A framework for assessing the relative vulnerability of species to land-use change

Conversion of native habitat to human-dominated uses is the main driver of global biodiversity loss, yet which species will be most impacted, and why, remain poorly known. There is thus an urgent need to develop frameworks for understanding, and predicting, the effect of habitat alteration on biodiversity. We develop an approach with three components: a demographic model, a regional database of life-history traits, and a sensitivity analysis of the model predictions. We use a spatially explicit model that predicts the fate of individual species in a human-dominated landscape. The model takes as parameters habitat affinity, population growth rate, annual dispersal, and dispersal behavior at the habitat edges. The model predicts the minimum area of native habitat that allows for persistence of a species. We apply the model to a regional community of species, the avifauna of Costa Rica. We gather life-history data (body mass, clutch size, breeding season length, number of broods per year, age at first breeding, life span, and dispersal distance) for Costa Rican birds. When data are not available for Costa Rican species, inferences are made from North American and European birds. We use these data to estimate the model parameters for each species. Minimum patch sizes predicted by the model are used to specify the relative degree of threat faced by each species. We perform a sensitivity analysis of patch size and relative vulnerability predictions to model assumptions and gaps in the data. Our predictions of relative vulnerability are robust to changes in model assumptions and agree with an independently derived empirical assessment. Our framework thus appears to be useful for understanding, and influencing, the fates of neotropical birds and possibly other taxa worldwide.