Regional analysis of the effect of paved roads on sodium and chloride in lakes.

Salinization of surface water from sodium chloride (road salt) applied to paved roads is a widely recognized environmental concern in the northern hemisphere, yet practical information to improve winter road management to reduce the environmental impacts of this deicer is lacking. The purpose of our study was to provide such information by developing baseline concentrations for sodium and chloride for lakes in watersheds without paved roads, and then determining the relationship between these ions and density, type, and proximity of paved roads to shoreline. We used average summer (June-September) sodium and chloride data for 138 lakes combined in a watershed based analysis of paved road networks in the Adirondack Park of New York, U.S.A. The watersheds used in our study represented a broad range in paved road density and type, 56 of which had no paved roads. Median lake sodium and chloride concentrations in these 56 watersheds averaged 0.55 and 0.24 mg/L, respectively. In contrast, the median sodium and chloride concentrations for the 82 lakes in watersheds with paved roads were 3.60 and 7.22 mg/L, respectively. Paved road density (lane-km/km(2)) was positively correlated with sodium and chloride concentrations, but only state roads were significantly correlated with sodium and chloride while local roads were not. State road density alone explained 84 percent of the variation in both ions. We also successfully modeled the relationship between road proximity to shoreline and sodium and chloride concentrations in lakes, which allowed us to identify sections of road that contributed more to explaining the variation in sodium and chloride in lakes. This model and our approach could be used as part of larger efforts to identify environmentally sensitive areas where alternative winter road management treatments should be applied.