On the Community Composition and Abundance of Delaware Forest Birds

Forest birds differentially respond to deforestation and fragmentation: some species are absent from small patches, some species increase in abundance in isolated patches, while others are not present in patches that appear to be of sufficient quality and area to meet their ecological requirements. Mechanisms currently associated with forest fragmentation, such as increased nest predation and nest parasitism, are insufficient to predict how populations of a given species, or individuals, respond to deforestation. Based on published data representing 573 forest patches, I examined the effects of deforestation on the abundance of forest birds and tested the hypothesis that species exhibit differential responses to deforestation based on their mean body mass. Very small, small, and large species were present in reduced numbers, or not at all, in small, isolated patches while medium and intermediate species often increased in abundance in these forests. Cavity-nesting species appear less affected than cup-nesting species of similar size. I found that forest bird communities can be generally predicted based on the home range needs and the average body mass of individual bird species; mass may be an indicator of competitive abilities. A natural experiment was conducted in the State of Delaware to verify and refine the above patterns detected in the literature. The effects of local forest cover, patch area, and patch isolation on the abundance of five mass classes of forest birds were examined with regression analyses based on data collected from 296 forest patches. Very small (e.g., most warblers), small (e.g., Ovenbird, Red-eyed Vireo) and large (e.g., American Crow) birds increased in abundance (number detected per survey) with increasing local forest cover, patch area, and decreasing patch isolation. Intermediate birds (e.g., Blue Jay) decreased in abundance as local forest cover and patch area increased, and increased in abundance as isolation increased. The abundance of medium (e.g., Wood Thrush) species decreased with increasing local forest cover and was not influenced by patch isolation. Abundance of medium species initially increased then decreased with increasing patch area. All of these patterns can be explained if forest resources are limiting and outcomes of interspecific competition, generally determined by body mass, either allow or restrict access to these resources. Body mass may be sufficient to predict, within bounds, avian community composition and the abundance of each species within a forest patch. (Abstract shortened by UMI.).