The formation of terrestrial food webs in glacier foreland: Evidence for the pivotal role of decomposer prey and intraguild predation
Introduction
The formation of terrestrial ecosystems during primary succession has been studied intensively, however, mainly from the view of the colonization of islands and adopting a plant perspective (Whittaker and Fernandez-Palacios 2007). Little is known on food webs formed early during primary succession. Investigating the structure of animal communities after glacial retreat Kaufmann (2001) noted that even at early successional stages the animal community appeared to be dominated by predators which has been confirmed from several other glacier forelands (Hodkinson et al., 2001, Gobbi et al., 2006, Vater, 2006). Dominance of predators prior to colonization by higher plants to provide primary productivity may be a widespread phenomenon in primary succession (Hodkinson et al. 2002). For instance it has also been reported from lava fields and pyroclastic flows (Ashmole and Ashmole, 1987, Edwards and Sugg, 2005). Predators may either live on allochthonous prey originating from nearby grassland or from prey out of the decomposer system. Indeed, it is increasingly recognized that prey out of the decomposer system significantly contributes to nutrition of aboveground predators (Settle et al., 1996, Oelbermann et al., 2008), thereby linking the above- and belowground food webs, and this may also be true early during primary succession (Wardle, 2002, Bardgett et al., 2005).
Studies investigating changes in food web structure during succession are hampered by the fact that the processes occur on time scales of decades, centuries or even millennia. Therefore, these processes typically are studied by investigating different sites in space which are taken to represent chronosequences (Bardgett et al. 2005). A classical example of such chronosequences which receive increasing interest are ecosystems which formed after the retreat of glaciers. The foreland of glaciers typically consists of sites of different age with the age increasing with distance from the glacial front allowing to accurately date sites of different distance from the current glacier forefront.
A particular problem of early terrestrial food webs is to identify the basal resources the food web is based on. Adopting an aboveground perspective it is commonly assumed the first consumer level consists of herbivores feeding on living plant tissue. However, this ignores that plants as the major primary producers rely on mineral nutrients which are provided by the decomposer system from decaying litter material which may be autochthonous but also originate from other ecosystems, i.e. being allochthonous. So far, however, the role of allochthonous resources for terrestrial food webs has been investigated mainly in coastal regions (Polis and Hurd, 1995, Polis et al., 2004). There is increasing evidence that the formation of terrestrial food webs during evolution started from simple systems comprising decomposing litter material and decomposer organisms with primary producers occurring later (Labandeira, 2005, Schaefer et al., 2010), and this may well also apply to food webs formed during primary succession.
In recent years understanding of the structure of food webs and identifying the basal resources of food webs has increased considerably due to analyzing variations in stable isotope ratios (Tiunov, 2007, Martinez del Rio et al., 2009). Variations in 15N/14N ratios have been shown to reflect the trophic position of consumers thereby allowing simple and reliable analyses of the trophic structure of food webs (Minagawa and Wada, 1984, Post, 2002, Martinez del Rio et al., 2009). Further, natural variations in 13C/12C ratios have been used intensively to identify basal resources of food webs as, in contrast to 15N/14N ratios, 13C/12C ratios change little with trophic position of consumers (Wada et al., 1991, Post, 2002, Martinez del Rio et al., 2009). However, so far stable isotope methodology has been used mainly in aquatic systems but recently is intensively employed for analysing terrestrial food webs (Ponsard and Arditi, 2000, Scheu and Falca, 2000, Tiunov, 2007).
Based on the rather enigmatic finding of Kaufmann (2001) that arthropod communities in glacier foreland comprise mainly predators we expected that analysis of natural variations in stable isotope ratios of potential food resources and consumers will allow to identify basal resources of the food webs and to understand how the structure of food webs change with age of terrestrial ecosystems. We hypothesized that the high abundance of predators in glacier foreland ecosystems in fact relies on prey out of the decomposer system, in particular Collembola. Further, by analyzing 13C/12C ratios we expected to be able to identify the basal resources of the food web. Finally, we expected that the trophic structure of food webs changes with the age of ecosystems and the associated increase in plant cover with prey out of the herbivore system becoming more important.
Section snippets
Sites
In 2005 samples were taken in the forefield of the Rotmoos Glacier (Austria). The Rotmoos valley is located in the Ötztal Alps of Northern Tyrol at 2250–2450 m asl (N46°50′, E11°02′). The valley is about 3 km in length and is surrounded by mountain peaks of up to 3000 m. It is located well above the timberline of the Central Alps which is at 1900–2000 m. The region was used for cattle and sheep grazing with sheep grazing continuing until today. Further, Marmots (Marmota marmota) and Alpine Goats (
Community composition
Average δ15N values of plants from the young, intermediate, and old site were −4.20 ± 0.87‰, −4.60 ± 1.85‰ and −3.97 ± 1.24‰ (means ± SD), respectively. To allow comparison of the signatures of animals between the sites their δ15N values were normalized to those of the young site according to the differences between plants, i.e. δ15N values of animals from the intermediate site were increased by 0.40‰ and those from the old site were reduced by 0.23‰. Only these adjusted data are presented and analyzed
Basal resources
Commonly studies on terrestrial food webs focus on the aboveground system assuming that plants function as basal resource with herbivores forming the first consumer level and predacious species feeding on herbivores the second (Pimm, 2002, Dunne, 2005). This view has been proven oversimplistic as it is becoming increasingly clear that the aboveground and belowground food web are closely interlinked (Polis and Strong, 1996, Oksanen et al., 1997, Scheu, 2001). Investigating the formation of
Acknowledgements
Data on temperature and precipitation were kindly provided by the Obergurgl Field Station and compiled by Meinhard Strobl.
References (42)
- et al.
A temporal approach to linking aboveground and belowground ecology
Trends Ecol. Evol.
(2005) - et al.
Feeding guilds in Collembola based on nitrogen stable isotope ratios
Soil Biol. Biochem.
(2005) Invasion of the continents: cyanobacterial crusts to tree-inhabiting arthropods
Trends Ecol. Evol.
(2005)- et al.
Stepwise enrichment of 15-N along food chains: further evidence and the relation between 15-N and animal age
Geochim. Cosmochim. Acta
(1984) Plants and generalist predators as links between the below-ground and above-ground system
Basic Appl. Ecol.
(2001)- et al.
The feeding ecology of elaterid larvae in central European arable land: new perspectives based on naturally occurring stable isotopes
Soil Biol. Biochem.
(2008) Wandering spiders limit densities of a major microbi-detritivore in the forest-floor food web
Pedobiologia
(2004)- et al.
On the relationship between productivity and food chain length at different ecological levels
Am. Nat.
(2007) - et al.
Arthropod communities supported by biological fallout on recent lava flows in the Canary Islands
Entomol. Scand.
(1987) - et al.
Primary faunal succession in volcanic terra lava cave studies on the Canary Islands
Biol. J. Linn. Soc.
(1992)