Effects of fragmentation and trampling on carabid beetle assemblages in urban woodlands in Helsinki, Finland
We studied the effects of fragmentation (edge effects and patch size) and trampling (path cover) on carabid beetle assemblages in urban woodland patches in Helsinki, Finland. We expected that (1) open habitat and generalist species would benefit and forest species would suffer from increased woodland fragmentation, and (2) most carabid species would respond negatively to increased levels of trampling. A total of 2088 carabid individuals representing 37 species were collected. A cluster analysis distinguished sites in the interior of large woodland patches, with low or moderate path cover, from the other sites. The other sites did not cluster meaningfully, suggesting increased variation in the carabid fauna with increasing human impact. All species and ecological species-groups decreased with increasing distance from the edge toward the woodland interior and with increasing patch size. This pattern is in accordance with our expectation for open habitat and generalist species but opposite to what we expected for forest species. The reason for these surprising results may be that (1) the species we collected are not true forest interior species, (2) urban woodland edges are optimal habitats for many forest carabids, or (3) edges are actually suboptimal, and high catches simply reflect increased activity of beetles moving away from the edge. Trampling did not have an overall negative effect on carabids as hypothesized. Species associated with moist forest habitat responded as predicted: they decreased in abundance with increasing path cover. Furthermore, open habitat species decreased with increasing path cover but more straightforward than we had predicted. Model elaboration, by dropping the highly trampled sites from the analyses, suggested that our data of high trampling may be too scarce: the results without these sites were more in accordance with our predictions than with the full dataset.