H059-0006
Sensitivity Analysis of the Colorado Basin River Forecast Center's Hydrologic Model
Sensitivity Analysis of the Colorado Basin River Forecast Center's Hydrologic Model
Wednesday, 9 December 2020
Poster
Abstract:
The Colorado Basin River Forecast Center (CBRFC) provides impact-based decision support services for a broad range of stakeholders over the Colorado River Basin and the eastern half of the Great Basin. Streamflow forecasts and water supply information developed by the CBRFC is particularly critical to reservoir operations, planning, and management decisions made by the Bureau of Reclamation (Reclamation), which impact state water resource agencies throughout the region. As active members of the Colorado River Climate and Hydrology Workgroup, the CBRFC and Reclamation identified the need to quantify the sensitivity of the CBRFC’s hydrologic forecasting model to perturbations of the model’s primary forcing parameters. To this end, the CBRFC perturbed historical precipitation, evapotranspiration, and soil moisture parameters by +2.5%, +5.0%, and +10.0%. Temperature was perturbed by +0.5 oF, +1.0 oF, and +2.0 oF. These perturbations were done over six headwater basins minimally impacted by anthropogenic influences. The impact of these perturbations to annual, April through July, and October through December water volumes were quantified and compared. Overall, precipitation had the largest impact on annual and seasonal volumetric flow amounts, though soil moisture played a significant role over the October through December timeframe. Temperature perturbations had minimal impacts to modeled annual streamflow volumes, though monthly streamflow distributions were significantly impacted by changes to historical temperature values. While unintuitive, the lack of sensitivity of temperature changes to annual streamflow volumes speaks to the nature of the CBRFC’s hydrologic model, where parameters are seldom functions of temperature, and highlights areas for potential improvement in the CBRFC’s hydrologic modeling paradigm.