MODELING GLOBAL SEA LEVEL CHANGE RESULTING FROM GLOBAL WARMING

If global warming is occurring and ice sheets are melting at an increased rate, average global sea level will rise at a rate of 2 to 5 mm/yr. However, measurement of this sea level rise is complicated because the rise will not be geographically uniform. Changing ice and meltwater loads, both past and present, cause deformation of the earth’s solid surface and geoid so that every ocean location will experience a different sea level signature. Therefore, we developed a model of glacial isostatic adjustment on a visco-elastic earth to assess modern sea level change and global warming.

Our model calculates sea level effects from changing ice and water loads on the earth’s surface from the last glacial maximum to present, using 137 ice grids and 454 ocean grids. The model is capable of both forward and inverse approaches. In the forward method, we assume a worldwide glacial history and an earth viscosity structure are known. The model then calculates global sea levels that can be compared to observations for the past 17,000 years. The inverse calculation does the opposite: observed sea level data, both past and present, is used to determine an ice sheet history that best fits the data in the least squares sense. Using tide gauge records, proglacial and postglacial lake shorelines, and ocean shoreline data, we calculated an optimum ice sheet with a maximum thickness of 3062 meters over southern Hudson Bay 17,000 years ago.

For the past 11 years the TOPEX/Poseidon satellite has provided highly accurate (to 2.5 cm) radar altimetry data of sea level height over the majority of the ice-free oceans every 10 days. We can use this data in the inverse model and a preliminary inverse calculation indicates that the Greenland ice sheet is thinning by 21 cm/yr. We also summarized the satellite data using empirical orthogonal functions, which decompose the data into orthogonal eigenvectors, or trends, that efficiently summarize the total variance in the data. The first eigenvector, responsible for 27% of the total variance, indicates that global sea level is rising at a rate of 2.5 mm/yr. The spatial pattern of sea level change for this eigenvector is similar to that expected for the case in which the Greenland ice sheet thins at a rate of 25 cm/yr. Hence the global sea level signature is consistent with the global warming hypothesis.