ (Rsample Rstandard)/Rstandard 1000. Rsample and
Rstandard represent the 13C/12C and 15N/14N ratios corresponding to
the samples and standard, respectively. For 13C, Pee Dee Belemnite
(PDB), and for 15N, atmospheric nitrogen, served as primary standard. Acetanilide (C8H9NO, Merck, Darmstadt, Germany) was used
for internal calibration. Due to the insignificant effect of nitrogen
addition on d15N signature of investigated taxa, data from
control, þCN and þCNP plots were pooled (ANOVA, F(2,187) ¼ 0.22,
p ¼ 0.802).
2.5. Carbon fluxes
In both CN- and CNP-treatments, carbon was added as cane
sugar (saccharose). Since sugarcane is a C4 plant, it has a higher
d13C signature (13&) than the C3 plants present at the study site
(27& on average) (Martinelli et al., 2002; Tiunov, 2007). We used
this isotopic difference to evaluate the relative contribution of
added sugar to animal tissue carbon, and to trace the flux of added
carbon through the food web. We calculated the relative contribution of cane sugar carbon to animal tissue carbon (f) using
a simple mixing model (Tiunov, 2007):
f ¼
dsample dsource2
.
ðdsource1 dsource2 Þ 100
dsource2 ¼ mean animal d13C value in control plots
dsource1 ¼ d13C signature of cane sugar ¼ 13&
dsample ¼ d13C signature of the sample
We measured the relative contribution of the added carbon into
the tissue of mesofauna grazers (Oribatida and Collembola), mesofauna predators (Gamasina and Uropodina) and ants. We segregated the analyzed ant species in 5 trophic groups: nectar eaters/
honeydew tenders, fungus growers, omnivores, large predators and
small predators (based on their d15N signature and a data survey of
the literature, see Ryder Wilkie et al., 2010; Davidson, 2005;
Davidson et al., 2003).
3. Results
3.1. Trophic position of BFW taxa
Omnivore
Omnivore
Omnivore
Omnivore
Omnivore
Fungus grower
Fungus grower
Yeast grower
Nectar/Honeydew eater
Nectar/Honeydew eater
Nectar/Honeydew eater
Nectar/Honeydew eater
Predator
Predator
Predator
Predator
Predator e Trap-jaw ant
Predator e Trap-jaw ant
Predator e Trap-jaw ant
Predator e Trap-jaw ant
The average d15N signatures of the animals ranged
between 0.43 and 9.88& (Fig. 1). Trophic levels were plotted
relative to the d15N signature of G. emarginata (1.15 0.13& SD) as
basal resource (Illig et al., 2005). Assuming that two trophic levels
are separated by a difference of 3.4& d15N due to fractionation
(Post, 2002), the gradient of 10.32 d units encompassed 4 trophic
levels. The ant Cyphomyrmex cf. rimosus, known as a yeast-eating
species, had the lowest d15N signature. Trophic levels I
(1.15e2.25 &) and II (2.25e5.65 &) were occupied mainly by
microbivores (Oribatida and Collembola), by fungus-eating ants
(Myrmicocrypta cf. tuberculata, Apterostigma sp.01) and by nectaror honeydew-eating ants (Dolichoderus sp.01, Crematogaster nigropilosa, Acropyga fuhrmanni, Brachymyrmex sp.02). Trophic level III
(5.65e9.05 &) contained most of the omnivorous (Hypoponera
sp.01, Pheidole and Solenopsis spp, presenting high intercolonial
variability) and predatory ant species. Two Uropodina taxa shared
J. Jacquemin et al. / Soil Biology & Biochemistry 46 (2012) 10e17
13
Fig. 2. Proportion of constitutive carbon in mesofauna (a) and ant (b) tissues originating from the cane sugar input varied according to the taxa considered and its
feeding habits. Small predatory ants show the same range of values as the mesofauna,
suggesting predation. By contrast, fungus-eating ants did not integrate the carbon
input, as they exclusively fed on their own fungus. Each point represents one replicate.
