Do molt-migrant songbirds optimize migration routes based on primary productivity?

Abstract

Migratory animals often go to extraordinary lengths to follow broad-scale changes in the spatial distributions of food resources over time. An example is the molt-migration system of songbirds in the central and western United States, wherein birds interrupt migratory movements to molt their feathers in the ephemerally productive North American monsoon region in Northwestern Mexico. Within this large-scale migration system, the efficacy of individual migration behaviors relative to dynamics of primary productivity en route is unknown. We performed a comparison of real migration tracks from a common molt migrant, the painted bunting (Passerina ciris), with the results from millions of random-walk simulations to determine if real birds demonstrate optimal migratory behavior with respect to seasonal changes in primary productivity across the landscape. We scored migration routes of both real and simulated birds using an index of new plant growth, wherein daily values for each day of the autumn migration period were summed to indicate the total exposure to primary productivity. We then compared the highest-scoring simulated routes with real bird tracks and observed a high degree of correspondence between them, with the monsoon region featuring prominently in most of the high-ranking simulations. However, real birds generally left the breeding grounds later than virtual birds, perhaps because of life-history constraints not represented in the simulations. Our work supports the contention that painted bunting migration maximizes exposure to new plant growth and provides a novel approach for exploring migratory behaviors in other species and other regions.