Unraveling the structure of soil food webs

In this work available methods for soil food web analysis were studied and new methods were evaluated. Soil biota consist of more species than the biota of any other environment on earth, however, despite their importance they are the least understood. Indirect methods are needed to get insight into soil food webs, because of the cryptic habitat and the small size of soil animals. A common used method is the analysis of the variation of natural occurring stable isotopes, mainly 15N/14N and 13C/12C. Nitrogen isotopes can be used to obtain trophic level information and carbon is used to determine food sources. Commonly the shift per trophic level is considered to be about 3.4permill for 15N/14N and 1permill for 13C/12C. But in this study delta15N varied from 2.4 to 6.3permill and delta13C from -1.0 to -3.3permill due to fungal food quality, starvation or age of soil animals (Collembola). This proves the impact of physiological status and food quality on stable isotope fractionation which should be considered when analysing food webs. Since the stable isotope method has its drawbacks, fatty acid (FA) analysis as alternative method for analysing soil food webs was investigated. FA analysis is a relatively new approach for investigating soil food webs. The FA pattern or specific biomarker FAs in consumers allow to infer food sources. To successfully apply this method, the influence of physiological and of environmental effects had to be tested. The FA pattern of Collembola was not significantly influenced by starvation, but food quality, life stage and temperature had an impact. Nevertheless, the specific marker FA (linoleic acid) for fungal food source made up over 20% of all FAs at all tested conditions compared to Collembola reared on other diets (2 - 14%). Other studies proved specific FA markers for leaves, nematodes and fungi; the present work revealed specific marker FAs for bacterial diet. Methyl branched FAs (i14:0, i15:0, a15:0 and i17:0) characterized Collembola reared on gram-positive bacteria, and a cyclic form (cy17:0) and 16:1omega5 were characteristic for gram-negative bacteria. A sophistication of FA analysis is the compound-specific $^{13}$C fatty acid analysis. The advantage of this method is that it delivers the 13C/12C ratio of specific dietary marker FAs. A similar 13C/12C ratio in the same FA in consumer and potential diet indicates trophic transfer and routing of the FA into the tissue of the consumer. The applicability of this analysis for soil food web investigations was tested in a field experiment including major functional groups, i.e. microorganisms as primary decomposers, euedaphic and epedaphic Collembola as secondary decomposers and cursorial and web-building spiders as top predators. Results of this study demonstrate that 13C/12C ratios of individual FAs in potential prey (diet) and consumers allow to identify carbon fluxes and trophic links. This work contributed to the establishment of tools to unravel the "enigma" of soil animal diversity and of the structure of soil food webs