Declining Groundwater Storage in the Indus basin Revealed Using GRACE and GRACE-FO Data

Groundwater storage changes were calculated using the average of three GRACE mascon solutions (JPL, GSFC, and CSR) and ERA5-Land-derived TWSA. However, there are 33 months of gaps in the GRACE and GRACE-FO data, which significantly impact regional trends and predictions of water mass changes. Therefore, we applied a machine learning-based MissForest algorithm to fill these gaps at 0.25° resolution using eight input variables.本研究采用GRACE（重力恢复与气候实验，Gravity Recovery and Climate Experiment）质量浓度块（mascon）的三套解算结果（JPL、GSFC与CSR）与ERA5-Land衍生的陆地水储量异常（Terrestrial Water Storage Anomalies, TWSA）的平均值，计算地下水储量变化。然而GRACE与GRACE-FO（GRACE后续任务，GRACE Follow-On）数据存在33个月的数据间隙，这会显著影响水质量变化的区域趋势分析与预测。为此，本研究采用基于机器学习的MissForest算法，以8个输入变量在0.25°分辨率下填补上述数据间隙。