Fig. 1. d15N signatures (mean SD, n ¼ 2e9) of the litter-dwelling ants and mesofauna. Since there was no effect of the nutrient treatment on the d15N signatures,
samples from control, CN- and CNP-enriched plots were pooled.
the top trophic position (trophic level IV, >9.05&) with two other
ant species, Strumigenys sp.01 and Brachymyrmex sp.03.
3.2. Response to the nutrient addition experiment
3.2.1. Carbon fluxes in the BFW
We distinguished 4 groups among the mesofauna: Oribatida,
Collembola, Gamasina and Uropodina (the two first are grazers, the
last two are predators). 19e62% of the constitutive carbon of oribatid mites came from the sugar input; 27e47% for Collembola;
21e62% for Uropodina; and the highest values, 60e72%, for
Gamasina (Fig. 2a).
We segregated ants in 5 trophic groups: fungus eaters, honeydew
tenders or nectar eaters, omnivores, large predators and small
predators. Fungus-growing ants did not integrate carbon of cane
sugar origin. Between 2 and 22% of the constitutive carbon of large
predators and 0e25% of that of nectar/honeydew eaters originated
from the sugar input; these values ranged between 0 and 50% for the
omnivores and between 17 and 60% for small predators (Fig. 2b).
3.2.2. Response of microorganisms and leaf-litter fauna
The litter volume was on average 25% and 28% lower in þCN
and þCNP plots than in control (1.44 0.74 l/m2; 1.39 0.6 l/m2
and 1.93 0.76 l/m2 in þCN, þCNP and control plots respectively)
(ANOVA: F2,93 ¼ 5.587, p ¼ 0.005; post-hoc LSD: p ¼ 0.007 and
p ¼ 0.003, n ¼ 32, for þCN and þCNP plots respectively).
The increase of mesofauna density (all taxa pooled) with
nutrient addition was not significant in CN-enriched plots
(p ¼ 0.139) and significant in CNP-enriched plots compared to
control (Dunnett’s test, n ¼ 8, p ¼ 0.048). Among the grazers, Oribatida density did not change with the treatments, but the density
of collembolans increased significantly in CNP-enriched plots
compared to control (Dunnett’s test, n ¼ 8, p ¼ 0.002). Among the
predators of the mesofauna (Mesostigmata), the density of Gamasina increased significantly in both CN- and CNP-enriched plots
compared to control (Dunnett’s test, n ¼ 8, p ¼ 0.013 and 0.026
respectively) but not that of Uropodina (ANOVA: F2,21 ¼ 1.199,
p ¼ 0.321) (Fig. 3a).
We distinguished 5 trophic groups among leaf-litter ants: nectar
or honeydew eaters, omnivores, fungus or yeast growers, predators
and specialized predators (trap-jaw ants) (Table 1). We observed
a differential response among ant trophic groups. Omnivores,
fungus/yeast growers and nectar/honeydew eaters have not been
affected by nutrient additions. By contrast, we observed a significant decrease of the densities of predatory ant species, both in CNand CNP-enriched plots compared to control: all predatory species
pooled (Dunnett’s test, n ¼ 8, p ¼ 0.0005 and 0.002 for CN and CNP
treatments respectively), and trap-jaw ants only (Dunnett’s test,
n ¼ 8, p ¼ 0.009 for CN treatment) (Fig. 3b).
We also studied the effect of nutrient addition on the density of
non-ant arthropods from the leaf-litter macrofauna: two omnivore
taxa (Blattodea and Dermaptera), and two predator taxa, Arachnida
(i.e. Pseudoscorpiones, Opiliones, Araneae) and predatory beetles
(i.e. Staphylininae, Pselaphinae, Scydmaeninae). Blattodea density
did not change with the treatments (ANOVA: F2,21 ¼ 1.504,
p ¼ 0.245), but Dermaptera densities were significantly higher in
CN-enriched plots (Dunnett’s test, n ¼ 8, p ¼ 0.005) and marginally
significant in CNP-enriched plots compared to control (p ¼ 0.06).
