Characterization of Recent Greenland Melt Events in Atmospheric Analyses and Satellite Data

Monday, 15 December 2014
Richard I Cullather1, Sophie Nowicki2, Bin Zhao1, Lora Koenig3 and Samiah Moustafa4, (1)NASA/GSFC Code 610.1, Greenbelt, MD, United States, (2)NASA GSFC, Greenbelt, MD, United States, (3)NASA, Greenbelt, MD, United States, (4)Rutgers University, Piscataway, NJ, United States
Data from a variety of observational and modeling sources have indicated enhanced, widespread melting of the Greenland Ice Sheet (GrIS) surface in recent years. In 2012, this melting was punctuated by the circumstance on 11 July when almost the entirety of the ice sheet simultaneously experienced surface melt, including Summit. While such an event has been considered as the result of unique meteorological conditions, the melting record for the season also occurred in 2012 based on spatial extent and duration. Previous melt extent records also occurred in 2002, 2007, and 2010. Melt extent may be estimated from remote sensing methods, but runoff volume may only be obtained from select in situ measurement locations or modeling methods. The aim of this study is to assess differences in available estimates of melt extent and runoff volume, and to characterize the spatial and temporal variability of surface melt. Significantly, the evaluation is conducted over the full available period from 1980 to the present, rather than focusing on one event. The GEOS-5 global atmospheric model with an improved surface representation for the GrIS is replayed against the NASA Modern-Era Retrospective Analysis for Research and Applications (MERRA) to produce an historical reanalysis. The use of GEOS-5 offers the potential for applying a global model with a realistic GrIS surface representation for the assessment of melt events and their relation to the large-scale climate. These values are compared with output from the Modèle Atmosphérique Régional (MAR) regional climate model and passive microwave remote sensing data. The approach is to spatially average values by drainage basins and evaluate the resulting time series. Seasonally, numerical analyses and models typically indicate less melt coverage during the early summer and more in late summer in comparison to passive microwave data. The relation between melt area, melt duration, and runoff volume differs markedly by drainage basin, with a greater amount of estimated runoff per melt area occurring in southern drainage basins while northern basins are characterized by shorter duration melt periods and a larger percentage of the basin area experiencing melt. An assessment of differences in the spatial and temporal variability of surface melt during enhanced melt years is presented.