a b s t r a c tFunctional traits (FT) offer a new framework to understand the ecology of organisms and overcome taxonomic difficulties that currently limit the study of minute soil taxa. FT are likely to be selected by environmental filters and hence they may provide more direct information on ecosystem characteristics than the species composition of a community.We tested the potential of testate amoeba (TA) functional traits as bioindicators of selected ecosystem processes in the context of a restored floodplain in north-western Switzerland. The floodplain was divided into six functional process zones (FPZs) associated to distinct post-restoration successional stages. We selected TA FT and computed three functional indices: functional richness (FRic), divergence (FDiv), evenness (FEve), and dispersion (FDis). We then compared the patterns of functional indices and classical diversity indices such as species richness, diversity and evenness. We assessed whether traits converged or were over-dispersed in the different FPZs using a randomization procedure. Finally, we related environmental variables and functional traits using the "Fourth Corner" statistic. This procedure enabled us to highlight relations that can potentially be used for bioindication. Promising candidates include the relationships between shell biovolume and vegetation structure and between shell compression and plant litter input variables.

No abstract

Size-structured food webs form integrated trophic systems where energy is being channelled from small to large consumers. Empirical evidence suggests that size structure prevails in aquatic ecosystems while in terrestrial food webs trophic level is largely independent of body size.Compartmentalisation of energy channeling according to size classes of consumers was suggested as a mechanism that underpins functioning and stability of terrestrial food webs including those belowground, but their structure has not been empirically assessed across the whole size spectrum.Here we used stable isotope analysis and metabolic regressions to describe size structure and energy use in eight belowground communities with consumers spanning 12 orders of magnitude in living body mass, from protists to earthworms. We showed a community-wide decline in trophic level with body mass in invertebrates and a remarkable non-linearity in community metabolism and trophic positions across all size classes. Specifically, we found that correlation between body mass and trophic level is positive in small-sized (protists, nematodes, arthropods below 1 µg in body mass), neutral in medium-sized (arthropods of 1 µg to 1 mg) and negative in large-sized consumers (large arthropods, earthworms), suggesting that these groups form compartments with different trophic organization. Based on this pattern, we propose a concept of belowground food webs being composed of (1) size-structured micro-food web driving fast energy channeling and nutrient release, e.g. in microbial loop, (2) arthropod macro-food web with no clear correlation between body size and trophic level, hosting soil arthropod diversity and subsidizing aboveground predators, and (3) 'trophic whales', sequestering energy in their large bodies and restricting its propagation to higher trophic levels in belowground food webs. The three size compartments are based on a similar set of basal resources, but contribute to different ecosystem-level functions and respond differently to variations in climate, soil characteristics and land use. We suggest that widely used vision of resource-based energy channeling in belowground food webs can be complemented with size-based energy channeling, where ecosystem multifunctionality, biodiversity and stability is supported by a balance across individual size compartments.

We investigated how the land-use change from rainforest into jungle rubber, intensive rubber and oil palm plantations affects decomposers and litter decomposition in Sumatra, Indonesia. Litterbags containing three litter types were placed into four land-use systems and harvested after 6 and 12 months. Litter mass loss and litter element concentrations were measured, and different microbial groups including bacteria, fungi and testate amoebae were studied. After 12 months 81, 65, 63 and 53% of litter exposed in rainforest, jungle rubber in oil palm and rubber plantations was decomposed. In addition to land use, litter decomposition varied strongly with litter type and short-term effects differed markedly from long-term effects. After 6 months, oil palm and rubber litter decomposed faster than rainforest litter, but after 12 months, decomposition of rainforest litter exceeded that of oil palm and rubber litter, reflecting adaptation of bacteria and fungi to decompose structural compounds in rainforest litter but not (or less) in rubber and oil palm litter. Bacterial and fungal community composition and testate amoeba species number and density varied strongly with litter type, but little with land use. However, community composition of testate amoebae was mainly affected by land use. Generally, changes in bacteria, fungi and testate amoebae were linked to changes in litter element concentrations, suggesting that element ratios of litter material as basal resource for the decomposer food web shape the structure of decomposer communities and decomposition processes via bottom-up forces. Overall, changing rainforest to monoculture plantations shifts the decomposer community structure and negatively affects litter decomposition.