Small-scale effect of habitat heterogeneity on invertebrate assemblages in sandy grasslands (Hungarian Great Plain)

Numerous environmental factors are confirmed to have significant influence on the habitat choice of invertebrates and thus on the assemblage structure. In dry, sandy grasslands the vegetation cover and height are assumed to be the most important factors in shaping the distribution of spiders and true bugs. The present study was carried out at a natural sand dune area in the Kiskunsag region of the Hungarian Great Plain. Two adjacent sand dunes and the dune valley between them were sampled using a transect consisting of pitfall traps. The traps were arranged in 4 parallel transects, running from the sand dune top through the dune valley to the adjacent sand dune. Each row consisted of 40 traps, three meters apart. The effect of microhabitat parameters on the species richness and abundance of invertebrate assemblages were tested with linear regressions with forward selection procedure. A total number of 1447 spider and 1580 true-bug individuals of 58 and 55 species were collected, respectively. The mean number of spider species along the transects was 10.5 [plus or minus] 3.7 and 9.8 [plus or minus] 3.0 for true bugs. Although our data did not show a significant effect of the plant species number on species richness and abundance of the two studied taxa along this gradient, the results of the canonical correspondence analysis and the Mantel test emphasized the importance of the total coverage and vegetation composition on the distribution of invertebrate species. The correspondence analysis and the multivariate ANOVA revealed different spider and true-bug assemblages on the two slopes (multivariate ANOVA: Araneae: F = 3.609, P <0.001, Heteroptera: F = 5.248, P <0.001), possibly due to the more dense and diverse vegetation on the north facing slope, which is presumably brought about by the different insolation and moisture conditions of the slopes.