Neither Arachnid nor predatory Coleoptera density changed with
treatments. The non-ant macrofauna density (all taxa pooled) did
not increase with treatments (Fig. 3c).
4. Discussion
4.1. Trophic position of BFW taxa
The d15N values of the mesofauna taxa and ant species spanned
over 10.32 d units. This is similar to the range of signatures
measured by Sandmann (2007) in a litter food web (11.1 d units), in
a study conducted in the same region as ours but at a higher
14
J. Jacquemin et al. / Soil Biology & Biochemistry 46 (2012) 10e17
Fig. 3. Density (number of individuals per litter volume unit) of mesofauna (a), ants (b) and non-ant macrofauna (c) in control, CN- and CNP-enriched plots. Despite higher density
of potential prey (mesofauna) in CNP-enriched plots, predatory ant densities decreased. In contrast, Dermaptera density increased with treatments.
altitude (1850 m a.s.l.). The d15N signatures of the ant species we
analyzed spanned over 10.30 d units, which is close to the value of
9.5 found by Davidson et al. (2003) in a Peruvian ant assemblage.
We distinguished 4 trophic levels, which is consistent with the
number of trophic levels commonly observed in terrestrial food
webs - between 3 and 4 - in both tropical and temperate systems
(Scheu and Falca, 2000; Illig et al., 2005; Maraun et al., 2011). This
suggests that the number of trophic levels is not determined by the
productivity and energy flow, but that it is rather limited by the low
energy-use efficiency of consumers in detrital systems (Illig et al.,
2005). We did not analyze the isotopic signature of non-ant predators from the macrofauna, such as spiders and predatory beetles,
but we assume that they are part of the third trophic level,
according to Sandmann (2007) who conducted a similar study in
the Ecuadorian Andes. A fifth trophic level could be constituted by
scavengers, if they were included in our study. Indeed, Oelbermann
and Scheu (2010) demonstrated that scavengers (Diptera and
Coleoptera) had the highest d15N signature in the arthropod
community of a temperate forest soil, and were significantly
enriched in 15N compared to predators.
4.2. Response of the leaf-litter organisms to the nutrient addition
experiment
One of the main goals of this study was to study the response of
leaf-litter ants to an increased availability of prey, by adding
resources without modifying the habitat (in contrast with litter
addition experiments, see Ponge et al., 1993; McGlynn, 2006).
J. Jacquemin et al. / Soil Biology & Biochemistry 46 (2012) 10e17
However, the manipulations of our study ultimately also resulted in
alterations of habitat volume. At the end of the experiment, the
amount of litter in the plots was lower in both CN and CNP treatments, probably due to increased microbial activity in enriched
plots, as shown in previous studies (Scheu and Schaefer, 1998;
Joergensen and Scheu, 1999; Krashevska et al., 2010). Indeed,
Krashevska et al. (2010) performed in the Ecuadorian Andes the
same nutrient addition as we did but at a higher altitude, and they
measured an increase of the fungal biomass by a factor of 6.2 and
7.9 in treatments with CN and CNP, respectively, and an increase of
the microbial biomass (mainly bacteria) by a factor of 8 and 16 in
treatments with CN and CNP, respectively. The treatments thus
enhanced food availability (i.e. microbes) for grazers (Collembola
and Oribatida). Collembola density increased with P addition,
suggesting they were either limited by P availability or food-limited
since microbes were limited by P availability. By contrast, Oribatida
density did not increase in enriched plots. Among the predators,
Uropodina did not respond to the nutrient additions, but Gamasina
did, both in CN and CNP treatments.
