Improved Stream Temperature Simulations in SWAT Using NSGA-II for Automatic Multi-Site Calibration

Abstract

<abstract> <bold><sc>Abstract.</sc></bold> Stream temperature is one of the most influential parameters impacting the survival, growth rates, distribution, and migration patterns of many aquatic organisms. Distributed stream temperature models are crucial for providing insights into variations of stream temperature for regions and time periods for which observed data do not exist. This study uses a relatively new stream temperature model incorporated into a modified version of the Soil and Water Assessment Tool (SWAT) in order to simulate stream temperatures at five sites on the Calapooia River within the Calapooia basin in northwest Oregon. The Nondominated Sorting Genetic Algorithm II (NSGA-II) is used to calibrate flow at a single outlet and stream temperatures at five sites. Few studies have calibrated this stream temperature model for different basins, and this is the first demonstration of an automatic, subbasin-level calibration for stream temperature at multiple sites. The subbasin calibration is shown to better match the observed data than the original SWAT temperature model as well as the modified temperature model calibrated basinwide. In addition to providing improved stream temperature simulations for the Calapooia River, the subbasin-level automatic calibration technique extends the applicability of the model, especially for complex basins with large spatial variability of topography, land use, and soil type.