Long-term increases in station-derived drought near Apalachicola National Forest, Florida

James B Elsner 

Long local weather records can reveal changes in drought exposure, but a drought index calculated at one station must be distinguished from soil-water measurements across a forest. We updated daily temperature and precipitation observations at Tallahassee airport and recomputed the Keetch-Byram Drought Index (KBDI) with explicit quality control, rainfall interception, and missing-data rules. Across 69 complete eligible annual records spanning 1945-2024, mean moisture deficit increased by 2.63 mm per decade (95% interval: 0.58-4.67), and the annual number of days with KBDI above 650 increased by 2.59 days per decade (0.67-4.51). May showed the clearest monthly drying signal after adjustment for 12 monthly comparisons: 4.64 mm per decade (1.68-7.61; adjusted p = 0.031). April’s trend remained uncertain. May slopes stayed positive across 27 tested KBDI settings, although uncertainty in high-deficit-day trends depended on the record’s starting year. A historical 2009-2018 PRISM comparison showed airport-forest daily anomaly correspondence, with an area-weighted squared correlation of 0.72. As an application, April 30 moisture deficit reduced point-estimate hindcast error for May-July natural-fire counts, but the paired skill interval included no improvement. The strongest evidence therefore concerns increasing station-derived drought exposure and its seasonal structure. It does not establish a forest-wide soil-moisture trend, an anthropogenic attribution, or a validated operational wildfire forecast