Bayesian Survival Trajectory Analysis

Recent studies show that there is an outstanding variety of age-specific trajectories of mortality and fertility in nature. Neglecting these age-specific trends can result in biased estimates of measures of population performance such as population growth rates (Colchero et al. in progress). However, obtaining age information from wild animals can be extremely challenging. To overcome these limitations, we have developed a hierarchical Bayesian model to estimate not only survival parameters but also the unknown individual times of birth and death (Colchero & Clark 2012).

To make these methods widely available, we have put together an R package called BaSTA (Bayesian Survival Trajectory Analysis) which allows users to perform age-specific survival analysis from capture-recapture/recovery data when age information is scarce or even absent (see video; Colchero et al. 2012). For further information on the package, please visit the BaSTA website.

We are applying these methods to understand patterns of mortality in a number of species for which the estimation of age-specific vital rates is extremely challenging. For instance, Francisco Villavicencio extended these methods to reconstruct mortality from historical parish wedding records in Barcelona in the XVI and XVII centuries, in collaboration with Prof. Ana Cabré from the Centre d'Estudis Demografics in Barcelona. We have also expanded the BaSTA framework to model mortality on organisms that either one or both sexes disperse out of the study area (Barthold et al. 2016a, 2016b, Colchero et al. 2016).