Supraglacial Drainage Efficiency of the Greenland Ice Sheet Estimated From Remote Sensing and Climate Models

Abstract

AbstractSupraglacial stream/river catchments drain large volumes of surface meltwater off the southwestern Greenland Ice Sheet surface. Previous studies note a strong seasonal evolution of their drainage density (Dd), a classic measure of drainage efficiency defined as open channel length per unit catchment area, but a direct correlation between Dd and surface meltwater runoff (R) has not been established. We use 27 high‐resolution (∼0.5 m) satellite images to map seasonally evolving Dd for four GrIS supraglacial catchments, with elevations ranging from 1,100 m to 1,700 m. We find a positive linear correlation (r2 = 0.70, p < 0.01) between Dd and simulations of runoff production from two climate models (MAR v3.11 and MERRA‐2). Applying this R‐Dd empirical relationship to climate model output enables parameterization of spatial and temporal changes in supraglacial drainage efficiency continuously throughout the melt season, although temporal and spatial skewness of Dd observations likely affects the application of this R‐Dd relationship on crevasse fields and snow/firn surfaces. Incorporating this information into a simple surface routing model finds that high runoff leads to earlier, larger diurnal peaks of runoff transport on the ice surface, owing to increased Dd. This effect progressively declines from low (∼1,100 m) to high (∼1,700 m) elevation, causing a roughly order‐of‐magnitude reduction in diurnal runoff variability at the highest elevations relative to standard climate model output. Combining intermittent satellite Dd mapping with climate model output thus promises to improve characterization of supraglacial drainage efficiency to the benefit of supraglacial meltwater routing and subglacial hydrology models.