Seasonal variation and driving factors analysis of Greenland ice sheet mass based on GRACE combination model data

Abstract

Abstract In recent years, the increasing mass loss of the Greenland Ice Sheet (GrIS) has a profound impact on the global sea level and climate change. Although some studies have investigated the rate and acceleration of mass loss in GrIS, research on its seasonal variation is relatively limited. In this regard, we firstly use Gravity Recovery and Climate Experiment (GRACE) multi-sourced data to estimate the GrIS mass changes from January 2003 to December 2016. Then, we study the annual variations of ice sheet mass and its driving factors (Modèle Atmosphérique Régionale 3.9 model variables and ice discharge) through singular spectrum analysis (SSA) and wavelet transform analysis. Results show the following: (1) Based on the generalized three-cornered hat method (TCH), the ice sheet mass change obtained from the combined data from different mechanisms, the seasonal annual amplitude variation extracted by SSA is within 200 Gt. (2) In terms of the seasonality, it is also found that the dominant factor of the GrIS mass change is runoff (RU) and ice melt (ME), while snowfall (SF) and rainfall (RF) have little effect on it. Among them, the mean annual amplitudes of the RU and ME are 1.1 and 1.5 times of GRACE respectively, and are not in phase with GRACE. (3) Through the analysis of relative phase and phase approximation, GRACE-derived data has an annual oscillation period with the RU, ME and RF, which lags their phases by 5.2, 5.4 and 4.7 months, respectively, while SF takes only 1.8 months. In addition, the average phase approximation values of the ME and GRACE are larger than those of the RU and RF, while that of the SF is positive. The change trend of the SF is basically opposite to that of the ME, RU and RF, and only consistent in the period 2006–2012.