The initial analysis of the GEOS-3 satellite radar-altimeter data showed an accuracy of about two meters (Brooks and others, 1978). The major residual uncertainty was due to inaccuracies in the determination of the satellite orbital position. This uncertainty has been reduced by minimization of differences in the elevations observed at intersections of the satellite-nadir tracks. By using only intersections over the adjacent ocean, a solution is obtained that is independent of any actual elevation changes during the time intervals between the intersection of the tracks. The improved accuracy thus obtained has made it possible to analyze the data for elevation changes that might have occurred during the three-year period of data collection by GEOS-3.
A topographic map having a contour interval of 10 m has been constructed for most of the Greenland ice sheet south of lat. 65° N. The present results have not been corrected for the effect of surface slope (Robin, 1966; Brooks and others, 1978). Therefore, the true surface may lie below the indicated surface by as much as ∆h = ha
/2 (e.g. ∆h = 4.7 m for surface slope α = 1/300 and satellite altitude h = 844 km). Since the altimeter measurement is the distance to the mean surface within the altimeter footprint (footprint diameter is 3.6 km), slopes or undulations having horizontal extent small compared to the footprint have little effect. Correction for the effect of slopes having a larger horizontal scale can be made provided the data set is sufficiently dense.
Surface undulations or waves of various amplitudes and wavelengths are observed on the ice sheet. Over much of the surface, waves on the order of 10 m amplitude and 10 km in wavelength appear to be predominant. These observed wavelengths are about six times the ice thickness, which is consistent with Budd's (1969) conclusion that surface undulations having wavelengths 2 to 10 times the ice thickness should be predominant with the shorter and longer wavelengths generated by bedrock irregularities being more severely damped within the ice. The surface topography is also observed to be more irregular near the southernmos portion of the ice sheet due to the more irregular bedrock topography.
Brooks, R. L., and others. 1978. Ice sheet topography by satellite altimetry, [by] Brooks, R. L.
Campbell, W. J.
Ramseier, R. O.
Stanley, H. R. and Zwally, H. J.
Nature, Vol. 274, No. 5671, p. 539–43.
Budd, W. F.
1969. The dynamics of ice masses. ANARE Scientific Reports. Ser. A(IV). Glaciology. Publication No. 108.
Robin, G. de Q.
1966. Mapping the Antarctic ice sheet by satellite altimetry. anadian Journal of Earth Sciences, Vol. 3, No. 6, p. 893–901.
T. J. Hughes: The altitude changes you might detect over a decade or so by this method can reflect either mass-balance changes or snow-density changes related to climatic warming or cooling over the decade. Can other satellite remote-sensing methods allow you to distinguish between these possibilities.
H. J. Zwally: Not directly, but microwave techniques can provide information on firn grain size, accumulation, and temperature which may be related to firn compaction. However, understanding of the magnitude of possible firn compaction effects due to climatic temperature changes, or accumulation changes, requires additional field and theoretical studies of firn compaction processes.
C. R. Bentley: How many satellite crossing lines are there in the large-scale (10 m contour line) map of surface elevations around the dome area?
Zwally: This preliminary map includes only about 12 lines. Therefore, some of the smaller features are probably artificial.