Digitizing the Williamstown, Australia Tide‐Gauge Record Back to 1872: Insights Into Changing Extremes.

For Williamstown tide gauge, at the northern‐most point of Port Phillip Bay (PPB), Melbourne, Victoria, tide registers from 1872 to 1966 and marigrams from 1950 to 1966, were digitized to extend sea‐level records back almost 100 years. Despite some vertical datum issues in the early part of the record, the data set is suitable for extreme sea‐level trend analysis after removal of the annual mean sea level. The newly digitized data was combined with the digital record to produce a combined record from 1872 to 2020. Analysis of this record revealed known problems of siltation of the tide gauge stilling well and associated reduction in tidal range at times during 1880–1895 and 1910–1940. A positive trend in tidal amplitude of 0.41 ± 0.01 mm yr−1 was found over 1966–2020, likely due to reduced hydraulic friction at the narrow entrance to PPB. Extreme sea‐level trends were examined over 1872–2020 for storm tides (the combination of storm surge and tide) after removal of the annual mean, and residuals after subtraction of the predicted tides. A non‐stationary Gumbel distribution with a time‐varying location parameter revealed statistically significant declining trends in the residuals of −0.73 ± 0.02 mm yr−1, consistent with the observed poleward movement of storm surge‐producing mid‐latitude weather systems. For storm tides a smaller declining trend of −0.40 ± 0.01 mm yr−1 was found. These trends are approximately an order of magnitude smaller than the current positive rates of mean sea level rise, meaning that storm tide hazard will continue to increase in the future. This information is relevant for future adaptation planning. Plain Language Summary: Sea levels recorded by tide gauges can be studied to understand how sea levels are changing due to mean sea‐level rise (SLR) from ocean warming and melting of land‐based ice as well as due to changes in extreme sea levels (storm tides) due to the combination of astronomical tides and severe‐weather‐induced storm surges. Digitizing historical paper tide gauge records lengthens the digital records and enables more reliable trend analysis. In this study, tide records from 1872 to 1966 were digitized for the Williamstown tide gauge, within the semi‐enclosed Port Phillip Bay, in southeastern Australia, lengthening the existing digital record by almost 100 years. After removing the trend due to SLR, a small negative trend in extreme sea levels was found. Looking at the components separately, a small negative trend in storm surges was found, consistent with the observed poleward movement of the weather systems that cause them. A positive trend in tidal range was found, possibly due to more tidal in and outflow through the narrow entrance to the bay under SLR. The net effect of SLR and storm‐tide trends is an increasing trend in extreme sea‐levels. This information is relevant for coastal managers considering coastal adaptation options. Key Points: Sub‐daily data for the Williamstown, Australia tide gauge has been extended back to 1872 through digitization of paper recordsTrends in extreme sea level events show a small declining trend over the entire record although trends in tidal range are positive over recent decades [ABSTRACT FROM AUTHOR]