Calibration and validation of linked water temperature models for the Shasta Reservoir and the Sacramento River from 2000 to 2015

Mechanistic-based water temperature models simulating how environmental and operational conditions affect water temperature dynamics in the Shasta/Sacramento system can aid water resource management in the region. This report outlines the process of linking two water temperature models (an upstream reservoir model to a downstream river model), including calibration and validation from 2000 to 2015 and the procedures used to run the models operationally. Validation results (even years) indicated the reservoir model matched observed vertical temperature distribution and discharge temperature well, with a RMSE on the order of 1° C. When ran with known upstream boundary conditions, the river model accurately matched daily average river temperatures, with a RMSE near 0.5° C. Running the models in series (i.e. the reservoir model supplying boundary conditions to the river model) resulted in the RMSE of daily average river temperatures to increase to approximately 1° C. Both the reservoir and river models tended to perform poorer from May to October during the temperature management season and predict warmer temperatures than observed. While further model refinement is needed, the linked model framework represents a useful tool to evaluate temperature dynamics of the system under past and future environmental and operational conditions. Miles E. Daniels, Vamsi K. Sridharan, Sara N. John, and Eric M. Danner. "March 2018." Includes bibliographical references (pages 41-42).