Accordingly, Maraun et al. (2001), Scheu and Schaefer (1998)
and Krashevska et al. (2010) reported that all meso- and macrofauna taxa do not show a response parallel to that of microorganisms to the nutrient addition, suggesting a differential limitation
among litter biota. This may be due to disturbance caused by the
macrofauna. Indeed, Maraun et al. (2001) have shown that, in spite
of higher food availability, Oribatida and Uropodina were very
sensitive to the disturbance caused by earthworms which were
more abundant in treated plots. A second reason may be a topdown control of the mesofauna density by predators (spiders,
ants). Milton and Kaspari (2007) observed a top-down control of
mesofauna densities by ants at a very small scale (100 cm2 nutrient
patches) absorbing the local increase of mesofauna. The non
increase in Oribatida density in treatments may be due to a selective top-down control of some of their predators, e.g. Pheidole ants
(Wilson, 2005).
However, when considering all taxa pooled (i.e. Collembola,
Oribatida, Gamasina, Uropodina), we measured an increase of
mesofauna density in CNP treatments, that is, a higher availability
of potential prey for the predators of the macrofauna (ants and nonants).
Ants showed a differential response to the nutrient supply,
according to their trophic group. The density of omnivores, fungus
and nectar eaters did not change with the nutrient supply.
Surprisingly, in spite of higher prey availability, we observed
a strong decrease of the predatory ant density in both treatments.
Among predatory ants, trap-jaw ant density did not increase
despite the higher density of Collembola in treatment with CNP.
Predatory ants seemed to be limited by another factor, presumably
the habitat. This is consistent with the Ecosystem Size Hypothesis
(Post et al., 2000). Shik and Kaspari (2010) demonstrated in a field
experiment that food could only become limiting to the litter ant
community when nest sites were supplemented on food-enriched
plots. Milton and Kaspari (2007) conducted a nutrient (N and P)
addition experiment in a Panamanian rainforest at very small scale
(100 cm2) and observed a local increase in ant densities but no
increase of microbivores abundance. At that scale, ants probably
used nutrient patches only as cafeteria, absorbing the local increase
in microbivores without suffering from habitat loss.
The density of non-ant macrofauna did not increase with
nutrient additions. Among the detritivore/omnivore taxa, Blattodea
did not respond to the treatments, but Dermaptera density
increased in both treatments, suggesting they were food-limited.
Indeed, Dermaptera are polyphagous insects adapting their diet
according to the availability of the resources (Ohgushi, 1988), and
may have benefited from the increase of mesofauna density in
15
treated plots. Among predatory taxa, neither predaceous beetles
nor arachnid density increased significantly with treatments.
Though, silk-spinning invertebrates are known to have high needs
in N (Xu and Lewis, 1990; Kaspari and Yanoviak, 2009), but they
were presumably limited by another factor in our treated plots:
according to the Ecosystem Size Hypothesis, these large predators
may have suffered from the loss of habitat.
In the CNP enrichment, P was added as Na2HPO4, bringing both
phosphorus and sodium to the plots. Sodium is carried inland by
organic aerosols and the sodium content in rainfall diminishes
strongly with increasing distance from the ocean (Stallard and
Edmond, 1981). Our study site is 150 km away from Pacific Ocean,
on the Eastern slope of the Andes which act as a natural barrier
against oceanic nutrient input (Stallard and Edmond, 1981; Salati
and Vose, 1984). Furthermore, inland ecosystems with high rainfall may experience significant sodium loss through leaching
(Vitousek and Sanford, 1986). Hence, it is likely that our study area
is Na-limited. Sodium shortage is known to slow down the carbon
cycle. Indeed, Kaspari et al. (2009) gave experimental evidence that
Na addition to the leaf-litter of an inland Amazon forest enhanced
litter decomposition rates. Termites, as decomposers, and ants, as
predators, increased in enriched plots, suggesting for the latter
cascade effects upwards through the food web. However, ants are
also Na-limited directly through their metabolic requirements
(Kaspari et al., 2008b). In our study, ant density did not increase in
the plots enriched in Na (and P). The positive effect of higher Na
availability could have been counterbalanced by some other
limiting effect, presumably the negative effect of reduced space
(Shik and Kaspari, 2010).
