THE INFLUENCE OF MODEL STRUCTURE ON CONCLUSIONS ABOUT THE VIABILITY AND HARVESTING OF SERENGETI WILDEBEEST

We investigate how the viability and harvestability predicted by popul ation models are affected by details of model construction. Based on t his analysis we discuss some of the pitfalls associated with the use o f classical statistical techniques for resolving the uncertainties ass ociated with modeling population dynamics. The management of the Seren geti wildebeest (Connochaetes taurinus) is used as a case study. We fi tted a collection of age-structured and unstructured models to a commo n set of available data and compared model predictions in terms of wil debeest viability and harvest. Models that depicted demographic proces ses in striking different ways fitted the data equally well. However, upon further analysis it became a clear that models that fit the data equally well could nonetheless have very different management implicat ions. In general, model structure had a much larger effect on viabilit y analysis (e.g., time to collapse) than on optimal harvest analysis ( e.g., harvest rate that maximizes harvest). Some modeling decision, su ch as including age-dependent fertility rates, did not affect manageme nt predictions, but others had a strong effect (e.g., choice of model structure). Because several suitable models of comparable complexity f itted the data equally well, traditional model selection methods based on the parsimony principle were not practical for judging the value o f alternative models. Our results stress the need to implement analyti cal framework for population management that explicitly consider the u ncertainty about the behavior of natural systems.