Abstract text
|
There is growing interest in surface water bodies across Antarctic ice shelves as they impact the ice shelf mass balance. The filling and draining of lakes have the potential to flex and fracture ice shelves, which may even lead to their catastrophic break up. The study of ice shelf surface lakes typically uses optical satellite imagery to delineate their areas and a parameterised physically based light attenuation algorithm to calculate their depths. This approach has been used to calculate ponded water volumes and their changes over seasonal and inter annual timescales. The approach has been developed and validated using various in-situ data sets collected on the Greenland Ice Sheet, but so far, the approach has not been validated for Antarctic ice shelves. Here we use simultaneous field measurements of lake water depths made using water pressure sensors, and surface spectral properties made with fixed four channel radiometers (red, blue, green, panchromatic), to parameterise the light attenuation algorithm for use during the filling and draining of shallow surface lakes on the McMurdo Ice Shelf, Ross Sea Sector, Antarctica during the 2016/17 summer. We then apply the approach to calculate lake areas, depths and volumes across several surface water bodies observed in high resolution Worldview imagery and their changes over time. These calculations are used, in turn, to help validate the approach to calculating water volumes across the entire ice shelf using Sentinel-2 and Landsat 8 imagery. Results suggest that using parameter values relevant to the Greenland Ice Sheet may bias the calculation of water volumes when applied to Antarctic ice shelves, and we offer values that may be more appropriate. Furthermore, calculations of lake volume using Sentinel-2 and Landsat 8 imagery maybe underestimated when compared to the higher resolution Worldview imagery. The findings have implications for the calculation of water volumes across other ice shelves.
|