Distant land use affects terrestrial and aquatic habitats of high naturalness

Biotas from all ecosystems need to respond to factors that determine habitat suitability. These factors originate from different scales. Effects can be assumed to be hierarchical in the order large-scale geographic > regional > local > small-scale in-habitat factors. We aimed at the identification of general patterns by comparisons between ecosystems (forest floor snails, hololimnic stream macroinvertebrates) and across scales, and include potential seasonal effects. Sampling sites displayed signs of naturalness, such as high levels of deadwood accumulation in the forests, or a lack of artificial stream bed fixation plus a good to high score for the assemblage-derived Multimetric Index (MMI) in the streams. Terrestrial and aquatic assemblages of non-emergent taxa fluctuated independent of seasonal effects. They differed in their relative correlation with environmental matrices with quasi-concentric effects in forests, and longitudinal effects in streams. Large-scale factors, namely geographic position, strongly influenced assemblage turnover, but the effect is based on a high covariation between geographic position and environmental factors. We thus extracted variables that best explained species turnover after correcting for spatio-temporal effects. The terrestrial community assembling was habitat-based and mainly responded to soil acidification, distance to disturbances, and regional scale deforestation and deciduous/mixed forest cover. The stream assemblages were structured by regional pasture cover, organic pollution, regional deciduous forest cover and microlithal cover. Apparently, community assembly occurs along with changes in regional forest cover and the transport of nutrients and matter that can originate from a distance, irrespective of ecosystem and assumed naturalness.