High Resolution Greenland Ice-Sheet Interannual Mass Balance Estimates Combining GRACE Gravimetry and Envisat Altimetry

Tuesday, 16 December 2014
Xiaoli Su1, Junyi Guo1, C.K. Shum1,2, Jianbin Duan1, Ian M Howat1,2, Kenneth C Jezek2, Yuchan Yi1, Chung-Yen Kuo3 and Kuo-Hsin Tseng4, (1)Ohio State University Main Campus, Division of Geodetic Science, School of Earth Sciences, Columbus, OH, United States, (2)Byrd Polar Research Center, The Ohio State University, Columbus, OH, United States, (3)NCKU National Cheng Kung University, Tainan, Taiwan, (4)NCU National Central University of Taiwan, Jhongli, Taiwan
Knowledge of long-term present-day melting rate of the Greenland ice sheet (GrIS) is important towards understanding anthropogenic climate change. However, ice-sheet mass change at spatial scales longer than 350 km has been monitored only since 2002 by the GRACE twin-satellites, whose current data span of ~10 years are insufficient to infer the long-term melting rate with significant interannual or longer variations. Continuous polar observing satellite radar altimeter data became available since the 1990s, and have more than an order of magnitude (~50 km) better in spatial resolutions than GRACE. However, radar or laser altimetry measures ice-sheet elevation or volume changes and has no knowledge of the firn or ice column density of the measured ice-sheets. Here we estimate the firn density by combining GRACE and Envisat data—GRACE data provide mass changes, while Envisat data provide volume changes—with the objective of estimating higher resolution mass changes using altimeter data at the interannual scale. We use the interannual variation in both data sets, which is based on the assumption that the influence of the top snow layer is most likely in seasonal or shorter period band, primarily because the published contemporary density estimates via linear trends are unreasonably large. GRACE and Envisat data from January 2003 through December 2009 are used. We find high correlation on the order of 0.7 over 60% of the GrIS, in particular over western Greenland. Significant and anomalous negative correlations are found in regions in northeast and southeast regions. Similar negative or small interannual variation correlations in these regions are found between GRACE and snow depth from ECMWF Reanalysis (ERA-Interim) model or from the Regional Atmospheric Climate Model (RACMO2/GR). For the positively correlated interannual variation regions which cover over 60% of Greenland, the estimated composite firn/ice density values have a range of 383.7±50.9 to 567.9±42.0 kg m–3. The definitive explanation of these negatively correlated regions in interannual variations at present remains elusive. After ignoring the GRACE-density estimates in these anomalous regions, we present our estimate of the high-resolution interannual mass balance of the GrIS based on density-corrected Envisat altimetry, 2003–2010.