PARENT SESSION
Oral Session #4: Animal Population Ecology.
Presiding: L. Mitchell
Monday, August 5. 8:00 AM to 11:30 AM. Gila Meeting Room, TCC.
Does elasticity analysis predict population response? Results from an experimental test.
Benton, Tim*,1, Cameron, Tom1, 1 Institute of Biological Sciences, Stirling, UK
ABSTRACT- Life histories of organisms can be described by a matrix model. Assuming that matrix does not vary with time allows very straightforward analyses to be conducted. The most commonly used analyses in conservation (and management) are elasticity or sensitivity analysis, which estimates the way that population growth rate, 
, varies with the matrix entries. Under certain conditions, the elasticities of also predict the change in equilibrium size for organisms subject to density dependence. The predictions of elasticity analysis, however, have never been experimentally tested. Here we report the results of replicated laboratory experiments on populations of a soil mite, Sancassania berlesei. For replicated populations, living in constant, very variable or low density environments, we experimentally perturbed vital rates, examined the population-level responses, and compared these to elasticities estimated from two different matrix models. The elasticities of estimated from a simulation model predict the population size effects well under constant conditions, but do less well with the very variable conditions (where the elasticities of the invasion exponent, 
, do better). A matrix model derived using quadratic programming from population time series gave elasticitites of that did not compare well with the empirically-derived elasticities. The empirically-derived elasticitites of population growth rate were all negative, suggesting density-dependence was still important even though population size was well below carrying capacity. These results suggest that the elasticities of can predict population-level responses to perturbations in the vital rates, but this can depend on the variability of the population, the strength of density dependence and the precision with which the projection matrix is estimated.
KEY WORDS: elasticity, population, sensitivity, density