Different NAO slopes gave no clear winter sea-level reconstruction gain at 60 European gauges in 2005-2024

We tested whether separate response slopes for positive and negative North Atlantic Oscillation (NAO) improve winter monthly sea-level reconstruction beyond a linear NAO term and recent local sea level. We reused fixed PSMSL and NOAA data at 60 European gauges. Each fit used earlier years only. In 4,790 gauge-month tests in 2005-2024, the added slope term reduced mean squared error by 2.19%. The conditional 95% three-year block interval was -1.85% to +6.20%. The test thus did not resolve an added gain. Equal cell weights, a 20-year fit window and removal of the local lag term also gave intervals that include zero. Both models use the target month NAO, so these are reconstructions, not forecasts. The result does not prove equal physical response slopes. The short test, selected network, simple model and fixed revised data limit inference. The package gives code, fixed inputs, source hashes, all model outputs, uncertainty draws and an independent audit.

NAO · sea level · different response slopes · PSMSL · monthly reconstruction · null result · chronological validation