Timing and regional patterns of snowmelt on Antarctic sea ice from passive microwave satellite observations

Tuesday, 15 December 2015
Poster Hall (Moscone South)
Stefanie Arndt1, Marcel Nicolaus1, Sascha Willmes2 and Wolfgang Dierking1, (1)Alfred Wegener Institute Helmholtz-Center for Polar and Marine Research Bremerhaven, Bremerhaven, Germany, (2)Organization Not Listed, Washington, DC, United States
The timing and regional distribution of surface properties of Antarctic sea ice is crucial for the atmosphere-ocean interaction and characterizes the mass and energy budgets of sea ice. Therefore, it is important to map and analyze changes and trends of the related processes and parameters. Since Antarctic sea ice is covered with snow during most of the year, inter-annual and regional variations in summer surface melt can be described through the timing of snowmelt onset. So far, the melt onset was described through the amplitude of diurnal freeze-thaw cycles detected by microwave brightness temperatures using a fixed threshold. However, other studies reveal that the strength of the diurnal variations is differing between the perennial snowpack characterized by strong snow metamorphism and the thinner and less complex seasonal snow cover.

Therefore, we present two complementary approaches to improve the existing melt onset algorithms: (1) We consider regional differences of the diurnal variations in the brightness temperature. (2) We combine brightness temperature measured at different polarizations and frequencies in order to describe also subsurface melt processes. Our analysis includes a comparison with autonomous measurements from snow buoys and previous studies on snow melt onset detection of Antarctic sea ice.

In doing so, we derive a distinct latitudinal dependence of the surface and subsurface snow melt onset. The major part of the East-Antarctic sea ice is dominated by lateral and bottom melt with negligible diurnal surface variations. Although a positive trend in sea-ice extent and concentration of Antarctic sea ice is observed, our melt onset time series do not indicate a significant trend from 1988/89 to 2014/15. Instead its inter-annual variability is not changing over time. From the assumed dynamically induced sea-ice growth in the Southern Ocean we expect an increasing importance of surface freeze-thaw cycles.