Regional and Local Influences on Species Richness in Coral Communities Across a Pacific Diversity Gradient

University Of Delaware, Newark DE

Investigators

Abstract

Species richness in coral communities has traditionally been explained in terms of processes operating within the local community over relatively short periods of time. Recent studies of other assemblages have made it clear that communities can be organized by a variety of processes operating at different spatial and temporal scales. Since coral communities are imbedded in larger geographic regions, regional/historical phenomena and local processes may jointly influence local richness and should be analyzed simultaneously. This study will sample corals in six island groups along a gradient of increasing regional species richness across the Pacific Ocean and in three different habitats at local sites to quantitatively evaluate the relative influence of regional and local processes on local richness. The Pacific Ocean gradient is the best possible choice for this test because of the large unequivocal trend in regional richness. Traveling from west to east, regional richness drops from over 450 species in the Philippines and Indonesia to fewer than 100 species in eastern French Polynesia. A hierarchical sampling design will be employed to measure variation in species richness, composition, and relative abundance at four spatial scales spanning 5-6 orders of magnitude (among replicate samples at a site, among sites on an island, among islands in a region, among regions). Data will be collected on species presence and relative abundance as well as on habitat breadth and depth range of selected species. This information will be used to test the effects of the increasing regional pool on local richness, evenness, habitat compression and habitat overlap at different scales. Percent cover of colonizable substrate will also be quantified and used to test the effects of the "openness" of local assemblages on their sensitivity to regional influences. The hierarchical design will provide a powerful framework for quantifying the proportion of total variation among samples that is attributable to each spatial scale and thus will permit a rigorous estimate of scale-dependency in patterns of diversity and abundance. This synthesis of regional and local perspectives will have relevance not only for basic studies of community ecology but also for applied problems associated with species extinctions, habitat loss and fragmentation, and ecological restoration.