Decomposition-based reconstruction scheme for GRACE data with irregular temporal intervals

• The GRACE satellites derived terrestrial water storage anomaly products exhibit irregular temporal intervals due to incomplete monthly coverage. • A decomposition-based reconstruction scheme is developed to address GRACE data with irregular temporal intervals. • The scheme effectively mitigates the adverse effects of irregular temporal intervals on the reconstruction process and applications of GRACE data. Irregular temporal intervals (ITI) in the Gravity Recovery and Climate Experiment (GRACE) data, caused by mission gaps and incomplete monthly sampling, introduce uncertainties into hydrological applications. This study introduces a decomposition-based reconstruction scheme specifically developed to preserve and leverage the ITI features of GRACE data for improved reconstruction performance. The scheme integrates multiple predictor processing techniques. It includes dynamic resampling to align predictor variables with the time bands of GRACE-derived terrestrial water storage anomaly (TWSA) products. It also incorporates an improved signal decomposition algorithm tailored for ITI series, a daily-scale lag correction to enhance temporal alignment, and exponential decay filtering to suppress high-frequency noise in precipitation inputs. Using data from 95 Mascon grids in southern China, the proposed approach improves median PCC by 5 %, NSE by 22 %, and reduces RMSE by 12 % compared to baseline schemes that disregard ITI. Reconstruction uncertainty is also reduced by 56 % (trend), 178 % (seasonal), and 12 % (subseasonal) compared to baseline schemes. Further comparison across multiple schemes confirms the method’s robustness. The results highlight the effectiveness and generalizability of ITI-aware modeling, offering a practical solution for GRACE-based hydrological studies under irregular temporal sampling conditions.