Extracting terrestrial water storage signals from GRACE solutions in the Amazon Basin using an iterative filtering approach
Lun Pu, Dongming Fan, Wei You, Xinchun Yang, Zemede M. Nigatu, Zhongshan Jiang
Abstract
Estimating terrestrial water storage (TWS) from the Gravity Recovery and Climate Experiment (GRACE) solutions is an essential technique for water resources management. TWS estimation accuracy from the GRACE spherical harmonic coefficients (SHCs) is limited by the north–south stripes noise. We developed an iterative filtering (IF) method with high accuracy and efficiency to extract TWS signals in the Amazon Basin . The results show that the root mean squared error of TWS anomalies calculated using IF is ~52% smaller than FM (forward modelling) in synthesized data, while the signal-to-noise ratios (SNR) of TWS anomalies calculated using IF is improved by ~30%. The annual amplitude of TWS anomalies using IF (10.51 ± 0.28 cm) is close to the true signal (10.57 ± 0.28 cm). From the real GRACE experiment, both FM and IF can extract most of the TWS signals. However, the IF’s anti-noise ability (i.e., average SNR = 24.08) is significantly better than FM’s ability (average SNR = 19.01), which indicates the result using IF can achieve the equivalent level of accuracy as mass concentration solutions in the Amazon Basin. Thus, the IF is a robust and efficient method for extracting TWS signals in the Amazon Basin from GRACE SHCs.