Chloride pollution from road salt is a common issue in urban and suburban communities in the snow belt. Regional studies of chloride often include only a small number of sites or utilize grab samples to demonstrate stream impacts. We undertake this study to more fully assess both the spatial and temporal dynamics of this pollutant that is known to be toxic to aquatic life. Data loggers deployed at 20 stream sites throughout St. Louis County, Missouri, are used to characterize the extent of chloride pollution in the region. This extensive continuous dataset is evaluated using concentration-duration-frequency (CDF) curves in order to more fully characterize in-stream conditions.
In addition, we conduct a laboratory evaluation of the toxicity of chloride to freshwater Gammarus (Crustacea, Amphipoda). Toxicity studies generally place test organisms in environmental chambers with a temperature of 18-22OC; these temperatures are unrealistic for the natural setting where road salt is applied in response to snow events. To account for this difference in the literature, this study evaluates the lethal toxicity of chloride at 6OC and 19OC. Feeding success is also evaluated as a non-lethal endpoint.
Results/Conclusions
Results show highly variable chloride concentrations within the region, both spatially and temporally. While pre-winter baseline chloride concentrations were below 100 mg/l in many streams, storm-related spikes of over 7,000 mg/l chloride are found in some of the more urban watersheds.
Differences in the toxicity of chloride to amphipods based on temperature indicate that the existing chloride toxicity literature may be inadequate to characterize in-stream impacts of winter road salt application. Decreased leaf consumption at chloride concentrations over 1,400 mg/l is indicative of the high potential for non-lethal impacts of long-term increases in chloride concentration.