A variable edge effect on trees of Bwindi Impenetrable National Park, Uganda, and its bearing on measurement parameters

Abstract

With tropical forests under increased pressure from fragmentation and nascent effects of climate change, it is imperative to reliably predict change for timely reduction of impacts. This depends on measuring the right variables and analyzing them in ways that best detect change. I compared the usefulness of density, species richness, species diversity, and size in detecting edge penetration; and other parameters in revealing edge effect on tree species in Bwindi Impenetrable National Park, Uganda. Trees were sampled in 104 1 km edge-interior transects set around the entire park. Density and species richness were better indicators of edge-width than girth size and diversity measurements, both showing edge penetration as most intense within 300 m, although effect was detectable up to 900– 1000 m. This estimate compared closely with an earlier estimate of 300–350 m obtained using incidences of resource harvest. Among other measures, measures of tree species abundance and frequency showed significant response to the edge by both early succession and primary forest species; while analyses of density showed edge response primarily by forest interior species. Parameters of tree size and their derivatives showed edge response by the lowest number of species. These results suggest that to measure edge effects reliably, it is important to select the right parameters.