Monthly mean removal reduced changes in three-day runs of high daily minima at six US gauges in 1960-2025

We tested changes in three-day runs of high daily minimum water levels at six US tide gauges. Such runs indicate loss of low-water relief relative to a statistical threshold. We used 48 fixed NOAA CO-OPS files for 1960-2025. The main comparison was 1996-2025 minus 1960-1989. Each site and calendar month used its early-period 95th percentile. The observed increase ranged from 12.22 to 149.12 windows per 1,000 valid three-day windows. After removal of observed monthly mean sea level, changes ranged from -1.51 to +2.39. All six centered 95% three-year block intervals included zero. Five-year blocks gave the same interval pattern. A ten-year block check and a shorter-period check gave small positive intervals at Sewells Point. These results extend earlier campaign studies of daily maxima and unusually low minima. They do not show that all event dynamics stayed fixed or identify a physical cause. Tides remain in the data. Selected sites, missing days, hourly sampling, fixed thresholds, and long tide cycles limit inference. These are statistical thresholds, not drainage or flood limits. The package gives daily evidence, annual counts, code, source hashes, figures, checks, and repeat instructions.

sea level · daily minimum · low water · drainage · NOAA CO-OPS · three-day runs