Abstract
Deicing salts are causing freshwater wetlands to become increasingly saline near roadways, with cascading impacts on local ecosystems. Understanding the full reach of salt into surrounding landscapes can inform the decisions made every winter about roadway management. We measured conductivity and surveyed for vernal-pool obligate species at 541 wetlands identified as potential vernal pools in western Massachusetts, USA, estimating that the salt effect zone extends as far as 167 m to 251 m from roadways. For the smallest wetlands with perimeters under 100 m, the salt effect zone extends to between 81 and 128 m. The mean conductivity of wetlands beyond 251 m was 91 μS/cm (SD = 109 μS/cm), whereas mean conductivity was 168 μS/cm (SD = 180 μS/cm) between 167 m and 251 m, and 274 μS/cm (SD = 340 μS/cm) at wetlands within 167 m of roads. Occupancy and N-Mixture models found that the threefold higher conductivities in average wetlands within the salt effect zone would cause 14% (SE = 5%) lower predicted rates of site occupancy for spotted salamander (Ambystoma maculatum), 15% (SE = 5%) lower occupancy for wood frogs (Lithobates sylvatica), 29% (SE = 11%) lower population sizes for spotted salamanders and 19% (SE = 14%) lower population sizes for wood frogs, although the wood frog abundance model did not meet the threshold for statistical significance. Compared to average wetlands, the mean conductivity was lower in wetlands with marbled salamanders (Ambystoma opacum) and fairy shrimp (Eubranchipus sp.) and approximately the same for Jefferson salamanders (Ambystoma jeffersonianum and associated unisexual Ambystoma), but data for these species were insufficient for formal occupancy modeling. We estimate that 78% of all vernal pools in Massachusetts fall within the road salt effect zone, underscoring the importance of integrating decision-making surrounding roadways and conservation.