Using variable-resolution grids to model precipitation from atmospheric rivers around the Greenland ice sheet

Abstract. Atmospheric rivers (ARs) are synoptic-scale features that transport moisture poleward and may cause short-duration, high-volume melt events on the Greenland ice sheet (GrIS). In contrast with traditional climate modeling studies that rely on coarse (1 to 2°) grids, this project investigates the effectiveness of variable-resolution (VR) grids in modeling ARs and their subsequent precipitation using refined grid spacing (0.25 and 0.125°) around the GrIS and 1° grid spacing for the rest of the globe in a coupled land–atmosphere model simulation. VR simulations from the Community Earth System Model version 2.2 (CESM2.2) bridge the gap between the limitations of global and regional climate models while maximizing computational efficiency. ARs from CESM2.2 simulations using three grid types (VR, latitude–longitude, and quasi-uniform) with varying resolutions are compared to outputs from two observation-based reanalysis products, ERA5 and the Modern-Era Retrospective Analysis for Research and Applications, version 2 (MERRA-2), using a study period of 1 January 1979 to 31 December 1998. The VR grids produce ARs with smaller areal extents and lower area-integrated precipitation over the GrIS compared to latitude–longitude and quasi-uniform grids. We hypothesize that the smaller areal AR extents in VR grids are due to the refined topography resolved in these grids. In contrast, topographic smoothing in coarser-resolution latitude–longitude and quasi-uniform grids allows ARs to penetrate further inland on the GrIS. Precipitation rates are similar for the VR, latitude–longitude, and quasi-uniform grids; thus the reduced areal extent in VR grids produces lower area-integrated precipitation. The VR grids most closely match the AR overlap extent and precipitation in ERA5 and MERRA-2, suggesting the most realistic behavior among the three configurations.