Bottom-up or top-down control in forest soil microcosms? Effects of soil fauna on fungal biomass and C/N mineralisation

A major question in soil ecology is whether soil food webs are regulated by resources or by predators, i.e. bottom-up (donor) or top-down controlled. We tested the hypothesis that meso- and macrofaunal soil predators can regulate fungivore populations and, thereby cause a top-down cascade effect on fungal biomass and decomposition/mineralisation processes in boreal forest soils. The study was performed as a microcosm experiment with two contrasting soils (humus layers), one poor and one rich in N, and with different combinations of fungivore and predator soil fauna added to "defaunated" soil. In comparison with control microcosms lacking mesofauna (but with nematodes and protozoans), the presence of a diverse Collembola and Oribatida fungivore community significantly reduced the FDA-active fungal biomass or tended to reduce the ergosterol fraction of the fungal biomass in the N-poor humus, but no clear effect could be detected in the N-rich humus. Fungivores as well as fungivores plus predators (a predator community consisting of gamasids, spiders and beetles or a subset thereof) reduced C mineralisation and increased net N mineralisation in both soils. The presence of predators (particularly gamasid mites) reduced collembolan numbers and alleviated the negative effect of fungivores on fungal biomass in the N-poor soil. In the N-rich soil, the presence of predators increased fungal biomass (ergosterol) in relation to the "defaunated" soil. Therefore, a top-down trophic cascade could be detected in the N-poor humus but not in the N-rich humus. Our results suggest that the degree of top-down control in soil fauna communities depends on resource quality and soil fertility.