Predictive Modeling\nReveals Elevated Conductivity\nRelative to Background Levels in Freshwater Tributaries within the\nChesapeake Bay Watershed, USA

Elevated conductivity (i.e., specific conductance or\nSC) causes\nosmotic stress in freshwater aquatic organisms and may increase the\ntoxicity of some contaminants. Indices of benthic macroinvertebrate\nintegrity have declined in urban areas across the Chesapeake Bay watershed\n(CBW), and more information is needed about whether these declines\nmay be due to elevated conductivity. A predictive SC model for the\nCBW was developed using monitoring data from the National Water Quality\nPortal. Predictor variables representing SC sources were compiled\nfor nontidal reaches across the CBW. Random forests modeling was conducted\nto predict SC at four time periods (1999–2001, 2004–2006,\n2009–2011, and 2014–2016), which were then compared\nto a national data set of background SC to quantify departures from\nbackground SC. Carbonate geology, impervious cover, forest cover,\nand snow depth were the most important variables for predicting SC.\nObservations and modeled results showed snow depth amplified the effect\nof impervious cover on SC. Elevated SC was predicted in two-thirds\nof reaches in the CBW, and these elevated conditions persisted over\ntime in many areas. These results can be used in stressor identification\nassessments to prioritize future monitoring and to determine where\nmanagement activities could be implemented to reduce salinization.