Pool-breeding amphibian populations operate at multiple scales, from the individual pool to surrounding upland habitat to clusters of pools. When metapopulation dynamics play a role in long-term viability, conservation efforts limited to the protection of individual pools or even pools with associated upland habitat may be ineffective over the long term if connectivity among pools is not maintained. Connectivity becomes especially important and difficult to assess in regions where suburban sprawl is rapidly increasing land development, road density, and traffic rates. We developed a model of connectivity among vernal pools for the four ambystomatid salamanders that occur in Massachusetts and applied it to the nearly 30,000 potential ephemeral wetlands across the state. The model was based on a modification of the kernel estimator (a density estimator commonly used in home range studies) that takes landscape resistance into account. The model was parameterized with empirical migration distances for spotted salamanders (Ambystoma maculatum), dispersal distances for marbled salamanders (A. opacum), and expert-derived estimates of landscape resistance. The model ranked vernal pools in Massachusetts by local, neighborhood, and regional connectivity and by an integrated measure of connectivity, both statewide and within ecoregions. The most functionally connected pool complexes occurred in southeastern and northeastern Massachusetts, areas with rapidly increasing suburban development. In a sensitivity analysis estimates of pool connectivity were relatively insensitive to uncertainty in parameter estimates, especially at the local and neighborhood scales. Our connectivity model could be used to prioritize conservation efforts for vernal-pool amphibian populations at broader scales than traditional pool-based approaches.

No abstract

Models of habitat selection have been developed primarily for mobile animals with well-defined home ranges. The assumptions made by traditional techniques about habitat availability are inappropriate for species with low mobility and large home ranges, such as the wood turtle. We used paired logistic regression, typically used in medical case Ϫ control studies, to model selection of habitat within activity areas in a population of wood turtles in a watershed in western Maine. We also modeled selection of activity areas within the watershed, using nonpaired logistic regression. Within activity areas, wood turtles selected nonforested locations close to water with low canopy cover. Within the watershed, they selected activity areas close to streams and rivers with moderate forest cover and little open water. The difference between selection at these two scales suggests that wood turtles select forest edges to balance thermoregulatory and feeding needs. The model of selection of activity areas within the watershed correctly classified 84% of activity areas and random areas. This model may be useful for identifying wood turtle habitat across the landscape as part of regional conservation efforts. We suggest that paired logistic regression shows promise for analysis of habitat selection of species with movement patterns that violate the assumptions of traditional habitat selection models.

Recent studies suggest that freshwater turtle populations are becoming increasingly male-biased. A hypothesized cause is a greater vulnerability of female turtles to road mortality. We evaluated this hypothesis by comparing sex ratios from published and unpublished population surveys of turtles conducted on-versus offroads. Among 38 166 turtles from 157 studies reporting sex ratios, we found a consistently larger female fraction in samples from on-roads (61%) than off-roads (41%). We conclude that female turtles are indeed more likely to cross roadways than are males, which may explain recently reported skewed sex ratios near roadways and signify eventual population declines as females are differentially eliminated.

Models of habitat selection have been developed primarily for mobile animals with well-defined home ranges. The assumptions made by traditional techniques about habitat availability are inappropriate for species with low mobility and large home ranges, such as the wood turtle. We used paired logistic regression, typically used in medical case Ϫ control studies, to model selection of habitat within activity areas in a population of wood turtles in a watershed in western Maine. We also modeled selection of activity areas within the watershed, using nonpaired logistic regression. Within activity areas, wood turtles selected nonforested locations close to water with low canopy cover. Within the watershed, they selected activity areas close to streams and rivers with moderate forest cover and little open water. The difference between selection at these two scales suggests that wood turtles select forest edges to balance thermoregulatory and feeding needs. The model of selection of activity areas within the watershed correctly classified 84% of activity areas and random areas. This model may be useful for identifying wood turtle habitat across the landscape as part of regional conservation efforts. We suggest that paired logistic regression shows promise for analysis of habitat selection of species with movement patterns that violate the assumptions of traditional habitat selection models.