Increased salinization of fresh water in the northeastern United States

Abstract

Chloride concentrations are increasing at a rate that threatens the availability of fresh water in the northeastern United States. Increases in roadways and deicer use are now salinizing fresh waters, degrading habitat for aquatic organisms, and impacting large supplies of drinking water for humans throughout the region. We observed chloride concentrations of up to 25% of the concentration of seawater in streams of Maryland, New York, and New Hampshire during winters, and chloride concentrations remaining up to 100 times greater than unimpacted forest streams during summers. Mean annual chloride concentration increased as a function of impervious surface and exceeded tolerance for freshwater life in suburban and urban watersheds. Our analysis shows that if salinity were to continue to increase at its present rate due to changes in impervious surface coverage and current management practices, many surface waters in the northeastern United States would not be potable for human consumption and would become toxic to freshwater life within the next century.

Fig. 1.

Examples of significant, long-term increases in baseline concentration of chloride for streams and rivers of the northeastern United States. The R² values are given for linear regressions. All streams and rivers are located in rural areas but contain roads within their watersheds. (A) LMR0015 (Little Morgan Run), MDE0026 (Middle Run), and BEA (Beaver Run) are sampling stations for tributaries to Liberty Reservoir, a drinking water supply for Baltimore. (B) Wappinger Creek and the Mohawk River are tributaries to the Hudson River in the Hudson River Valley. (C) The streams in the White Mountains drain into Mirror Lake; one is located near an interstate highway in the Hubbard Brook Valley, and the forested reference stream is watershed 6 of the Hubbard Brook Experimental Forest (10).

Fig. 2.

Relationship between impervious surface and mean annual concentration of chloride in streams of the Baltimore LTER site during a 5-year period (R² = 0.81). Sites are located along a gradient of urbanization. Dashed lines indicate thresholds for damage to some land plants and for chronic toxicity to sensitive freshwater life (6, 8).