4.3. Tracing carbon fluxes in the BFW
Since root-associated vesicular arbuscular mycorrhizal fungi
receive carbon from the plant and are not able to incorporate the
added cane sugar (Smith and Read, 1997), mesofauna organisms
feeding on them did not incorporate the added carbon. On the
contrary, saprophytic fungi were able to integrate the cane sugar
carbon, and grazer organisms feeding on that resource hence
integrated the added carbon (Ekblad et al., 2002). Since 13C/12C
ratio changes very little from one trophic level to another, it is
therefore possible to evaluate the ultimate sources of carbon for an
organism when the isotopic signatures of the sources are different
(Post, 2002; Peterson & Fry, 1987).
The rate of assimilation of carbon originating from the added
cane sugar in grazer tissue (Oribatida and Collembola) indicates
that they integrated up to 62% of C4 carbon, suggesting that they
fed to some extent on saprophytic fungi and on bacteria, but also
partly on the vesicular arbuscular mycorrhizal fungi. Uropodina
integrated between 21 and 62% of C4 carbon, the same range of
values as grazers (19e62%), suggesting that Uropodina probably fed
on these organisms. Gamasina, the other group of predators of the
mesofauna, showed higher densities in both treatments, and up to
72% of C4 carbon assimilation rate, suggesting that they fed on prey
relying strongly on saprophytic fungi and bacteria. The results of C4
carbon assimilation rates suggest that all the investigated mesofauna taxa integrated at least partly the added carbon, directly by
feeding on saprophytic fungi, or indirectly by predation on the
grazers, but it did not result in an increase of density, except for
Collembola and Gamasina.
Even if we did not observe any increase of global ant density,
tracing C4 carbon flux gave evidence that ants differentially integrated the added carbon by feeding on food resources relying or not
on saprophytic fungi. Only fungus growers did not integrate the
added C4 carbon at all, as they feed specifically on their own fungus.
Nectar and honeydew eaters probably integrated small amounts of
16
J. Jacquemin et al. / Soil Biology & Biochemistry 46 (2012) 10e17
C4 carbon (0e25%) by feeding partly directly on the sugar.
However, the rate of integration is very low as they principally fed
on nectar or honeydew, and as plants did not integrate the added
carbon. Omnivores are opportunistic feeders and probably fed
partly on microbivores or on small arthropods feeding partly on
saprophytic fungi (0e50%). Large predators integrated the cane
sugar carbon in a low range (2e22%) since they were large and
mobile, and they probably fed partly outside the plots on prey that
was not in contact with the added carbon. In contrast, small
predatory ants integrated C4 carbon to the largest extent (17e60%)
since they mainly fed inside the plots (due to their small size and
limited foraging extent) on the mesofauna taxa (Oribatida and
Collembola also showed 19e62% of C4 carbon integration in their
tissue). d15N signatures of omnivorous and predatory ants did not
change with the sugar (and nutrient) addition, suggesting these
animals did not feed directly on the sugar.
These results give a general overview of the food fluxes inside
the BFW, from microbes to ants, but we would need further
investigation to highlight more specific trophic relationships.
However, the combined analysis of isotopic ratios and invertebrate
density allowed us to trace the food fluxes in the Brown Food Web
and to detect linkages to the decomposer system, even in species
which did not respond numerically. The isotopic approach confirms
that predatory ants benefited from the nutrient supply to some
extent, but some competition may have occurred with other
predatory taxa of the macrofauna, reinforcing the unfavorable
effect of habitat loss on leaf-litter ants.
