Effects of landscape composition and configuration on migrating songbirds: inference from an individual-based model

Abstract

The behavior of long-distance migrants during stopover is constrained by the need to quickly and safely replenish energetic reserves. Replenishing fuel stores at stopover sites requires adjusting to unfamiliar landscapes with little to no information about the distribution of resources. Despite their critical importance to the success of songbird migration, the effects of landscape composition and configuration on fuel deposition rates (FDR g/d]), the currency of migration, has not been tested empirically. Our objectives were to understand the effects of heterogeneous landscapes on FDR of forest dwelling songbirds during spring migration. The results of field experiments were used to parameterize a spatially explicit individual-based model of forest songbird movement and resulting FDR. Further field experiments were used to validate the results from the individual based model. In simulation experiments, we altered a Gulf South landscape in a factorial design to predict the effects of future patterns under different scenarios of land use change in which the abundance of high-quality hardwood habitat and the spatial aggregation of habitat varied. Simulated FDR decreased as the amount of hardwood in the landscape decreased from 41 to 22 to 12 percent. Further, migrants that arrived in higher quality habitat types gained more mass. Counter to our expectations, FDR was higher with lower spatial aggregation of habitat. Differences in refueling rates may be most influenced by whether or not an individual experiences an initial searching cost after landing in poor quality habitat. Therefore, quickly locating habitat with sufficient food resources at each stopover may be the most important factor determining a successful migration. Our findings provide empirical evidence for the argument that hardwood forest cover is a primary determinant of the quality of a stopover site in this region. This study represents the first effort to empirically quantify FDRs based on the configuration of landscapes.