Snow accumulation variability and random walk:: how to interpret changes of surface elevation in Antarctica

Because snow accumulation above an ice sheet is a cumulative process, a random fluctuation of snow accumulation can lead to significant variability in ice sheet mass or elevation, without being linked to a long-term climatic change. Moreover, ice sheet mass balance cannot be measured directly. We can determine the ice sheet surface elevation through altimetric data but, due to the densification process, the snow accumulation variability affects the surface elevation variations more than the mass variations. We investigated the effect of the recent snow accumulation variability on both ice sheet mass and volume at the time scale of the satellite era. There is more than 10% chance of measuring an artificial trend greater than 15% of the mean accumulation rate, from a 10-year elevation series. In our simplified model, the impact of snow accumulation variability only depends on the ratio between the snow densification time and the variability time period: It decreases with the densification rate and increases with the variability frequency, so that the induced error is maximum for high-frequency noise in the central part of Antarctica. Finally, our results showed that knowledge of snow accumulation variability for the past 5-10 years is needed to eliminate the delayed induced error. (C) 2004 Elsevier B.V. All rights reserved.