Mass losses of the Antarctic Peninsula. New constraints from stereoscopic imagery and laser altimetry.

Along with glaciers, polar ice sheets are a major contributor to sea level rise and their losses are accelerating. Since 2012, intercomparison exercises have combined estimates of ice sheet mass change from various methods (gravimetry, altimetry and input/output method). However, the consensus displayed in these intercomparisons hides sometimes strong divergences between these different methods because each one presents drawbacks. In particular, the altimetry method, whether based on radar or laser measurements, has a resolution of generally one kilometer. This resolution, although perfectly suited in the central and flat areas of the polar ice sheets, does not allow to solve the complexity of the elevation changes of the coastal glaciers, especially along the sloping coasts of the Antarctic Peninsula and Greenland. Yet, it is at their margins that ice sheets respond dynamically to rising atmospheric and oceanic temperatures.

The objective of the study is to build high resolution estimates of ice sheet elevation changes. It exploits an archive of stereo pairs acquired by the SPOT5-HRS sensor mostly during the International Polar Year (IPY, 2007-2009) to build a topography of the polar ice sheet periphery. A vertical correction of each digital terrain model (DEMs) is performed using the elevation measurements, partly simultaneous, of the ICESat laser altimeter (2003-2008). This IPY topography is then used as a reference to estimate more than 15 years of volume changes of the ice sheet periphery by comparison with measurements from recent missions, in particular ICESat-2 and REMA (Reference Elevation Model of Antarctica) /ArcticDEM.

The Antarctic Peninsula was selected to develop the methodology and to estimate 15 years of evolution. This is one of the regions where recent estimates of mass loss diverge the most and where glacier dynamics are complex. The elevation change maps reveal, at a high resolution, the spatial pattern of changes over the past 15 years.