9627766 Cohen This project will examine the patterns and timing of benthic paleocommunity change in Lake Tanganyika, a large and ancient aquatic ecosystem. Prior investigations in the lake have yielded conflicting views of the nature of ecological interactions in the lake, whether they are tightly linked and coevolved, or whether they are much looser assortments of taxa, formed either through the individualistic habitat requirements of components species or even more randomly by recruitment from patchy metapopulations. This disagreement is a microcosm of the broader and long-standing argument among ecologists of what a community is and how it is assembled. The dispute is not merely academic, since the design of underwater reserves in a lake implicity requires assumptions about the permanence or ephemerality of communities and their component species' populations. Given its intrinsic interest as a major reservoir of biodiversity (>1400 spp., ~600 endemic), Lake Tanganyika provides an excellent site to address this problem using paleoecolgical techniques. Preliminary studies have demonstrated that dateable cores with abundant and diverse faunas can be obtained from the nearshore sediments of this lake, with stratigraphic resolution on the decadal (and perhaps even annual) level. Cores taken at strategic localities in the lake will be taken to create an interface between the time-scales investigated by Long Term Ecological Research Studies (decadal) and the time spans to which marine paleoecologist are normally constrained by time-averaging taphonomic processes (millenia). These cores can be used to answer questions of interest to both conservation biologist and theoretical paleoecologists. The primary goal of this project will be to document patterns of local extinction and recolonization of habitat patches by individual benthic species (crustaceans, molluscs and sponges) and to determine the degree to which the various species histories at each site are interrelated. This will be ach ieved by coring portions of the lake where high resolution paleoecological records can be obtained, obtaining a closely-spaced series of radiometric dates (210 PB and AMS 14C) to provide an accurate chronology of the cores, and then sampling the fossil fauna at ~decadal scale intervals. A series of cores at both distrurbed (river delta) and undisturbed (platform environments away from major rivers) localities will be taken along the Tanzanian coast of the lake along depth transects (25,50 and 75m) and analyzed quantitatively for fossils benthos content and various paleoenvironmental indicators (sedimentology, stable isotope history and diatom paleoecology) to address the following questions: -Do species groups colonize and dissappear from localities in synchrony or individualistically? -Are appearances and disappearances of species correlated with changes in overall diversity and/or environmental shifts? -Do particular clusters of taxa replace each other consistently in biotic overturn, in the absence of paleoenvironmental change? -Does the paleobiologic record in Lake Tanganyika better support a model of extreme niche specialization and species integration (little relative biotic overturn, under relatively constant environmental conditions or one of continuous non-equilibrium reassortment of species assemblages? -At what rate do local extinction and recolonization events naturally occur and how "permanent" or temporary are populations of local endemics? -What are the implications of the observed presence or absence of community cohesion for the establishment of underwater reserves in adjacent parts of the lake? Community paleoecologist and conservation biologists share a host of questions about community structure and stabilty, although explicit linkages between the two fields have rarely been articulated. The research proposed here has the potential to demonstrate ways in which these linkages can be made.
