Satellite gravity and the mass balance of the Antarctic ice sheet

Changes in the Earth's gravity field with time have important applications to a broad range of disciplines. Any process that involves a large enough h orizontal redistribution of mass, either within the Earth or on or above it s surface, is potentially detectable. In particular, when ice sheets: grow or shrink, gravity changes as mass is redistributed in the solid earth and between the oceans and the ice sheets. The sources of global sea-level rise (about 2 mm a(-1) over the last century) and in particular the contributio n of the Antarctic ice sheet thereto are not well understood. Gravity measu rements can help to diminish this uncertainty. The technology currently exists to measure gravity with high accuracy by a dual-satellite mission in which the distance between the satellites is prec isely monitored. We estimate from recent studies that temporal changes in t he gravity. field as determined by a satellite gravity mission lasting 5 ye ars at an orbital height of 400 km would be sensitive to changes in the ove rall mass of the Antarctic ice sheet to a precision corresponding to better than 0.01 mm a(-1) of sea-level change. However, the effects of three othe r phenomena that could each produce a temporally varying gravity signal wit h characteristics comparable to that caused by a change in Antarctic ice - postglacial rebound, inter-annual variability in snowfall, and atmospheric pressure trends - also need to be evaluated. Postglacial rebound could be p altry separated from ice-mass changes with the aid of global positioning sy stem campaigns and numerical models of rebound that use improved determinat ions of mantle viscosity also provided by the gravity mission. Determinatio n of inter-annual ice-mass changes will be aided by measurements of moistur e-flux divergence around the perimeters of the ice sheets and direct observ ations of inter-annual changes by the gravity satellite itself The removal of pressure effects over Antarctica will become more effective as the numbe r of automatic weather stations in the interior of the continent increases. Even after, corrections are made for these factors, the uncertainties they cause limit the accuracy in the determination of the contribution of the An tarctic ice sheet to sea-level change to about 0.5 mm a(-1). However, there is a strong complementarity between gravity measurements and the surface-h eight measurements that will be produced by NASA's laser altimeter mission early next century. Together, they should be able to determine that contrib ution to an accuracy of about 0.1 mm a(-1).