EFFECTS OF GRASSLAND FRAGMENTATION ON INSECT SPECIES LOSS, COLONIZATION, AND MOVEMENT PATTERNS

Abstract

Ecological theory predicts that isolated habitat patches will experience greater rates of species loss and lower rates of recolonization compared to less isolated habitats. As a corollary, corridors of suitable habitat should reduce patch isolation, thereby decreasing species loss and enhancing colonization. Little ecological evidence exists, however, to compare species loss and colonization in habitat patches with and without corridors under field conditions. Using insects in a native grassland habitat, I performed a three-year field experiment that varied fragment size and connectivity to test the hypotheses that corridors influence patterns of insect species loss, rates of recolonization, and insect movement among habitat fragments. I detected few effects of corridors consistent with theoretical predictions. First, corridors generally failed to reduce insect species loss from otherwise isolated grassland patches, except for subtle effects in medium-sized patches. Species richness was slightly higher in medium patches with corridors than in medium isolated patches when averaged over the three years of the study. I observed the strongest positive effect of corridors in the third and driest year of the study. Second, corridors did not significantly influence overall rate of patch colonization, but slightly increased the probability of colonization by less vagile species. Interestingly, corridors did not affect recolonization by rare species, which are often the subject of conservation concern. Third, I characterized individual movement pathways of three insect species in plots with and without corridors. One of the three species significantly preferred corridors, while the other two species moved independently of corridors. Taken together, these results suggest that corridors have the potential to promote movement of organisms among habitat patches, but that their function may depend upon species characteristics, landscape context, patch size, and environmental variation.