Regional Patterns of Blowing Snow Dynamics on the Antarctic Ice Sheet from the Modèle Atmosphérique Régionale (MAR), Assessed with In Situ and Remote Sensing data (2000-2011)

Abstract:

The erosion and sublimation of blowing snow in Antarctica is a considerable source of error in surface mass balance estimates of the Antarctic Ice sheet. These processes are also crucial to an understanding of near-surface atmospheric processes at the margins. Because ground measurements of blowing snow are difficult to capture, regional climate models (RCMs) are crucial to an understanding of the phenomenon. Here, we present a comparison of outputs of two implementations of the Modèle Atmosphérique Régionale (MAR) RCM at a 50 km resolution over Antarctica for the period 2000-2011. While previous versions of MAR have been run for time periods beginning in 1958, this abbreviated run was performed specifically to test the blowing snow routine. The model is forced at the boundaries with 6-hourly reanalysis data provided by the European Centre for Medium-Range Weather Forecasts (ECMWF). The two MAR configurations are identical except for the implementation of blowing snow dynamics. The differences in surface and atmospheric outputs between the models can therefore isolate the seasonal and regional effects of blowing snow. We determine spatio-temporal patterns of covariance between variables over the entire 11-year period, as well as inter-seasonal trends within each year using Empirical Orthogonal Function (EOF) analysis. The most prominent of these patterns pinpoints specific regions and time periods for model outputs which are compared with available temperature, surface pressure, wind speed and relative humidity data from Automatic Weather Stations as well as with radiosonde data generally available at the ice sheet margins (made available by the University of Wisconsin-Madison). Additionally, model outputs in selected areas are compared with satellite data (MODIS surface temperature and pressure) to assess the robustness of these trends. The analysis shows that there are specific regions and weeks in the winter where increased blowing snow flux corresponds to significant humidity differences between the two models.