Abstract

Classic results of matrix models predict that, in species with a long generation time, adult survival should be the demographic parameter driving population dynamics whereas, in species with a short generation time, adult survival should not be of such importance. Nonetheless, Yoccoz et al. (1998, Research Population Ecology 40: 107-121) hypothetized that, in small rodents, adult survival should be the demographic parameter driving the population growth rate if one considered a time scale of one month (instead of one year). As far as we know, this hypothesis has not yet been tested with empirical data. To test this hypothesis, we used five years of capture-mark-recapture data to estimate maturation and survival of a rodent, the leaf-eared mice, Phyllotis darwini, in a population of central Chile. This analysis revealed that the probabilities of survival decreased with the average rainfall by year and that the probabilities of maturation decreased with the abundance of the population. Using the probabilities of survival and maturation, we built up a seasonal matrix model and use perturbation analysis (elasticity) to be able to actually measure the relative importance of each demographic parameter to the population growth rate. Environmental seasonality was incorporated in the model by using a rainfall season, of five months long, and a dry season. Adult survival was indeed the demographic parameter with the highest elasticity. Such a result plainly supported thus the hypothesis of Yoccoz et al. (1998). © 2006 Sociedad de Biología de Chile.