A fixed lunar-cycle model bounds excess-span trends to 12.1 mm per decade at six US gauges in 1960-2025

We tested how strongly six US tide-gauge records constrain change in the span between monthly highest and lowest water after subtraction of mean high water minus mean low water. We used NOAA monthly data from 1960-2025. With season, linear time and a fixed 18.61-year term, all six trend intervals include zero. A 95% simultaneous normal bound limits the absolute slopes to 12.04 mm per decade across the six sites. Without the lunar term, the largest bound is 38.08 mm per decade. A 2000 start raises it to 50.19 mm per decade. Thus the tight full-record bound depends on model and time span. This extends a campaign study of separate extreme gaps on the same observations. It is not independent evidence of stable storms or flood risk. Source changes, missing observations, six selected ports and asymptotic error estimates limit the result. The package gives monthly evidence, exact raw source paths and hashes, code, figures and repeat instructions.

tide gauges · monthly extremes · simultaneous trend bounds · lunar nodal cycle · model sensitivity