A comparison of historical and recent sea level measurements at Port Arthur, Tasmania

Estimates of anticipated sea level rise as a consequence of 'greenhouse' warming depend both on the increased global temperatures, and on the way in which this heat and the water formed from melting ice are absorbed in the global ocean (Church et a/, 2001). Various numerical models of ocean responses have shown that the increase in sea level will not be uniform worldwide; the validity of these models can be confirmed if these regional differences are consistent with direct observations. For this, measurements of actual mean sea level changes over long periods are needed. Unfortunately, very few early sea level measurements have survived, especially in the Southern Hemisphere. A unique series of sea level measurements, made at Port Arthur, Tasmania, between 1837 and 1842, and linked to a benchmark which still exists, has been used to estimate sea level changes in the region over the past 160 years. The estimated rise (0.13 +/- 0.03 metres) gives an average rate of increase of 0.8 +/- 0.2 mm/year. Correction for local vertical land movement, based on best available models and observations, increases this value by about 0.0 to 0.2 mm/year. This rate of increase is significantly less than the observed globally averaged mean sea level rise over the same period (Church et al, 2001). However, it is broadly consistent with the reduced rate of rise at high southern latitudes shown by numerical models (Gregory et al, 2001). A major uncertainty arises in the estimate of recent vertical land movement, which will be resolved only when our geocentric measurements of the benchmark coordinates are repeated after a sufficiently long interval.