Comparison of direct and indirect methods of estimating migration and dispersal routes of the marbled salamander, Ambystoma opacum

Monitoring migration among natural populations is important because migration is a crucial component contributing to the viability of populations, particularly for rare or endangered species. Accurate estimates of migration are necessary to implement conservation management plans; however, there is still little consensus on the best method to estimate migration. Given that such measures of connectivity and dispersal are essential for assessing the status of populations, accuracy should be at the forefront of migration studies. The goal of this study was to determine the most accurate method for estimating migration as well as dispersal routes.;Direct and indirect methods of estimating migration have been separately implemented in migration studies; however, both types of methods pose limitations. While the direct capture-mark-recapture method can better estimate migration, it is time consuming and difficult to implement for many species. On the other hand, while implementing indirect methods of estimating migration using genetic data and a variety of programs is relatively simple and easy, they may not always be an accurate indicator of migration. This study was aimed at determining the best method to monitor migration within a metapopulation of marbled salamanders, Ambystoma opacum. Comparing direct and indirect estimates of migration revealed that the measures are in fact not congruent. We conclude that direct measures are more accurate estimates of the movement of individuals, while indirect measures more accurately reflect gene flow.;In addition, migration estimates in conjunction with least-cost path analysis via Geographical Information Systems are increasingly employed to determine dispersal routes. Given that conservation management plans may refer to least-cost path dispersal maps to implement dispersal corridors, determining which data will yield the most biologically plausible habitat costs and therefore dispersal routes is crucial for implementing effective corridors. We compared dispersal maps of marbled salamanders, Ambystoma opacum, constructed on two sets of habitat costs derived from (1) field estimates of migration and (2) genetic estimates of migration. We show that the two sets of habitat costs predict different dispersal maps and that field data likely represent biologically plausible habitat costs and therefore dispersal routes.