Blowing snow detection from ground-based ceilometers: application to East Antarctica

• Published in: The cryosphere Vol. 11; no. 6; pp. 2755 - 2772
• Main Authors: Gossart, Alexandra, Souverijns, Niels, Gorodetskaya, Irina V, Lhermitte, Stef, Lenaerts, Jan T. M, Schween, Jan H, Mangold, Alexander, Laffineur, Quentin, van Lipzig, Nicole P. M
• Format: Journal Article
• Language: English
• Published: Katlenburg-Lindau Copernicus GmbH 05.12.2017; Copernicus Publications
• Subjects: Algorithms; Animal attacks; Annual variations; Antarctic ice sheet; Atmospheric models; Availability; Bears; Blowing rate; Blowing snow; Blowing time; Ceilometers; Clouds; Detection; Distribution; Environmental conditions; Glaciation; Ice sheets; Interannual variability; Lasers; Mass balance; Mass balance of ice sheets; Mathematical models; Numerical models; Observations; Polar environments; Polar regions; Precipitation; Remote sensing; Satellites; Snow; Storms; Sublimation; Synoptic conditions; Visual observation; Wind; Wind speed; Antarctica
• ISSN: 1994-0424; 1994-0416; 1994-0424; 1994-0416
• DOI: 10.5194/tc-11-2755-2017

Blowing snow impacts Antarctic ice sheet surface mass balance by snow redistribution and sublimation. However, numerical models poorly represent blowing snow processes, while direct observations are limited in space and time. Satellite retrieval of blowing snow is hindered by clouds and only the strongest events are considered. Here, we develop a blowing snow detection (BSD) algorithm for ground-based remote-sensing ceilometers in polar regions and apply it to ceilometers at Neumayer III and Princess Elisabeth (PE) stations, East Antarctica. The algorithm is able to detect (heavy) blowing snow layers reaching 30 m height. Results show that 78 % of the detected events are in agreement with visual observations at Neumayer III station. The BSD algorithm detects heavy blowing snow 36 % of the time at Neumayer (2011–2015) and 13 % at PE station (2010–2016). Blowing snow occurrence peaks during the austral winter and shows around 5 % interannual variability. The BSD algorithm is capable of detecting blowing snow both lifted from the ground and occurring during precipitation, which is an added value since results indicate that 92 % of the blowing snow is during synoptic events, often combined with precipitation. Analysis of atmospheric meteorological variables shows that blowing snow occurrence strongly depends on fresh snow availability in addition to wind speed. This finding challenges the commonly used parametrizations, where the threshold for snow particles to be lifted is a function of wind speed only. Blowing snow occurs predominantly during storms and overcast conditions, shortly after precipitation events, and can reach up to 1300 m a. g. l.  in the case of heavy mixed events (precipitation and blowing snow together). These results suggest that synoptic conditions play an important role in generating blowing snow events and that fresh snow availability should be considered in determining the blowing snow onset.