Wednesday, August 4, 2010: 10:10 AM
335, David L Lawrence Convention Center
Benjamin G. Van Allen, Ecology and Evolutionary Biology, Rice University, Houston, TX, Chris M. Asquith, Biology, Virginia Commonwealth University and Volker H.W. Rudolf, Department of Ecology & Evolutionary Biology, Rice University, Houston, TX
Background/Question/Methods Anthropogenic habitat degradation, resource exploitation and climate change have led to unparalleled rates of population decline and species extinctions. Yet, there is considerable variation in what species become rare. Current approaches in conservation often require large amounts of data that are difficult to obtain for the many species potentially at risk. Identifying widely applicable factors useful in determining which species are likely to decline is therefore critical for conservation strategies and management. We present a method for using existing data from life table level information to generate a single numerical predictor of relative decline under stochastic environmental conditions. To generate our predictor we examine the distribution of elasticity values for evenness in a stage based population projection matrix for. This value is hypothesized to correlate with species decline in the face of multi-stage stochasticity in survival and reproduction rates. To further test our hypothesis against real data we compare evenness of elasticity values for least concern and threatened species on the IUCN redlist.Results/Conclusions
Analytical tests of our hypothesis show that evenness of elasticities significantly predicts decline under simulated stochastic conditions. We then test model predictions by analyzing populations of threatened and least concern species for our predictor. We find that our model predictions are significantly correlated with conservation status for our known populations. Together, simulations and correlative tests suggest that populations of species which spread their reproductive potential evenly through their life history will be more resilient to stochastic variation in survival and fecundity. This simple metric indicates that general knowledge of life history traits can be used to help predict the severity of conservation threats and to prioritize conservation research when little is known of many species.
