Detection of lake drainage events in Antarctica from SAR imagery

Supraglacial lakes are expected to play a crucial role in determining the ice sheet mass balance in a warming climate. Water ponding lowers the albedo of the ice surface, establishing a positive feedback of melting processes that might be further enhanced by the projected rising temperatures. Lake drainages can have particularly large impacts on ice shelves depending on their location and surrounding topography. Drainage events on grounded ice can transport water to the ice/bedrock interface, affecting the sliding of the ice sheet. On floating ice shelves, lake drainage events have been linked to fracture formation potentially leading to ice shelf collapse.

Over the past decade, observations of supraglacial lake drainage events have mainly been gathered from the Greenland ice sheet, while observations of drainage events remain rare in Antarctica. While some limited examples have been reported in the literature, it is not yet known how common these events are, the likelihood of their formation from the grounding line, and how their recurrence could impact Antarctic ice shelves. Observations of Antarctic supraglacial lake drainages are challenging as the lakes often have lids of ice covering liquid water, and drainages can occur in winter when low light levels preclude the use of optical sensors. Since Sentinel-1 SAR imagery works independently from light and cloud conditions, it enables continuous monitoring throughout the year providing further insights into their spatial and temporal evolution. The use of Google Earth Engine (GEE) platform for analyzing SAR images and detecting drainage events has shown the value of this platform as a tool to monitor changes over several locations and to efficiently deal with the increasing workload of satellite data. Here we demonstrate the use of SAR backscatter to reliably detect drainage events to map their location also during the winter months and to locate their prevalence around the Antarctic coastline.