n Integrated Framework for Improved Stream Temperature Predictions to Mitigate Fish Mortality

Jason Caldwell1,2, Eric Danner, Andrew Pike, Balaji Rajagopalan2,1, and Forrest Melton

1CIRES, University of Colorado, Boulder, CO
2Civil Environmental and Architectural Engineering, University of Colorado, Boulder, CO
University of California
In 2004, the National Marine Fisheries Service (NMFS) issued a Biological Opinion (BiOp) to outline the decision support system for water allocations in the Central Valley Project (CVP) with respect to impacts on threatened and endangered species in the Sacramento River Basin. Peer-review of the BiOp identified fundamental flaws in two critical components, the stream temperature and fish mortality models, due to limitations of the proposed methods in both temporal and spatial resolution. To address these issues, an integrated framework was proposed that would result in the development of a suite of decision support tools (DSTs) for resource managers. The overall approach is to utilize satellite-derived inputs in ecological and numerical weather prediction models to provide environmental inputs to the stream temperature models at increased temporal and spatial resolutions. The higher-resolution stream temperature forecasts can then be implemented in the fish mortality models. Additionally, the framework includes the development of stochastic weather generator software and statistical modeling tools to address both short- (e.g., daily) and long-term (e.g., seasonal, annual) predictions of a suite of hydrometeorological variables, including stream temperature. By integrating state-of-the-art modeling systems with statistical analysis and prediction methods, a comprehensive set of DSTs can be developed that will best guide water resource management decisions in the CVP. We will describe the proposed decision support system framework in an overview fashion to highlight the integrated and easily-transferable design of the project.