ICESat-2 derived discharge estimates of Petermann meltwater stream

Abstract

Mass loss from the Greenland Ice Sheet due to runoff from surface ablation and ice- berg calving contributes to global sea level rise. Throughout annual melt seasons, a complex hydrological system comprising supraglacial meltwater ponds, lakes, streams, rivers, and moulins develops on top of the ice sheet. This supraglacial river and stream network is significant for the transport and removal of meltwater off the surface of the Ice Sheet towards the ocean and influencing the stability of ice shelves. The Petermann glacier in North-Western Greenland is a marine-terminating out- let glacier with a floating ice tongue with a prominently developed meltwater stream forming in the centre of the ice shelf. For a better understanding about the spatiotem- poral dynamics of the Petermann meltwater stream, a method was developed that uses ICESat-2 ATL03 photon cloud data to derive river discharge estimates using the Manning-Strickler equation for open-channel flow. The method is based on manu- ally delineated surface and bottom of the Petermann meltwater stream to derive river width, cross-section area, wetted perimeter, as well as the hydraulic radius of the channel. ArcticDEM of 100m resolution was used to derive average slope val- ues in flow direction completing the parameters feeding into the Manning-Strickler equation. Flux estimates are derived for the years of 2020 and 2023, as those two years were characterized by rather low and high melt anomalies, respectively. Estimated discharge values of the Petermann meltwater stream located on the frontal part of the ice tongue obtained by two nearby ground-tracks with Manning’s roughness co- efficient n of 0.3 are 271 m3/s and 55 m3/s, respectively. That shows that the derived flux estimates of the two beams (one weak and one strong beam) of one ground-track pair can differ considerably, inducing uncertainty into the manual stream-picking method. Future attention needs to be paid to reducing the uncertainties feeding into the Manning-Strickler equation for receiving more accurate flux data. For validation of manually stream-picked data, the Watta-algorithm can be used to automatically delineate surface and bottom of supraglacial melt-features and implement these data into the proposed method for retrieving discharge estimates. The algorithm was developed and applied for melt lakes by Datta & Wouters (2021) but can also be applied and tuned for streams and rivers and data can be implemented into the proposed method for retrieving discharge estimates.