Using ICESat-2 and Operation IceBridge altimetry for supraglacial lake depth retrievals

Abstract

Supraglacial lakes and melt ponds occur in the ablation zones of Antarctica and Greenland during the summer months. Detection of lake extent, depth, and temporal evolution is important for understanding glacier dynamics. Previous remote sensing observations of lake depth are limited to estimates from passive satellite imagery, which has inherent uncertainties, and there is little ground truth available. In this study, we use laser altimetry data from the Ice, Cloud, and land Elevation Satellite-2 (ICESat-2) over the Antarctic and Greenland ablation zones and the Airborne Topographic Mapper (ATM) for Hiawatha Glacier (Greenland) to demonstrate retrievals of supraglacial lake depth. Using an algorithm to separate lake surfaces and beds, we present case studies for 12 supraglacial lakes with the ATM lidar and 12 lakes with ICESat-2. Both lidars reliably detect bottom returns for lake beds as deep as 7 m. Lake bed uncertainties for these retrievals are 0.05–0.20 m for ATM and 0.12–0.80 m for ICESat-2, with the highest uncertainties observed for lakes deeper than 4 m. The bimodal nature of lake returns means that high-confidence photons are often insufficient to fully profile lakes, so lower confidence and buffer photons are required to view the lake bed. Despite challenges in automation, the altimeter results are promising, and we expect them to serve as a benchmark for future studies of surface meltwater depths.