Monitoring snow melt characteristics on the Greenland ice sheet using a new MODIS land surface temperature and emissivity product (MOD21)

Land Surface Temperature (LST) and emissivity are sensitive energy-balance parameters that control melt and energy exchange between the surface and the atmosphere. MODIS LST is currently used to monitor melt zones on glaciers and can be used for glacier or ice sheet mass balance calculations. Much attention has been paid recently to the warming of the Arctic in the context of global warming, with a focus on the Greenland ice sheet because of its importance with sea-level rise. Various researchers have shown a steady decline in the extent of the Northern Hemisphere sea ice, both the total extent and the extent of the perennial or multiyear ice. Surface melt characteristics over the Greenland ice sheet have been traditionally monitored using the MODIS LST and albedo products (e.g. MOD11 and MOD10A1). Far fewer studies have used thermal emissivity data to monitor surface melt characteristics due to the lack of suitable data. In theory, longwave emissivity combined with LST information should give a more direct measure of snow melt characteristics since the emissivity is an intrinsic property of the surface, whereas the albedo is dependent on other factors such as solar zenith angle, and shadowing effects. Currently no standard emissivity product exists that can dynamically retrieve changes in longwave emissivity consistently over long time periods. This problem has been addressed with the new MOD21 product, which uses the ASTER TES algorithm to dynamically retrieve LST and spectral emissivity (bands 29, 31, 32) at 1-km resolution. In this study we show that using a new proposed index termed the snow emissivity difference index (SEDI) derived from the MOD21 longwave emissivity product, combined with the LST, will improve our understanding of snow melt and freezeup dynamics on ice sheets such as Greenland. The results also suggest that synergistic use of both thermal-based and albedo data will help to improve our understanding of snow melt dynamics on glaciers and ice sheets, and reduce uncertainties in estimating magnitudes and trends.