Acknowledgments
We thank Catherine Vits and Boudewijn De Roover, owners of
the Nature Reserve Copalinga, Zamora, Ecuador, for allowing us to
conduct research in their estate. We thank Jaime Peña for his help
in the field and Delia Magaly Montalvan Carrión from the Universidad Técnica Particular de Loja for help in the lab. This work
was supported by fellowships from the National Fund for Scientific
Research (FNRS, Belgium) to ML and JJ (PhD grant, Fonds pour la
Formation à la Recherche dans l’Industrie et l’Agriculture e FRIA,
Belgium) who also received a grant from the Fonds Agathon de
Potter (Académie Royale de Belgique, Belgium) to conduct the
collection of samples in Ecuador. We would like to thank also
Dorothee Sandmann from Georg-August University of Göttingen,
Germany, for her help in mite identification. We thank two anonymous reviewers for their helpful comments and suggestions.
References
Blüthgen, N., Feldhaar, H., 2010. Food and shelter: how resources influence ant
ecology. In: Lach, L., Parr, C.L., Abbott, L. (Eds.), Ant Ecology. Oxford University
Press, New York, USA, pp. 115e136.
Davidson, D.W., Cook, S.C., Snelling, R.R., Chua, T.H., 2003. Explaining the abundance
of ants in lowland tropical rainforest canopies. Science 300, 969e972.
Davidson, D.W., 2005. Ecological stoichiometry of ants in a New World rain forest.
Oecologia 142, 221e231.
Demoling, F., Figueroa, D., Bååth, E., 2007. Comparison of factors limiting bacterial
growth in different soils. Soil Biology & Biochemistry 39, 2485e2495.
Ekblad, A., Nyberg, G., Högberg, P., 2002. 13C-discrimination during microbial
respiration of added C3-, C4- and 13C-labelled sugars to a C3-forest soil.
Oecologia 131, 245e249.
Elser, J.J., Dobberfuhl, D.R., MacKay, N.A., Schampel, J.H., 1996. Organism size, life
history, and N: P stoichiometry. BioScience 46, 674e684.
Enriquez, S., Duarte, C.M., Sandjensen, K., 1993. Patterns in decomposition rates
among photosynthetic organisms e the importance of detritus CeNeP content.
Oecologia 94 (4), 457e471.
Gallardo, A., Schlesinger, W.H., 1994. Factors limiting microbial biomass in the
mineral soil and forest floor of a warm temperate forest. Soil Biology &
Biochemistry 26, 1409e1415.
Haug, I., Lempe, J., Homeier, J., Weiss, M., Setaro, S., Oberwinkel, F., Kottke, I., 2004.
Graffenrieda emarginata (Melastomataceae) forms mycorrhizas with Glomeromycota and with a member of the Hymenoscyphus ericae aggregate in the
organic soil of a neotropical mountain rain forest. Canadian Journal of Botany
82, 340e356.
Hobbie, S.E., Vitousek, P.M., 2000. Nutrient limitation of decomposition in Hawaiian
forests. Ecology 81, 1867e1877.
Illig, J., Langel, R., Norton, R.A., Scheu, S., Maraun, M., 2005. Where are the decomposers? Uncovering the soil food web of a tropical montane rain forest in southern
Ecuador using stable isotopes (15N). Journal of Tropical Ecology 21, 589e593.
Joergensen, R.G., Scheu, S., 1999. Response of soil microorganisms to the addition of
carbon, nitrogen and phosphorus in a forest Rendzina. Soil Biology &
Biochemistry 31, 859e866.
Kaneda, S., Kaneko, N., 2004. Growth of the collembolan Folsomia candida Willem in
soil supplemented with glucose. Pedobiologia 48, 165e170.
Kaspari, M., 2004. Using the metabolic theory of ecology to predict global patterns
of abundance. Ecology 85, 1800e1802.
Kaspari, M., Garcia, M.N., Harms, K.E., Santana, M., Wright, S.J., Yavitt, J.B., 2008a.
Multiple nutrients limit litterfall and decomposition in a tropical forest. Ecology
Letters 11, 35e43.
