Determining Reservoir Tolerance to Forecast Errors in the Southwest U.S. in Support of Forecast-Informed Reservoir Operations (FIRO)

Código

I-EBHE0182

Autores

Eric Shearer, Rachel Weihs, Tyler Roberts, Joseph Forbis

Tema

WG 2.6: Water systems analysis for integrated planning & management

Resumo

The United States Army Corps of Engineers (USACE) is currently supporting the Forecast-Informed Reservoir Operations (FIRO) strategy, which aims to integrate meteorological and hydrologic forecasting into the decision-making process of reservoir operations across the portfolio of USACE dams. A persistent challenge within this project is determining the limits of leverageable forecast skill at specific reservoir sites. To address this, we have developed a methodology that uses reservoir and watershed characteristics to assess lead times of runoff volumes under a "worst-case scenario" approach. Specifically, we analyze the 90th percentile of accumulated Global Ensemble Forecast System (GEFS) under-forecast errors (R_err) over every multi-day accumulation period between 1 to 10 days (e.g., 1 day, 1-2 days, 1-3 days, etc.), averaged across a reservoir's drainage basin. This error is then converted into inflow error (I_err) by considering the drainage basin size (A_w) and basin-averaged USDA curve numbers (CN) during wet conditions. I_err= R_err×A_w×CN Lead times are identified as the last day when the inflow error remains below a reservoir?s flood storage capacity (S_fl). I_err< S_fl This metric has been calculated for approximately 90 sites in the South Pacific Division of USACE, encompassing regions in California, Nevada, Arizona, New Mexico, and parts of Colorado and Texas, to illustrate the distribution of safe lead times. The assumptions, limitations, and sensitivities associated with this metric are also thoroughly discussed. The results help identify locations where forecast errors could be rectified with flood capacity under a potential FIRO water management strategy.