El Niño Southern Oscillation and Chill Accumulation Variability in the Southeastern United States

ABSTRACT

Winter chill deficits disrupt dormancy release and synchronised budbreak, reducing fruit set and yields and increasing management costs. The southeastern United States is especially vulnerable because many areas already lie near cultivar chilling thresholds and warming is amplifying year‐to‐year chilling deficit hazard. Because the El Niño–Southern Oscillation (ENSO) offers rare seasonal predictability for winter conditions, quantifying its effects on chill accumulation alongside long‐term warming can improve cultivar selection, site planning and in‐season risk advisories. This study evaluates the influence of ENSO phase and event type, distinguishing between Eastern Pacific (EP) and Central Pacific (CP) events, on chill accumulation using 74 seasons (1950–2024) of ERA5‐Land temperature data across Florida, Georgia, Alabama and South Carolina. Chill accumulation was quantified using the Chill Hours model (0°C–7.2°C) and the Dynamic Model (Chill Portions), and temporal trends were assessed through the Mann–Kendall test and Sen's slope estimator. Results show that chill accumulation varies substantially among ENSO phases. El Niño events generally enhance chill accumulation, particularly in northern Georgia, Alabama and South Carolina, while La Niña events lead to notable reductions across the region. Neutral seasons exhibited values close to long‐term averages. Separating EP and CP types refined this response, with CP La Niña events producing the largest and most spatially extensive reductions in both models, whereas among El Niño events EP produced the largest Chill Portions increases, and the two types differed mainly in the location of the maximum Chill Hours response. The Dynamic Model revealed smoother spatial gradients and reduced coastal anomalies relative to Chill Hours. Trend analyses indicate significant declines in chill accumulation, especially during the early season (October–December), with Florida and southern Georgia exhibiting the largest areas of statistically significant negative trends. Comparisons between models showed that Chill Portions identified broader areas of significant long‐term decline, reflecting the Dynamic Model's sensitivity to physiologically effective chilling rather than total hours below a temperature threshold. Overall, both ENSO variability and long‐term warming contributed to reduced chill accumulation, highlighting increasing vulnerability of temperate fruit crops and the need for region‐specific adaptation strategies in the southeastern United States.