These datasets accompany the paper titled "Decreasing surface albedo signifies a growing importance of clouds for Greenland Ice Sheet meltwater production". We used machine learning to merge accurate Cloudsat/CALIPSO cloud profiling retrievals with high temporal and spatial resolution MODIS visible-infrared imagery, enabling retrieval and investigation of fine-scale cloud radiative effects (CRE) across the Greenland Ice Sheet surface from 2003 to 2020. The high spatial resolution (1 x 1 km) of our CRE product represents a major advance for understanding interactions between clouds and specific areas of the ice sheet surface.