Collembolan communities in shrublands along climatic gradients in Europe and the effect of experimental warming and drought on population density, biomass and diversity

Sampling of arthropods from plants, soil surface and soil was carried out at six sites in shrublands across Europe as part of a multi-disciplinary, EU-sponsored ecosystem research project: ‘VULCAN: Vulnerability assessment of shrubland ecosystems in Europe under climatic changes’. Climatic gradients spanned from cold temperate heath- and moorland (Netherlands, Denmark, Wales) to Mediterranean maquis (Catalonia, Sardinia) and from moist moorland (Wales) to dry continental forest-steppe shrubland (Hungary). The ex- perimental setup at each site included three warming plots where radiation during the night was impeded, resulting in an average air temperature increase of 0.3–1.3 °C. (April–June 2003), three drought plots where precipitation was prevented during 1–4 months in the growing season and three control plots. Collembolan biomass and, less distinctly, population density and average individual weight in the control plots decreased from the northern to the southern sites, suggesting a latitudinal cline controlled by tempera- ture or interaction between temperature and moisture. In contrast, species richness tended to increase from North to South. The extremely dry Hungarian site with low biomass and number of species differed from this pattern and fit into a moisture-controlled cline from humid Atlantic to dry continental climates. These observations may reflect long-term adaptations to climate. Both warming and drought treatments resulted in significant effects on collembolan density and biomass. More significant effects were observed in the drought treatments than in the warming treatments. The most significant effects for total Collembola and individual collembolan species were found in the Dutch and the Spanish sites, while only few were found in the British, Hungarian and Italian sites. Nearly all significant effects of both warming and drought treatments were negative. Differential responses, i.e. negative, positive or unaffected, of individual species resulted in changes in community structure. The vertical distribution between plants, soil surface and soil showed an increasing proportion of collembolan biomass in the soil layer concurrently with a decreasing proportion in the soil surface layer from the northern to the southern sites. The same change in vertical distribution was observed in several sites as a response to the warming and drought treatments. It is concluded that the effects on collembolan communities resulting from clima- tic manipulations are very complicated and the result of many direct and indirect factors, partly acting in opposite directions.