Characteristics of intra-, inter-annual and decadal sea-level variability and the role of meteorological forcing: the long record of Cuxhaven

This paper addresses the role of meteorological forcing on mean sea level (MSL) variability at the tide gauge of Cuxhaven over a period from 1871 to 2008. It is found that seasonal sea level differs significantly from annual means in both variability and trends. The causes for the observed differences are investigated by comparing to changes in wind stress, sea level pressure and precipitation. Stepwise regression is used to estimate the contribution of the different forcing factors to sea level variability. The model validation and sensitivity analyses showed that a robust and timely independent estimation of regression coefficients becomes possible if at least 60 to 80 years of data are available. Depending on the season, the models are able to explain between 54 % (spring, April to June) and 90 % (winter, January to March) of the observed variability. Most parts of the observed variability are attributed to changes in zonal wind stress, whereby the contribution of sea level pressure, precipitation and meridional wind stress is rather small but still significant. On decadal timescales, the explanatory power of local meteorological forcing is considerable weaker, suggesting that the remaining variability is attributed to remote forcing over the North Atlantic. Although meteorological forcing contributes to linear trends in some sub-periods of seasonal time series, the annual long-term trend is less affected. However, the uncertainties of trend estimation can be considerably reduced, when removing the meteorological influences. A standard error smaller than 0.5 mm/year requires 55 years of data when using observed MSL at Cuxhaven tide gauge. In contrast, a similar standard error in the meteorologically corrected residuals is reached after 32 years.