Satellite observations of snowfall regimes over the Greenland Ice Sheet

Abstract

The mass of the Greenland Ice Sheet (GrIS) is decreasing due to increasing surface melt and ice dynamics. Snowfall both adds mass to the GrIS and has the capacity to reduce surface melt by increasing surface brightness, reflecting additional solar radiation back to space. Modeling the GrIS's current and future mass balance and potential contribution to future sea level rise requires reliable observational benchmarks for current snowfall accumulation as well as robust connections between individual snowfall events and the large-scale atmospheric circulation patterns that produce them. Previous work using ground-based observations showed that, for one research station on the GrIS, two distinct snowfall regimes exist: those associated with exclusively ice-phase cloud processes (IC) and those involving mixed-phase processes indicated by the presence of supercooled liquid water (CLW). The two regimes have markedly different accumulation characteristics and dynamical drivers. This study leverages the synergy between two satellite instruments, CloudSat's Cloud Profiling Radar (CPR) and CALIPSO's Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP), to identify snowfall cases over the full GrIS and partition them into the IC and CLW regimes. We find that, overall, most CPR observations of snowfall over the GrIS come from IC events (70 %); however, during the summer months, close to half of the snow observed is produced in CLW events (45 %). IC snowfall plays a dominant role in adding mass to the GrIS, producing ∼ 80 % of the total estimated 399 Gt yr-1 accumulation. Beyond the cloud phase that defines the snowfall regimes, the macrophysical cloud characteristics are distinct as well; the mean IC geometric cloud depth (∼ 4 km) is deeper than the CLW geometric cloud depth (∼ 2 km), consistent with previous studies based on surface observations. Two-dimensional histograms of the vertical distribution of CPR reflectivities show that IC events demonstrate consistently increasing reflectivity toward the surface while CLW events do not. Analysis of ERA5 reanalyses shows that IC events are associated with cyclone activity and CLW events generally occur under large-scale anomalously high geopotential heights over the GrIS. When combined with future climate predictions, this snapshot of GrIS snowfall characteristics may shed light on how this source of ice sheet mass might respond to changing synoptic patterns in a warming climate.