Influences of local hydroclimatology and teleconnections on Florida's precipitation and temperature variability

Understanding the influences of local hydroclimatology and two large-scale oceanic-atmospheric oscillations (i.e., Atlantic Multidecadal Oscillation (AMO) and El Niño-Southern Oscillation (ENSO)) on seasonal precipitation (P) and temperature (T) relationships for a tropical region (i.e., Florida) is the focus of this study. The warm and cool phases of AMO and ENSO are initially identified using sea surface temperatures (SSTs). The associations of SSTs and regional minimum, maximum and average surface air temperatures (SATs) with precipitation are then evaluated. The seasonal variations in P-SATs and P-SSTs associations considering AMO and ENSO phases for sites in (1) two soil temperature regimes (i.e., thermic and hyperthermic); (2) urban and non-urban regions; and (3) regions with and without water bodies, are analysed using two monthly datasets. The analyses are carried out using trend tests, two association measures, nonparametric and parametric statistical hypothesis tests and kernel density estimates. Decreasing (increasing) trend in precipitation (SATs) is noted in the recent multi-decadal period (1985–2019) compared to the previous one (1950–1984) indicating a progression towards warmer and drier climatic conditions across Florida. Spatially and temporally non-uniform variations in the associations of precipitation with SATs and SSTs are noted. Strong positive (weak negative) P–T associations are noted during the wet (dry) season for both AMO phases and El Niño, while significant (positive) P–T associations are observed across southern Florida during La Niña in the dry season. The seasonal influences are predominant in governing the P–T relationship over the regions with and without water bodies; however, considerable variations between El Niño and La Niña are noted during the dry season. The climate variability influences on P–T correlations for hyperthermic and thermic soil zones are found to be insignificant (significant) during the wet (dry) season. Nonparametric clustering is performed to identify the spatial clusters exhibiting homogeneous P–T relationships considering seasonal and climate variability influences.