|A considerable amount of photosynthates is released into soil via living roots as rhizodeposits. This root-derived carbon (C) is an important source of energy and nutrients for microorganisms and higher trophic levels in soil food webs. However, as soil food webs are highly complex with considerable proportions of generalist feeders, it is not well studied how root-derived C affects trophic interactions between microorganisms and microbivores, such as Collembola. Therefore, in this thesis I traced the root-derived C flux into Collembola species belonging to three functional groups, i.e. epedaphic, hemiedaphic and euedaphic, in three cropping systems (oil-seed rape, grass and willow) using 13CO2 pulse labeling and bulk stable isotope 13C analysis. The contribution of bacterial and fungal channels to the incorporation of root-derived C into Collembola was investigated in more detail using compound specific 13C lipid analysis. To study the influences of vertical heterogeneity and root energy supply on the trophic structure of soil food webs in forest systems, I identified the trophic niches of Collembola functional groups in a root-trenching experiment using compound-specific isotope analysis of carbon and nitrogen (N) in amino acids of Collembola.
In the first study (chapter 2), using a pulse labeling experiment with 13CO2, I identified the incorporation of root-derived C into nine abundant species of Collembola belonging to three functional groups (epedaphic, hemiedaphic and euedaphic) over a period of 28 days in agricultural fields with different plant types, i.e. rape, grass and willow. The incorporation of 13C into Collembola was slower in rape than in grass and willow, suggesting a lower incorporation efficiency of root-derived C into soil food webs in rape as compared to grass and willow. In addition, the utilization of root-derived C and litter/old soil organic matter differed between functional groups/species of Collembola, with the differentiation being dependent on cropping system. These findings highlight that cropping system impacts the flux of root C into soil food webs, presumably due to differences in plant species, application of fertilizer and rhizosphere microbial communities. In addition, partitioning in the utilization of root-derived C and soil organic matter presumably is an important mechanism for local coexistence of Collembola species in soil.
A shortcoming of the first study (chapter 2) was the inability to disentangle the contribution of bacterial and fungal channels to the incorporation of root-derived C into Collembola. Therefore, I further identified the incorporation of root-derived 13C into bacterial and fungal biomarkers in microbial phospholipid fatty acids (PLFAs) in bulk soil and neutral lipid fatty acids (NLFAs) in five Collembola species belonging to different functional groups (chapter 3). Generally, Collembola incorporated more root-derived C from the bacterial channel in rape than in grass and willow, where fungi were the dominant C source. The proportional abundance of bacterial and fungal biomarkers in Collembola NLFAs corresponded with the 13C incorporation into microbial PLFAs, while it did not correlate with the proportion of microbial PFLAs in different cropping systems. This suggests that the differences in dietary composition of Collembola among cropping systems are mainly driven by changes in microbial community in the rhizosphere but not in bulk soil. Finally, hemiedaphic Collembola incorporated more root-derived C from bacterial resources than eu-/epedaphic Collembola. Hence, changes in root C flux into bacterial and fungal channels among cropping systems resulted in differential utilization of these resources by soil microbivores, suggesting that in particular microorganisms fueled by rhizodeposits are vital resources for the nutrition of higher trophic levels in soil food webs.
In contrast to arable soils, the soil in forest systems receives a considerable amount of aboveground inputs via dead organic matter, resulting in a strong vertical heterogeneity of available resources and environmental conditions. To investigate the vertical heterogeneity of trophic interactions in forest soil food webs and its response to deprivation of root energy supply, a state-of-the-art method, i.e. compound specific isotope analysis (CSIA) of C and N in amino acids, was used to identify the trophic niches of different functional groups of Collembola and their responses to the deprivation of root-derived resources in a root-trenching experiment (chapter 4). The 13C fingerprinting suggested saprotrophic microorganisms rather than mycorrhizal fungi to be the dominant resource for Collembola. The 15N values of phenylalanine as indicator of isotopic baseline were higher in euedaphic Collembola than in ep-/hemiedaphic Collembola, suggesting that euedaphic Collembola mainly utilized N from processed organic substrate in deeper soil, while leaf litter was the dominant N source for ep-/hemiedaphic Collembola. In addition, the trophic position calculated from CSIA (TPCSIA) of euedaphic Collembola was constantly higher than that of epedaphic Collembola, reflecting a higher number of trophic transfers in the soil than in the litter layer. Further, the deprivation of root energy supply reduced the TPCSIA of Collembola, suggesting that root derived C increases the number of trophic transfers and food chain length in soil food webs. However, this varied among functional groups of Collembola and forest stands, suggesting that the association between root energy supply and trophic interactions is mediated by vertical resource heterogeneity, regional conditions and feeding strategies of consumers.
Overall, results of this thesis advance our understanding of the important role of root-derived resources for soil food webs; root-derived C as major resource fueling soil food webs profoundly influences the niche partitioning in soil consumers, trophic interactions between microbes and microbivores and trophic structure of soil food webs.