Kaspari, M., Yanoviak, S., Dudley, R., 2008b. On the biogeography of salt limitation:
a study of ant communities. Proceedings of the Natural Academy of Sciences of
the United States of America 105, 17848e17851.
Kaspari, M., Yanoviak, S.P., 2009. Biogeochemistry and structure of tropical brown
foodwebs. Ecology 90, 3342e3351.
Kaspari, M., Yanoviak, S.P., Dudley, R., Yuan, M., Clay, N.A., 2009. Sodium shortage as
a constraint on the carbon cycle in an inland tropical rainforest. Proceedings of
the Natural Academy of Sciences of the United States of America 106 (46),
19405e19409.
Krashevska, V., Maraun, M., Ruess, L., Scheu, S., 2010. Carbon and nutrient limitation
of soil microorganisms and microbial grazers in a tropical montane rain forest.
Oikos 119 (6), 1020e1028.
Macfadyen, A., 1961. Improved funnel-type extractors for soil arthropods. Journal of
Animal Ecology 30, 171e184.
Maraun, M., Alphei, J., Beste, P., Bonkowski, M., Buryn, R., Migge, S., Peter, M.,
Schaefer, M., Scheu, S., 2001. Indirect effects of carbon and nutrient amendments on the soil meso- and microfauna of a beechwood. Biology and Fertility
of Soils 34, 222e229.
Maraun, M., Illig, J., Sandmann, D., Krashevska, V., Norton, R.A., Scheu, S., 2008. Soil
fauna: composition and function. In: Beck, E., Bendix, J., Kottke, I., Makeschin, F.,
Mosandl, R. (Eds.), Gradients in a Tropical Montane Ecosystem of Ecuador.
Ecological Studies, vol. 198. Springer, Berlin, pp. 181e192.
Maraun, M., Erdmann, G., Fischer, B.M., Pollierer, M.M., Norton, R.A., Schneider, K.,
Scheu, S., 2011. Stable isotopes revisited: their use and limits for oribatid mite
trophic ecology. Soil Biology & Biochemistry 43, 877e882.
Martinelli, L.A., Carmargo, P.B., Lara, L.B.L.S., Victoria, R.L., Artaxo, P., 2002. Stable
carbon and nitrogen isotopic composition of bulk aerosol particle in a C4 plant
landscape of southeast Brazil. Atmospheric Environment 36, 2427e2432.
McGlynn, T.P., 2006. Ants on the move: resource limitation of a litter-nesting ant
community in Costa Rica. Biotropica 38 (3), 419e427.
McGlynn, T.P., Weiser, M.D., Dunn, R.R., 2010. More individuals but fewer species:
testing the ‘more individuals hypothesis’ in a diverse tropical fauna. Biology
Letters 6, 490e493.
Milton, Y., Kaspari, K., 2007. Bottom-up and top-down regulation of decomposition
in a tropical forest. Oecologia 153, 163e172.
Moore, J.C., Walter, D.E., Hunt, H.W., 1988. Arthropod regulation of micro- and
mesobiota in below-ground detrital food webs. Annual Review of Entomology
33, 419e439.
Oelbermann, K., Langel, R., Scheu, S., 2008. Utilization of prey from the decomposer
system by generalist predators of grassland. Oecologia 155, 605e617.
Oelbermann, K., Scheu, S., 2010. Trophic guilds of generalist feeders in soil animal
communities as indicated by stable isotope analysis (15N/14N). Bulletin of
Entomological Research 100, 511e520.
Ohgushi, T., 1988. Temporal and spatial relationships between an herbivorous lady
beetle Epilachna niponica and its predator, the earwig Anechura harmandi.
Researches on Population Ecology 30, 57e68.
Peterson, B.J., Fry, B., 1987. Stable isotopes in ecosystem studies. Annual Review of
Ecology and Systematics 18, 293e320.
