Vulnerability of River Systems to Climate Changes Across the Conterminous United States

As they operate over the scale of the river catchment, climate factors influence physical characteristics of rivers draining those catchments, and ultimately, their biological assemblages. Using a landscape approach, we assessed the vulnerability of river systems throughout the conterminous United States to predicted changes in climate in 10-year intervals to the year 2040. With modeled changes in air temperature and precipitation available for this effort, we assumed that changes in thermal and flow regimes would result directly from predicted changes in climate.  We began by assembling historical daily stream flow and temperature data available through the National Water Information System (NWIS) and attributed them to individual stream arcs represented by the National Hydrography Dataset Plus (NHDplus).  Using the hydrological index tool (HIT), long-term flow records were summarized into a large set of metrics characterizing stream flow regimes, and stream temperature data were also used to calculate metrics describing thermal regimes. Using an indicator analysis approach designed to recognize species to their matched flow and thermal characters, a subset of both flow and temperature metrics determined to be important to fish were identified. This analysis was conducted separately within nine ecologically-defined regions of the conterminous United States, and resulted in a list of regionally-specific fish species most responsive to stream flow and temperature as well as the identification of habitat metrics important to fish species. Using a set of fish records compiled from throughout the conterminous United States, we then performed a classification to group stream arcs supporting similar fish assemblages and having similar natural characteristics including catchment area, stream reach slope, surficial geology, and climate variables. By applying modeled changes on a reach-by-reach basis, we calculated a relative change in score and in some cases class to characterize vulnerability. Assembling these reach-by-reach information, we have the ability to map and compare the vulnerability of each reach to projected changes in climate and to understand vulnerability on a regional basis, providing managers with better information to plan for potential changes.