Combining Groundwater and Teleconnection Information to Enhance Wet-Season Runoff Forecasting in Central-Southern Oregon Basins
For the Deschutes and Klamath Projects, reservoir operations during wet-season are based on expected November to March runoff volumes, where the anticipation is currently informed by climatology and potentially polling of season-accumulated precipitation and snow conditions. This Science and Technology (S&T) Program research project asks:
* Can we improve this anticipation by also considering information from ground water monitoring and teleconnections signals?
These latter two information sources describe how longer-term climate trends and forecast-horizon climate conditions might influence runoff volume. Neither is significantly described by currently used information. Groundwater offers potential information given that base flow contributes a significant portion of runoff in both basins, and because ground water expresses the long-term climate trend. Teleconnections offer potential information given that they have been identified to predict wet-season climate conditions in both the Mid-Pacific (MP) and Pacific Northwest (PN) regions.
Need and Benefit
For PN's Deschutes and MP's Klamath projects, wet-season release operations are constrained primarily by environmental demand obligations rather than flood control objectives. For the Deschutes operators, the environmental need is flow stability through the wet-season, designed for supporting balanced populations stemming from the early- and late-season spawning periods of two competing fish species (i.e., Red Band Trout and Brown Trout), and for avoiding riverbed dewatering that would lead to erosion, frost heave, and water quality problems. For the Klamath operators, the concern is meeting prescribed downstream flow objectives that vary in time during the wet season. Regardless of demand specifics, both projects have the objective of scheduling monthly release patterns based on wet-season runoff anticipation (i.e., inflow into Wickiup/Crane Prarie reservoirs for the Deschutes Project and Upper Klamath Lake for the Klamath Project). Adequate scheduling of these release patterns affects Reclamation's ability to avoid conflicts related to ecosystem management, which could compromise capability to deliver contract water supplies and generate hydropower.
Operators of both projects establish wet-season release schedules based on anticipated wet-season runoff volume (generally November to March). Klamath Project operators currently anticipate wet-season runoff based on climatology and observed base flow leading into the wet season. Starting in January, they also get forecast information from the Natural Resources Conservation Service (NRCS) on volumetric runoff for the February-through-July period. Predictors in the NRCS model are based on season-accumulated precipitation and snow information prior to the January issue date.
Deschutes Project operators also consider climatology and base flow in their wet season anticipation. They also make use of a statistical forecast model developed by NRCS that predicts Upper Deschutes runoff volume from November to March. This model bases prediction on prior-to-issue information on season-accumulated precipitation and snowpack. It is meant to be used initially in November (i.e., the first issue is November) and allows for monthly forecast updates through March first. Deschutes operators have found the November to March model to be significantly uncertain at the earlier issues (November and December), which is reasonable given the weak relationship between the basin's pre-December precipitation and snow conditions and the November to March climate as a whole.
Both the Deschutes and Klamath projects would benefit from improved November to March runoff forecasts leading to more certain release scheduling, enabling better setup of spring water supplies and promoting better avoidance of environmental conflicts related to wet-season releases. The proposed work explores potential for improving November to March runoff forecasts through consideration of currently polled information (i.e., precipitation and snow) jointly with ground water information (i.e., indicators of longer-term climate trends) and teleconnections indices (i.e., indicators of climate during the forecast-horizon).
Ground water indices are of interest because base flow is a significant wet-season runoff component in both basins, and because ground water conditions vary interannually with longer-term climate trends.
Teleconnections indices are of interest due to evidence that upwind antecedent climate conditions are related to wet-season climate and snowmelt runoff volumes in MP and PN regions. Teleconnections sometimes offer information on snowmelt volumes at lead-times of up to six months or more. Given this lead-time relative to April-July runoff, it is thought that these signals may also provide information on precipitation-driven wet-season runoff during November to March.
Contact the Principal Investigator for information about these documents.