Assimilation of GRACE-derived oceanic mass distributions with a global ocean circulation model

作者：Saynisch J^*; Bergmann Wolf I; Thomas M

来源：Journal of Geodesy, 2015, 89(2): 121-139.

DOI：10.1007/s00190-014-0766-0

• Data assimilation
• GRACE satellite gravimetry
• Time variant ocean mass distribution
• GRACE error estimation
• Correction of atmospheric forcing

To study the sub-seasonal distribution and generation of ocean mass anomalies, Gravity Recovery and Climate Experiment (GRACE) observations of daily and monthly resolution are assimilated into a global ocean circulation model with an ensemble-based Kalman-Filter technique. The satellite gravimetry observations are processed to become time-variable fields of ocean mass distribution. Error budgets for the observations and the ocean model's initial state are estimated which contain the full covariance information. The consistency of the presented approach is demonstrated by increased agreement between GRACE observations and the ocean model. Furthermore, the simulations are compared with independent observations from 54 bottom pressure recorders. The assimilation improves the agreement to high-latitude recorders by up to 2 hPa. The improvements are caused by assimilation-induced changes in the atmospheric wind forcing, i.e., quantities not directly observed by GRACE. Finally, the use of the developed Kalman-Filter approach as a destriping filter to remove artificial noise contaminating the GRACE observations is presented.

• 出版日期2015-2