Uncertainty and bias in electronic tide-gauge records: Evidence from collocated sensors

Understanding noise and possible bias in tide-gauge sensors is important for determining the mean sea level, its fluctuations and their climatic, geophysical and engineering implications, but not an easy task. In the past, this problem has been examined through comparison of different sensors in the laboratory, or through correlations of neighbouring sensors. In this study we identified and studied 10 cases of harbours with fully collocated sensors. Transient differences were found between collocated records. Pressure gauges were found significantly more sensitive to noise than radar-type sensors, and with higher chances of long-term transient bias. The amplitude of the observed bias is important, of the same order of magnitude with tsunami waves in the open sea and with seismic ground displacements. Only 9% of the sensors analysed were found to satisfy the 1 cm accuracy criterion imposed by the Permanent Service for Mean Sea Level (PSMSL).