1. Rapid Thinning of Parts of the Southern Greenland Ice Sheet
2. Instrument performance was checked each flight by calibration measurements while on the ground and surveys of calibration sites. Comparison of derived ice-surface elevations at all flight-line crossing points for the same year gives rms differences of about 10 cm (3) and some of these differences represent real changes during the 3 to 4 weeks of the survey period. However assuming these differences to be totally caused by errors that are systematic to a flight all elevation changes inferred along any one flight line could include a bias of 10 cm or 2 cm/year which is independent of the biases for other flight lines. Interpolation between flight lines increases the error but this is offset to some extent by the inclusion of many flight lines and also the independence of errors in different parts of the ice sheet. Over higher elevation parts of the ice sheet dH / dt has low spatial variability and there are many flight lines so interpolated values should have similar errors to those measured along flight lines. Consequently errors for dH / dt averaged over large areas of higher elevation ice sheet are approximately ±2/ N 0.5 cm where N is the number of flight lines over the region. Nearer the coast spatial variability increases and interpolation errors can predominate. Errors that are systematic to an entire year's survey are caused mainly by uncertainty (of about 1 cm) in the location of GPS base stations introducing a possible bias to estimates of dH / dt of 0.3 cm/year.
3. Krabill W., Thomas R. H., Martin C. F., Swift R. N., Frederick E. B., Int. J. Remote Sens. 16, 1211 (1995).
4. Bromwich D. H., Chen Q-s., Li Y., Cullather R. I., J. Geophys. Res. 104, 22103 (1999).
5. McConnell J. R., Mosley-Thompson E., Bromwich D. H., Bales R. C., Kyne J. D., J. Geophys. Res. 105, 4039 (2000).