Published: Aug. 13, 2018 By

Castellani, Benjamin BÌý1Ìý;ÌýShupe, Matthew DÌý2Ìý;ÌýHudak, David RÌý3Ìý;ÌýSheppard, Brian EÌý4

1ÌýCooperative Institute for Research in Environmental Sciences, University of Colorado, and NOAA Earth Systems Research Laboratory
2ÌýCooperative Institute for Research in Environmental Sciences, University of Colorado, and NOAA Earth Systems Research Laboratory
3ÌýEnvironment Canada
4ÌýEnvironment Canada

Under the current warming trend of the Earth’s climate, the Greenland Ice Sheet (GIS) has been melting on its fringes and experiencing a net loss of mass. However, the mass balance of the GIS as a whole is not well understood due to a dearth of knowledge of the interior region. Central to understanding this mass balance is precipitation, of which non-satellite derived observations over the central GIS are sparse. Recently, a Precipitation Occurrence Sensor System (POSS) was deployed at Summit, Greenland, a research site located near the apex of the GIS in a region where the mass balance is thought to be near equilibrium or slightly positive. Here a conditional retrieval for the POSS based on qualitative ice particle habit information derived from relating periodic on-site ice crystal images to ranges of cloud base temperature derived from instantaneous lidar and interpolated radiosonde measurements is described. The resulting reflectivity and snowfall measurements are compared with a co-located vertical-pointing Ka-band cloud radar. Providing a broader perspective, this radar-based precipitation data is analyzed alongside weekly surface height change measurements from an accumulation forest. The annual cycle of snowfall at Summit is presented showing a clear summertime maximum and a smaller, late-winter maximum as well. The annual mean liquid equivalent snowfall was found to be 41mm, which is smaller than estimates provided by reanalysis products by a factor of 2. A comparison to the seasonal cycle of accumulation showed similar seasonal patterns, though the seasonal amplitude is dampened. A time series of accumulation and snowfall shows that while accumulation increases at a relatively steady rate throughout the year, snowfall occurs in two distinct and repeating regimes. The relatively low correlation (0.39) between snowfall and accumulation indicates that other factors like compaction and the redistribution of snow by wind are also important contributors to the seasonal variability of solid accumulation. The reduced amplitude of the accumulation seasonal cycle can be reasoned based on seasonal variations in latent heat flux and compaction, which help to counteract the seasonally changing nature of snowfall to provide a relatively consistent rate of accumulation throughout the year in lieu of the large variations in snowfall throughout the year. The density of the snow implied by the latent heat flux, solid accumulation, and liquid equivalent snowfall is far too low when compared to actual vertical density observations at Summit. Since the in-situ accumulation data is quite reliable, a large portion of this discrepancy is likely due to underestimates of the total snowfall by the POSS. However, this uncertainty alone cannot fully explain the rather large disagreement in mass. It is hypothesized that the only other source of mass, the deposition of snow by wind, contributes a sizable amount of mass to complete the balance.