Decreasing surface albedo signifies a growing importance of clouds for Greenland Ice Sheet meltwater production.

Clouds regulate the Greenland Ice Sheet's surface energy balance through the competing effects of shortwave radiation shading and longwave radiation trapping. However, the relative importance of these effects within Greenland's narrow ablation zone, where nearly all meltwater runoff is produced, remains poorly quantified. Here we use machine learning to merge MODIS, CloudSat, and CALIPSO satellite observations to produce a high-resolution cloud radiative effect product. For the period 2003-2020, we find that a 1% change in cloudiness has little effect (±0.16 W m) on summer net radiative fluxes in the ablation zone because the warming and cooling effects of clouds compensate. However, by 2100 (SSP5-8.5 scenario), radiative fluxes in the ablation zone will become more than twice as sensitive (±0.39 W m) to changes in cloudiness due to reduced surface albedo. Accurate representation of clouds will therefore become increasingly important for forecasting the Greenland Ice Sheet's contribution to global sea-level rise.