Experimental drought threat forecast for Florida

Short-term climate variations including the El Niño/Southern Oscillation (ENSO) cause anomalies in temperature and precipitation that could produce or contribute to extreme events including droughts and wildfires. Wildfire threat potential in the Southeast is based in part upon the Keetch–Byram Drought Index (KBDI), and the KBDI uses daily temperature and rainfall measurements. The KBDI responds to changing climate and weather conditions on time scales of days to months, and this index is high during dry warm weather patterns and low during wet cool patterns. In the Southeastern United States, these dry periods typically occur during the winter and spring of La Niña years while wet periods occur during the winter and spring of El Niño years. This study computes forecast probabilities of KBDI exceeding thresholds by using temperature and precipitation data generated by a stochastic weather generator and initial KBDI from the observed dataset. The weather generator produces multiple realizations of ENSO's impact on local temperature and precipitation. Relatively high probabilities of exceeding thresholds tend to occur during the winter and spring months of La Niña years in both Jacksonville and Belle Glade. Low probabilities occur during the winter and spring months of El Niño years. The distribution is reversed during late spring and early summer months, but the distribution is smaller during the summer than during the winter. Verification indicates high skill during January, February, and May and low skill in April and June. Although other wildfire decision-support tools exist, foresters in the Southeastern United States commonly use the KBDI since its development during the 1970s. This study generates probabilistic KBDI forecasts based on initial KBDI and the ENSO signal.