Dry water year correlation shows no clear change across 344 US divisions between 1901-1960 and 1966-2025

We tested whether dry water years became more synchronized across 344 climate divisions in the contiguous United States. We used fixed NOAA nClimDiv precipitation data and compared 1901-1960 with 1966-2025. Each division and period had its own 20th-percentile threshold. Mean correlation across 58,996 division pairs was 0.1182 in the early period and 0.1093 in the late period. The change was -0.0089, with a 95% five-year block interval of -0.0601 to 0.0358. All national year vectors stayed together in each sample, and local thresholds were refitted. Block length, tail threshold, calendar-year, state-weight, and early-decade checks also left changes uncertain. Some checks changed the point estimate sign. The result does not establish equal risk. It is specific to relative annual precipitation deficits and does not measure soil moisture, runoff, water demand, or land area in drought. One source product, only 60 years per period, division weighting, and slow climate variation limit inference. The package supplies raw data, exact source hashes, annual evidence, all bootstrap draws, code, a figure, independent arithmetic checks, and an offline repeat.

precipitation · drought synchrony · nClimDiv · spatial dependence · block bootstrap · null result