Effects of Selective Logging on Tropical Forest Butterflies on Buru, Indonesia

Abstract

1. The butterfly fauna of lowland monsoon forest on Buru, Indonesia was compared in unlogged forest and forest that had been selectively logged 5 years previously. 2. Seven variables relating to vegetation structure were measured in each habitat. Tree density and percentage cover of vegetation in the canopy and understorey were significantly higher, and vegetation cover 2 m above the ground was significantly lower, in unlogged forest. There were no differences between sites in the mean heights or girths of trees, but the ranges of both heights and girths were lower in logged forest. Percentage cover of vegetation at ground level was similar at the two sites. 3. Species richness, abundance and evenness of butterflies and an index of taxonomic distinctiveness were all significantly higher in unlogged forest. Two endemic species and a further four species with distributions restricted to Maluku Province were recorded only in unlogged forest. 4. Species abundance data for butterflies at both sites fitted a log-series distribution. Data for unlogged forest also fitted a log-normal distribution, whereas those for logged forest did not. This indicated the presence of a more complex butterfly community in unlogged forest. 5. These results indicate that the distributional pattern of species abundance of tropical butterflies may be used as an indicator of forest disturbance, and that selective logging of tropical forests in SE Asia may be associated with a significant decrease in biodiversity of butterflies, at least during the first 5 years of forest regeneration.