A simulation model of Mexican long-nosed bat ( Leptonycterisnivalis) migration

Abstract

We (1) describe a model that simulates migration of the Mexican long-nosed bat ( Leptonycteris nivalis) based on the flowering phenology (nectar–pollen production) of agaves (Agavaceae) and hypothesized ‘rules’ governing bat movements; (2) evaluate the model by comparing simulated seasonal and spatial patterns of nectar production and consumption, and bat movements and densities, to patterns observed in the field; and (3) use the model to examine various hypotheses concerning factors that control migration. A nectar production sub-model represents the flowering phenology of agaves in terms of the daily availability of nectar within each of four latitudinal intervals between 18 and 29°N. A bat migration sub-model represents the number of bats present within each latitudinal interval each day, with bat movements from one site to another depending on availability of nectar and season. Simulated patterns of nectar consumption are similar to observed patterns of nectar production based on the number of flowering plant species present at different latitudes. Simulated patterns of bat movements are similar to general patterns observed in the field for Leptonycteris curasoae and L. nivalis. Simulated fluctuations of bat density at the southernmost latitude exhibit the same general annual cycle observed at a southern roost of L. curasoae. The seasonal representation of nectar production in the model corresponds well with patterns of nectar production observed over a 2-year period at a site in northern Mexico, although nectar production in the model begins somewhat earlier. Simulations examining factors hypothesized to control migration indicate that predicted migration patterns correspond well with field observations only when model rules assume that both food availability and season limit migration, and that all bats with access to sufficient energy during the correct season migrate.