Effects of Hydrologic Regime Changes on a Taxonomic and Functional Trait Structure of Earthworm Communities in...

Disturbances, both natural and anthropogenic, influence the patterning of species and species traits. The shift in species composition and distribution pattern of functional traits can demonstrate if the community is resistant, sensitive or resilient to the disturbance. Based on species- and trait-based approaches, we examined the response of the earthworm community to changing hydrologic conditions caused by the artificial drainage of mountain fens, in which cumulative effects of disturbance events over space and time are much less dynamic than in riverine wetlands. We hypothesized that the drainage-related changes of mountain fen peat soils have an effect on the earthworm community composition and its functional structure. We assume that the shift in species composition and value of community-weighted functional traits reflect changes in the resilience or resistance of the earthworm community to environmental change. Our results demonstrate that the total density of earthworms was almost three times lower under drained conditions compared to natural ones. Artificial drainage of fens had a neutral effect on the species-based diversity indices. However, there were species-specific traits that responded to hydrologic changes and which led to the species' replacements and to the co-occurrence of eurytopic, surface-browsing and more drought- and low-pH-resistant earthworm species in the drained fens. Based on these results, we conclude that abiotic-based environmental filtering was the main process responsible for sorting earthworms according to species and traits in the disturbed hydrologic conditions. The greater earthworm functional trait variations in semi-natural hydrologic conditions emphasizes the impact of transient dynamics in an altered disturbance regime on the earthworm assembly. Results also showed that in the changing hydrologic conditions of mountain fens, the functional trait approach revealed only slightly more predictive power than the taxonomic one, but it proved better with processes responsible for earthworm species filtering.