Integration of local and regional perspectives on the species richness of coral assemblages

Integration of Local and Regional Perspectives on the Species Richness of Coral Assemblages1

SYNOPSIS. We have evaluated the relationship between regional species richness and the number of species occurring within local, quantitatively sampled assemblages of scleractinian corals. Our data have been extracted from the published literature describing richness patterns from over 100 locations around the world. In general, we find a positive relationship between local and regional richness. Local richness is not independent of regional richness as posited by conventional theory and there is no hard upper limit indicating saturation. Instead, local coral assemblages are regionally enriched. This result suggests that these assemblages are open to regional sources of species. The degree of regional enrichment is geographically variable. In the Indo-Pacific, assemblages in speciose regions appear to be less open and much more sensitive to local depth and habitat gradients than those in more depauperate regions. Other large-scale geographical and historical effects on local richness in the Indo-Pacific include the degree of isolation from high-diversity regions and distance from the equator. In contrast, local richness in the relatively homogeneous and depauperate western Atlantic is insensitive to the large-scale variables we examined. As in most ecological communities, membership in local assemblages of corals is not absolutely limited (by biotic interactions or local environmental factors) nor is it totally open to regional pools of species. Understanding the dynamics of coral communities will require integrating the local ecological perspective with large-scale phenomena (ie., physical TECO processes [Myers, 1994] and evolutionary history [Hugueny et al.1997]). Such an integration will necessarily encompass multiple spatial and temporal scales.

INTRODUCTION

In recent years, community ecologists have begun to recognize the influence of large-scale phenomena on the structure of local communities (e.g., Ricklefs and Schluter, 1993; Giller et al., 1994). These comprise a wide variety of historical and geographical factors including such things as regional climate, geological disturbances, unique speciation and extinction events, oceanographic transport processes, atmospheric-oceanographic coupling, and even orbital forcing of global climate. Traditional notions regarding the structure of ecological communities invoked local resource competition and niche diversification as central paradigms. We are now faced with the challenge of integrating the effects of these local factors with much larger scale processes in order to understand community structure and dynamics.

Expectations based on conventional theory include the prediction from MacArthur (1965) that the high diversity exhibited by many taxa in the tropics is due to habitat specialization (i.e., there are relatively large differences in species composition among habitats). Within habitats, species membership was thought to be restricted primarily because of strong biological interactions, habitat stratification, and partitioning of food resources. Consequently, MacArthur (1965) predicted that tropical habitats are saturated with species.

The notion that habitats are saturated dates back at least to Elton (1933). Based on animal surveys in similar habitats in quite different places, he noted that the number of species co-occurring in one place was far below the number occurring in the general area. Thus he suggested that the species composition (i.e., membership) within local assemblages is limited to relatively few species rather than being open to all. Conventional theory predicts that limited membership is a consequence of biological interactions (competition and predation) and physical environmental factors characteristic of the local habitat. This theory also predicts the following: 1) "local diversity should be correlated with features of the environment, especially the diversity of resources," 2) "independently assembled communities in similar habitats on different continents (or in different oceans) should contain similar numbers of species" (i.e., convergence), and 3) local "diversity within any habitat should reach a ceiling that is independent of the number of species in the regional pool" (Schluter and Ricklefs, 1993a). This last prediction describes the quantitative relationship between local and regional species richness and provides the focus for our work on coral species richness.

Species richness is the component of species diversity referring only to the number of species. At very large spatiotemporal scales, species richness describes biogeographical assemblages and is indicative of the influence of evolutionary phenomena and large-scale variation in the physical environment. We define regional richness for corals as the number of species measured over large geographical scales (102-104 km). At much smaller spatiotemporal scales, species richness describes ecological assemblages and is indicative of the effects of biological interactions and the local physical environment. We define local richness for corals at small spatial scales (10o103 m). At these scales, samples provide estimates of the within-habitat component of species diversity. Although local richness is sampled at one point in time, the recent history of the assemblage over the life span of the resident organisms is reflected in such samples. As recognized by Done (1999) and many others, local richness in ecological assemblages is highly dynamic. By examining the relationship between local and regional richness, we explore potential cross-scale linkages between ecological and biogeographical phenomena.