Supporting information for “Decadal evolution of ice-ocean interactions at a large East Greenland glacier resolved at fjord scale with downscaled ocean models and observations

As given in the main document (URS324295): In recent decades, the Greenland ice sheet has been losing mass through glacier retreat and ice flow acceleration. This mass loss is linked with variations in submarine melt, yet existing ocean models are either coarse global simulations focused on decadal‐scale variability or fine‐scale simulations for process‐based investigations. Here, we unite these scales with a framework to downscale from a global state estimate (15 km) into a regional model (3 km) that resolves circulation on the continental shelf. We further downscale into a fjord‐ scale model (500 m) that resolves circulation inside fjords and quantifies melt. We demonstrate this approach in Scoresby Sund, East Greenland, and find that interannual variations in submarine melt at Daugaard‐Jensen glacier induced by ocean temperature variability are consistent with the decadal changes in glacier ice dynamics. This study provides a framework by which coarse‐resolution models can be used to quantify glacier submarine melt for future ice sheet projections.

如主文档（URS324295）所述：近数十年来，格陵兰冰盖因冰川退缩与冰流加速持续发生物质亏损。该物质亏损与海底消融（submarine melt）的变化密切相关，但现有海洋模式要么是聚焦于年代际尺度变率的粗分辨率全球模拟，要么是面向过程机理研究的精细化尺度模拟。本研究通过一套降尺度框架整合两类尺度的模拟：先将15公里分辨率的全球状态估计产品降尺度至3公里分辨率的区域模式，该模式可解析大陆架区域的环流特征。在此基础上，本研究进一步开展降尺度推演，构建500米分辨率的峡湾尺度模式，该模式可解析峡湾内部环流并量化冰川消融量。我们在格陵兰东部的斯科斯比松峡湾（Scoresby Sund）验证了该方法，发现多亚德-延森冰川（Daugaard-Jensen Glacier）的海底消融年际变化由海洋温度变率驱动，且与冰川冰动力学的年代际变化相一致。本研究构建的这套框架，可利用粗分辨率模式量化冰川海底消融量，为未来冰盖预测研究提供支撑。