EFFECTS OF POLAR ICE ON THE EARTHS ROTATION AND GRAVITATIONAL POTENTIAL

Gridded values of net surface accumulation rates for the Antarctic and Green an ice sheets are used to estimate the contributions these two regions make to the Earth's gravity and rotation. The ice discharge from both ice sheets is adjusted so that the estimates of sea-level contribution from both ice sheets fall within limits for present mass balance suggested by the Committee on Glaciology (1985). These sea-level contributions are, for Antarctica, -0.6 +/- 0.6 mm yr-1, and for Greenland, -0.1 +/- 0.4 mm yr-1. The contributions to the low-order zonal harmonic coefficients of the Earth's gravitational potential from Antarctica are found to be between two and 10 times larger than the uncertainties of the zonal harmonics derived from satellite solutions. For Greenland, the coefficients are within an order of magnitude of the uncertainties of the satellite solutions. These results are not definitive. They are presented to demonstrate that the ice sheets contribute to gravity in amounts that are within the limits of detection by satellite ranging methods. The secular trends in the X and Y components of observed polar-motion excitation agree in magnitude and sign with the Antarctic contributions, for a model where the interior regions on the ice sheet undergo a net thinning and the coastal regions thicken. For this model, the boundary between the interior and coastal regions is set at the contour of constant accumulation of 0.28 m yr-1 water equivalent, and the magnitude of the thickness change is set equal to the local accumulation rate. The contributions from Greenland to the secular trends in polar-motion excitation and length-of-day are either smaller than or of opposite sign to the observed effects and are found to be smaller than those of Antarctica. The limited information on mass balance and ice discharge including calving of ice and refreezing of melt-water for both ice sheets and the lack of measured data for land-based water storage makes the results presented here highly speculative. In addition, the inelastic response of the Earth to changes in surface mass loading is not considered here. This delayed response can reduce the magnitude of the effects presented here by as much as 15 per cent if the loading is a linear trend spanning 100 yr and the correction can be larger for longer histories of loading, see Wahr et al. (1992). These results suggest very approximate upper and lower limits for both inter-annual and secular contributions of polar-ice mass balance to Earth rotation and gravity. These limits are useful where they fall within the range of detectability using satellites.