Distributed Hydrologic Modeling For Water Management Decision Support In The Russian River Basin, California

The Russian River watershed encompasses 1,485 square miles within Sonoma and Mendocino Counties, California. It is one of the most flood-prone rivers in the State of California because of the watershed's unique geography and its proximity to the coast, which together produce climatologically heavy wintertime rainfall, frequent flooding and mudslides. The Russian River basin is the most flood prone in California and periodically suffers high damages from the incidence of land falling atmospheric rivers (ARs) during the winter-spring seasons. In addition to the flood threat, ARs provide the main source of water volumes for water supplies; there is little precipitation during the summer-fall seasons. When flood events occur, reservoir operations for flood storage compete with water supply storage needed for a burgeoning population of more than 600,000 in Sonoma County. And diversions for vineyard irrigation and frost protection continue to expand in response to high value wine industry demands. Further, endangered salmon fisheries require tributary and main stem flows during the late summer and early fall spawning season. These water management information needs are being addressed using a high resolution distributed hydrologic model which can provide unregulated flow estimates at any location in the watershed and at small time increments. The Research Distributed Hydrologic Model (RDHM) developed by NOAA National Weather Service (NWS) Office of Hydrologic Development (OHD) is the primary tool for modeling unregulated flows at the 4-km HRAP and 6-hr time step resolution; a 1-km, 1-hr version of the RDHM has also been developed. The RDHM is being prototyped for nationwide deployment by the OHD and they have provided base data sets on terrain, grid connectivity, soils and a priori parameter values for the RDHM setup. The California-Nevada River Forecast Center (CNRFC) provided precipitation data. We used these data for the initial version of the RR model.

In this paper we summarize the RDHM setup, initialization and calibration-verification activities as applied to selected unregulated tributaries. The calibration approach is described and performance metrics for calibration and verification are summarized. In accomplishing the model calibrations we have attempted to characterize model performance for flood peaks and low flows, and to characterize a complete water budget for the simulation period. A primary issue for basin-wide application of the model is the influence that reservoir storage operations and diversions have on the water balance on specific stream reaches. Accordingly, we have sought to integrate the RDHM generated “natural” flows with water management models to obtain the "managed" flows. Reservoir operations strongly influence the main stem flood flows; and irrigation diversions, pond storage and return flows strongly influence low flows throughout the basin. In this paper we summarize alternative approaches for representing the flow management practices in the basin and coupling of these with the RDHM gridded “natural” flows.