A 13.66-day fit explained 34% and 45% of later daily midpoint variance at San Francisco and Seattle in 1993-2025

We tested the transfer of fixed lunar periods in daily high-low water-level midpoints at six US tide gauges. We fitted sine and cosine terms on 1960-1992 and evaluated them on 1993-2025. A single 13.66-day term explained 34.47% of later midpoint variance at San Francisco and 45.42% at Seattle, against an early-period mean. Conditional 95% three-year block intervals were 32.46% to 36.64% and 44.33% to 46.63%. Three lunar periods explained 42.93% and 66.46%. Nearby 13-, 14-, and 15-day controls explained less than 0.1% at these sites. Four Atlantic gauges had 0.25% to 1.99% explained by the 13.66-day term. An initial upper-1% recurrence test found no clear fortnightly excess for daily maxima at any of the six sites. Midpoint recurrence had a positive contrast at the two Pacific sites. This extends a campaign tide-subtraction check with a simple long-record transfer test. It supports lunar variation in daily midpoints, but does not isolate a tide constituent or storm surge. Selected gauges, retrospective monthly mean removal, fixed training fits, source revisions, and exploratory analysis limit inference.

daily midpoint · tide gauge · lunar periods · temporal transfer · NOAA CO-OPS · method check