Ponge, J.F., Arpin, P., Vannier, G., 1993. Collembolan response to experimental
perturbations of litter supply in a temperate forest ecosystem. European Journal
of Soil Biology 29, 141e153.
Ponsard, S., Amlou, M., 1999. Effects of several preservation methods on the isotopic
content of Drosophila samples. Comptes Rendus de l’Académie des Sciences
322, 35e41.
Post, D.M., Pace, M.L., Hairston, N.G., 2000. Ecosystem size determines food-chain
length in lakes. Nature 405, 1047e1049.
Post, D.M., 2002. Using stable isotopes to estimate trophic position: models,
methods, and assumptions. Ecology 83 (3), 703e718.
Ryder Wilkie, K.T., Mertl, A.L., Traniello, J.F.A., 2010. Species diversity and distribution patterns of the ants of Amazonian Ecuador. PLoS One 5 (10), e13146.
doi:10.1371/journal.pone.0013146.
Salati, E., Vose, P.B., 1984. Amazon basin: a system in equilibrium. Science 225
(4658), 129e138.
Sandmann, D., 2007. Effects of nutrient addition on the soil food web of a tropical
mountain rain forest in southern Ecuador. Master thesis. Technische Universität
Darmstadt, 56 pp.
Schaefer, M., 1990. The soil fauna of a beech forest on limestone: trophic structure
and energy budget. Oecologia 82, 128e136.
J. Jacquemin et al. / Soil Biology & Biochemistry 46 (2012) 10e17
Scheu, S., Schaefer, M., 1998. Bottom-up control of the soil macrofauna community
in a beechwood on limestone: manipulation of food resources. Ecology 79,
1573e1585.
Scheu, S., Falca, M., 2000. The soil food web of two beech forests (Fagus sylvatica) of
contrasting humus type: stable isotope analysis of a macro- and a mesofaunadominated community. Oecologia 123, 285e296.
Schneider, K., Maraun, M., 2009. Top-down control of soil microarthropods e
Evidence from a laboratory experiment. Soil Biology & Biochemistry 41,
170e175.
Shik, J.Z., Kaspari, M., 2010. More food, less habitat: how necromass and leaf litter
decomposition combine to regulate a litter ant community. Ecological Entomology 35, 158e165. doi:10.1111/j.1365-2311.2009.01165.x.
Smith, S.E., Read, D.J.,1997. Mycorrhizal Symbiosis, second ed. London Academic Press.
Stallard, R.F., Edmond, J.M., 1981. Geochemistry of the Amazon, 1: precipitation
chemistry and the marine contribution to the dissolved load at the time of peak
discharge. Journal of Geophysical Research 86, 9844e9858.
17
Sterner, R.W., Elser, J.J., 2002. Ecological Stoichiometry: the Biology of Elements
from Molecules to the Biosphere. Princeton University Press, Princeton, New
Jersey, USA.
Swift, M.J., Heal, O.W., Andersen, J.M., 1979. Decomposition in Terrestrial Ecosystems. University of California Press, Berkeley, California, USA.
Tillberg, C.V., Mccarthy, D.P., Dolezal, A.G., Suarez, A.V., 2006. Measuring the trophic
ecology of ants using stable isotopes. Insectes Sociaux 53, 65e69.
Tiunov, A.V., 2007. Stable isotopes of carbon and nitrogen in soil ecological studies.
Biology Bulletin 34 (4), 395e407.
Vitousek, P.M., Sanford, R.L., 1986. Nutrient cycling in moist tropical forest. Annual
Review of Ecology and Systematics 17, 137e167.
Wilson, E.O., 2005. Oribatid mite predation by small ants of the genus Pheidole.
Insectes Sociaux 52, 263e265.
Xu, M., Lewis, R.V., 1990. Structure of a protein superfiber: spider dragline silk.
Proceedings of the Natural Academy of Sciences of the United States of America
87, 7120e7124.