Detection of supraglacial lakes and melt ponds using airborne laser altimetry

Supraglacial lakes and melt ponds occur frequently over Alaska and Greenland glaciers. The presence of water on the ice produces a positive feedback loop as a consequence of albedo reduction, leading to further ice melt. Furthermore, lakes and ponds are important drivers of hydrofracture, or crevassing via water pressure, which may lead to bed lubrication or enhanced ice shelf breakup. Therefore, supraglacial lake and melt pond detection is important for understanding glacier dynamics, despite limited efforts in remote sensing studies. In this study, we use altimetry data from the Multiple Altimeter Beam Experimental Lidar (MABEL) and the Airborne Topographic Mapper (ATM) campaigns to detect lakes over Alaska and Greenland. To validate lake detection from lidar data, reflectance products from MODIS, Landsat and MABEL camera imagery were also analyzed. Initial results are promising, with both MABEL and ATM detecting multiple lakes during the chosen flights. However, lake signatures differ somewhat between the instruments. In MABEL, high-power photons reflect off water surfaces, and lower-powered photons reach the bottom of sufficiently shallow lakes. The ATM photons, while lacking a classification scheme, profile both lake beds and surfaces clearly. Further analysis will incorporate the development of a lake depth retrieval algorithm tuned to the signatures of both instruments, as well as a Monte Carlo model accounting for lake impurities. This project is in preparation for the launch of ICESat-2, so future analysis will also consider the satellite's observations.