MEASURES OF THE USEFULNESS OF SEASONAL PRECIPITATION FORECASTS FOR AGRICULTURAL APPLICATIONS

Authors: Schneider, Jeanne; Garbrecht, Jurgen

Submitted to: Transactions of the ASAE
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/31/2002
Publication Date: 5/31/2002
Citation: Schneider, J.M., Garbrecht, J.D. A measure of the usefulness of seasonal precipitation forecasts for agricultural applications. American Society of Agricultural Engineers. 2003. v. 26. p. 257-267.

Interpretive Summary: NOAA's Climate Prediction Center produces experimental climate forecasts that predict total precipitation over three-month periods out to a year in advance. The utility of these seasonal forecasts for agricultural applications will depend on a number of forecast characteristics, including dependability, effectiveness, and usefulness. This study examines the usefulness of the forecasts, defined as the ability of the forecasts to predict precipitation totals significantly different from those experienced during the previous 30 years (climatology). Four measures of usefulness were computed for six years of forecasts for the continental United States. Results vary with location, with some regions showing much larger and more frequent forecasts of departures from climatology. For example, departures larger than 10% of normal were issued in 40% of the forecasts for southeastern Arizona, but only 6% of the time for southern Nebraska. Usefulness also varies with ENSO state and intensity, and season, with the largest forecast departures issued during strong ENSO events, and during the fall, winter, and spring seasons. Even if seasonal precipitation forecasts are shown to be dependable and effective, the usefulness of these forecasts for individual, local agricultural planners and managers will be limited in regions where forecasts departures are small. Agricultural enterprises that operate at regional scales, and can profitably use predictions of small shifts in probable outcome (e.g., crop insurance programs, fertilizer production and distribution, and grain storage and transportation), are best suited to benefit at this time from these forecasts.

Technical Abstract: NOAA's Climate Prediction Center produces experimental climate forecasts that predict total precipitation over three-month periods out to a year in advance. The utility of these seasonal forecasts for agricultural applications will depend on a number of forecast characteristics, including dependability, effectiveness, and usefulness. Usefulness is defined as the ability of the forecasts to predict conditions significantly different from climatological norms. This definition assumes that producers would be more likely to "use" a forecast if it predicts a large departure from normal conditions. Four measures of usefulness are developed for the NOAA probability of exceedance seasonal precipitation forecasts. Based on the archived precipitation forecasts from 1995 through 2000, these four measures are evaluated for 102 forecast divisons covering the continental United States. Results vary significantly across the United States, with some regions showing much larger and more frequent forecasts of departures from climatological precipitation than others. For example, precipitation forecst departures larger than 10% of climatological norms were issued in 40% of the forecasts for southeastern Arizona, but were issued only 6% of the time for southern Nebraska. Usefulness also varies with ENSO state and intensity, and season, with the largest forecast departures issued during strong ENSO events, and during the fall, winter, ans spring seasons. Even if seasonal precipitation forecasts are shown to be dependable and effective, the usefulness of these forecasts for individual, local agricultural planners and managers will be limited in regions where forecasts rarely depart from climatology. Agricultural enterprises that operate at regional scales, and can profitably use predictions of small shifts in probable outcome (e.g., crop insurance programs, fertilizer production and distribution, and grain storage and transportation), are best suited to benefit in a direct fashion from these forecasts.