Forecast-Informed Reservoir Operations in Texas: what will it take to say “yes”?

Abstract:

Forecast-Informed Reservoir Operations (FIRO) is widely viewed as a potentially cost-effective approach for maintaining reliable water supply and mitigating the impacts of floods. Yet, it has not been widely adopted in Texas due to several unique challenges. Being a large state with diverse climates, forecast information needs vary greatly depending on the climate, the size and functions of reservoirs, downstream conditions, and the experience and training of operators. Key factors that hinder the adoption of forecasts include limitations in forecast accuracy and specificity, infrastructural and operational constraints facing reservoir operators, and information gaps between forecast providers and operators. This presentation outlines the science, technology, and operational gaps that need to be addressed to facilitate a wider adoption of the FIRO paradigm in Texas.

The first major challenge is insufficient predictive skills. The predictability associated with floods and droughts, which drive storage-release decisions, is limited. In addition, forecast products are not specifically tailored to be readily usable in reservoir operation. Second, there are large information gaps between reservoir operators and forecast purveyors — reservoir operators are often unaware of emerging forecast products and associated skill and are not typically equipped to interpret and utilize forecast uncertainty information. Third, there are infrastructure constraints (e.g., reservoirs have been designed to conserve water and releasing water quickly may require retrofitting or add adding more flood gates) that must be addressed if the proactive use of forecast information is to yield tangible benefits. Our recommendations include investment in advancing the predictability of weather systems that cause flooding in Texas, enhancing the skill and specificity of current hydrometeorological and hydrologic forecasts, improving methods for conveying uncertainty and forecast confidence, and identifying reservoirs for which integration of certain flavors of FIRO might be practical given the accuracy of current forecast products.