The largest three-year dry footprint across 344 US divisions was within a shuffled-year range in 1896-2025

We tested whether year order produced an unusually large common footprint of three consecutive dry water years in 344 US climate divisions. We used fixed NOAA nClimDiv precipitation for 1896-2025. A dry year was at or below each division's full-record 20th percentile. The largest count of divisions dry in all three years was 36, in 1964-1966. This was 10.47% of divisions, not US land area. We compared this count with 9,999 records that kept each national annual field together and reordered years within 20-year strata. Each draw included the search over all window dates. The null median was 32 divisions and the central 95% range was 19-80. The upper-tail p value was 0.3915. Thus, the test found no clear excess under this year-order reference. Planned checks changed strata, threshold, duration, year boundary, and division weights. All gave upper-tail p values above 0.05, although four-year and equal-state checks were closer to the upper tail. This is a null result conditional on within-stratum exchangeability. It does not prove independence or measure recovery of water stores. One product, annual aggregation, division weights, and slow climate variation limit inference. The package gives source data and hashes, exact annual evidence and permutations, all null scores, three figures, code, a separate arithmetic audit, and a repeat without network access.

precipitation drought · persistent drought · spatial dependence · year permutation · null result · nClimDiv · water year · national drought footprint