COMPARATIVE BREEDING BIOLOGY OF THE OVENBIRD IN LARGE VS. FRAGMENTED FORESTS: IMPLICATIONS FOR THE CONSERVATION OF NEOTROPICAL MIGRANT BIRDS (LANDSCAPE ECOLOGY, AVES: PARULINAE, NEW JERSEY, EASTERN DECIDUOUS)

The purpose of this study was to investigate whether recent, well-documented declines of neotropical migrant birds in patches of Eastern Deciduous Forest are related to factors internal to patches (e.g., increased adverse biotic interactions), to behavioral responses of the birds (e.g., edge avoidance), or to external causes (e.g., increased wintering-ground mortality). To accomplish this the breeding biology of Ovenbirds (Seiurus aurocapillus) on a 20-ha plot within an extensive forest was compared to that in 14 forest patches ranging in size from 1.7 to 37.2 ha in central New Jersey. Presence of other neotropical migrant species was monitored in both large forest and forest patches, and data from several other sources were also analyzed. Ovenbirds in forest patches exhibited a pattern of later arrival dates, lower density of territorial males and breeding pairs, and lower percentage of mated males compared to those in the large forest. This species' utilization of isolated and of small (<10 ha) forest patches was much lower than that of a closely grouped "archipelago" of patches, or of patches larger than 10 ha. However, compared to pairs in the large forest, Ovenbirds breeding in forest patches experienced no difference in levels of predation or brood parasitism or in reproductive success of breeding pairs. Edge avoidance behavior did not appear to be a significant cause of this species' low utilization of forest patches. Potential colonists of virtually all neotropical migrant species occurred in even very small and isolated forest patches, although nearly half of these species did not breed in any patch. That these species did breed in long-isolated forest patches as recently as the 1960s indicates that forest patches larger than 10 ha can provide appropriate habitat for most forest-interior species. Comparison of current and former breeding densities of neotropical migrants in extensive forests reveals that serious population declines have occurred in large forests as well as in patches. A model is suggested to explain the apparent differential rate of decline of neotropical migrants in large vs. small forests. The observed population declines are more probably related to loss of wintering-ground habitat rather than to fragmentation of breeding habitat.