Numerous species face redistribution and compression of habitat due to climate change. We coupled long-term movement and environmental data to assess how a freshwater species responds to changes in a coastal refuge habitat using Boosted Regression Trees (BRT) to predict distributional changes in the coming decades. Salinity, variation in salinity, and stage of the surrounding marsh habitat were the most important variables in BRT model, accounting for over half (56.6%) of the tree splits informing the final model. Interestingly, the habitat classified as conditional experienced the most variability (5.85 ± 6.2 km2), while core habitat remained relatively consistent (1.29 ± 0.98 km2) across years with varying hydrological conditions. These results suggest that varying environmental scenarios can drastically shift the amount of suitable habitat available for freshwater species using high salinity, conditional habitats at the coast. Climate change will likely result in large-scale reductions of critical dry season habitat for these species; while restoration efforts and adaptive management can bolster the resiliency of these habitats to ensure population persistence.
