Wu et al. (2026) Synergistic architecture of conventional GRACE filters: Global generality to region-specific optimal filtering via multiscale hydrologic validation

Identification

• Journal: Journal of Hydrology Regional Studies
• Year: 2026
• Date: 2026-01-27
• Authors: Xiaohui Wu, Yunlong Wu, Sulan Liu, Qi Liu, Danyi Hu
• DOI: 10.1016/j.ejrh.2026.103180

Research Groups

Hubei Key Laboratory of Regional Ecology and Environmental Change, School of Geography and Information Engineering, China University of Geosciences, Wuhan, China.

Short Summary

This study develops a latitude-dependent synergistic filtering framework for GRACE/GRACE-FO data to overcome limitations of traditional decorrelation filters, demonstrating improved noise suppression and signal preservation, especially in low-latitude regions, for enhanced hydrological monitoring.

Objective

• To develop and validate a latitude-dependent synergistic filtering framework that integrates conventional spatial decorrelation filters (Swenson, SSAS, MVMDS) to achieve superior noise suppression and signal preservation in GRACE/GRACE-FO data, particularly for regional hydrological applications, by overcoming the instability and boundary effects of traditional methods.

Study Configuration

• Spatial Scale: Global scope, with validation across 40 river basins and two lakes (Victoria and Nasser).
• Temporal Scale: April 2002 to June 2023, covering both GRACE and GRACE-FO missions.

Methodology and Data

• Models used: GLDAS NOAH V2.1 model (for simulating terrestrial water storage signals).
• Data sources:
  • GRACE Level-2 Release 06 and Release 06.3 GSM products from Center for Space Research (CSR), German Research Centre for Geosciences (GFZ), Jet Propulsion Laboratory (JPL), and Institute of Geodesy at Graz University of Technology (ITSG).
  • Degree-1 spherical harmonic coefficients (SHCs) replaced with TN-13.
  • C20 and C30 coefficients substituted with values from TN-14.
  • CSR Mascon data (reference for hydrological basin validation).
  • Satellite altimetry data (reference for lake validation).
  • DDK5 filter (for extracting noise components in synthetic simulations).

Main Results

• The proposed synergistic filtering framework (SAY, SAV, YAV) effectively overcomes boundary effects and instability inherent in traditional decorrelation filters.
• The YAV mode is optimal for low-latitude regions (below 20° latitude), effectively mitigating high-frequency striping noise to recover masked tropical signals and enable the monitoring of small-scale lakes, demonstrating minimal signal loss in this region.
• The SAY and SAV modes ensure signal stability in mid-to-high latitudes (40°S to 60°N) by avoiding leakage and preserving seasonal amplitudes, achieving optimal matching with benchmark data in the Dnieper basin (Root Mean Square Error (RMSE) of approximately 1.6 cm).
• The framework improves the proportion of high Signal-to-Noise Ratio (SNR) results by 10 %–15 % in noise-prone areas (20° to 50° latitude range) compared to traditional methods.
• All three proposed synergistic methods demonstrated robust temporal stability, effectively mitigating RMSE surges typically observed during the initial and final phases of the GRACE mission.
• In the spectral domain, synergistic methods significantly suppressed high-degree amplitudes (beyond degree 40) and high-frequency noise (wavenumbers 40-60), outperforming some spatiotemporal filters like CGS in certain aspects.
• The SAV+DDK6 method yielded the lowest Pearson correlation coefficient (0.0401) and the most moderate slope (0.1144) for signal-noise orthogonality, indicating superior signal-noise separation.
• For Lake Victoria (low-latitude, high noise), the YAV method achieved the highest correlation (0.96) with altimetry data and the highest correlation with the altimetry spectrum (0.918).
• For Lake Nasser (mid-latitude, smaller area), synergistic methods maintained correlations with altimetry data around 0.75, representing an improvement of approximately 0.1 over traditional methods.

Contributions

• Introduces a novel Latitude-Dependent Synergistic Filtering Architecture for GRACE/GRACE-FO data, providing a robust, latitude-dependent strategy for filter selection.
• Significantly enhances the accuracy of regional terrestrial water storage assessments and the monitoring of extreme hydrological events, particularly in noise-prone low-latitude regions and for small-scale lakes.
• Quantitatively demonstrates improved signal-to-noise ratio (10-15%) and better signal-noise orthogonality compared to traditional decorrelation methods.
• Offers a consistent and stable baseline for future multi-satellite gravity missions by generating global grids with consistent background noise levels, thereby facilitating subsequent spatial smoothing and long-term standardized monitoring of global terrestrial water storage.

Funding

• National Key R&D Program of China (2024YFF1308104)
• National Natural Science Fund of China (Nos. 42274111, 42442015, 42574073)

Citation

@article{Wu2026Synergistic,
  author = {Wu, Xiaohui and Wu, Yunlong and Liu, Sulan and Liu, Qi and Hu, Danyi},
  title = {Synergistic architecture of conventional GRACE filters: Global generality to region-specific optimal filtering via multiscale hydrologic validation},
  journal = {Journal of Hydrology Regional Studies},
  year = {2026},
  doi = {10.1016/j.ejrh.2026.103180},
  url = {https://doi.org/10.1016/j.ejrh.2026.103180}
}