Stream Conductivity: Relationships to Land Use, Chloride, and Fishes in Maryland Streams

Abstract

Stream conductivity reflects both landscape and anthropogenic interactions, although increasing chloride inputs from road salt in eastern North America is also important. Employing a spatially extensive database derived from the Maryland Biological Stream Survey (MBSS), we examined relationships of stream conductivity to landscape attributes and determined relationships between stream MBSS fish metrics, abundance, and biomass as well as fish assemblages by means of conductivity (as a chloride surrogate) to estimate potential effects. Background stream conductivity for the MBSS strata and Maryland L3 ecoregions ranged from 51 to 150 μS/cm, with the Piedmont having the highest background conductivity (145–160 μS/cm). For MBSS sites there were strong relationships of stream conductivity and chloride with both impervious surface and road density, and 0.26% of the MBSS sites exceeded the U.S. Environmental Protection Agency's acute chloride criterion and 1.5% the chronic chloride criterion. For the Maryland Piedmont species assemblage, observed conductivity values between 230 and 540 μS/cm caused important alterations in the biotic community as measured by the series of fish metrics, and thus affected fish species assemblages associated with chloride concentrations between 33 and 108 mg/L. Maryland coastal, coldwater, and highland species assemblages displayed variable responses of conductivity to fish metrics, abundance, and biomass. If regional conductivity and chloride levels continue to increase or remain high from road deicing owing to an increase in urbanization in a watershed or climatic changes, differences in stream fish assemblages and therefore fish diversity may become more apparent. Because of the linkage between salt usage and impervious road surfaces resulting from urbanization, it is important to manage effectively the use of road salt to protect biotic resources.

Received January 24, 2012; accepted June 5, 2012

ACKNOWLEDGMENTS

This paper is based on water quality analyses completed under numerous MBSS contracts (MANTA, Maryland Department of Natural Resources [MDNR]) given to R.P.M. from 1993 to 2009. In addition, we thank personnel associated with MDNR and Versar, Inc., for their assistance and excellent stream monitoring database, and the foresight, in the early 1990s, of the MDNR in development of this stream assessment program. This is Scientific Contribution Number 4664 from the Appalachian Laboratory of the University of Maryland Center for Environmental Science.