Evidence of anthropogenic climate effects in snow and firn of East Antarctica? Characterization of low accumulation areas using multiparameter-analysis from snow and firn cores

Abstract

Antarctic ice masses are a unique climate archive, but are also strongly affected by global climate change. Field data from the East Antarctic Plateau are necessary for both, studying the signal formation of climate proxies and validating results from remote sensing. This work provides data and analysis of snow cores, sampled on a traverse between Kohnen Station and Plateau Station during the Antarctic summer of 2016/17. X-ray CT was used to determine the density and stratigraphic properties, then the cores were analyzed for stable water isotopes and major ions. Multiple snow cores per sampling location allow a more representative determination of the investigated parameters. Along the traverse route, the mean surface snow density is 355 kg m-3 and shows a lower dependence on temperature and accumulation rate than assumed. Modeled values show a significant discrepancy of about -10% from the measured density. In this work, the first dataset on the spatial distribution of crusts in polar snow is presented. Contrary to the assumption of finding more crusts in locations with lower accumulation rates, the total number of crusts per meter decreases with decreasing accumulation. The results suggest a relationship between the number of crusts and the logarithmic accumulation rate. Cycles in δ18O around Kohnen Station can still be interpreted as seasonal signals, but below an accumulation rate of 50 kg m-2a-1 they are no longer suitable for dating the snowpack on short time scales. A comparison with ECHAM6-wiso validates the model trend along the traverse, but shows a constant offset in δ18O. Modeled snow profiles with precipitation values from ECHAM6-wiso and a diffusion model represent the measured profiles well at 1-2 m depth. However, at the very surface, redeposition and sublimation appear to contribute significantly to (postdepositionally) shaping of the δ18O signal. The change in surface snow density between samples from 2005/06 and samples from this study can be attributed to different volume errors in sampling, but mean δ18O values over the recent decades show an increasing trend. While these values are within the range of natural variability, they may be early indications that climate proxies in the snow of the East Antarctic Plateau have already recorded the increase of the global temperature.