Cai et al. (2026) Fusion-Based Regional ZTD Modeling Using ERA5 and GNSS via Residual Correction Kriging

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Remote Sensing
• Year: 2026
• Date: 2026-03-23
• Authors: Yang Cai, Hongyang Ma, Zheng Wang, Shuaishuai Jia, Xin Duan, Ge Shi, Chuang Chen
• DOI: 10.3390/rs18060963

Research Groups

Specific author affiliations or research groups are not explicitly provided in the paper text.

Short Summary

This study proposes a Residual Correction Kriging (RK ZTD) method to fuse sparse Global Navigation Satellite System (GNSS) Zenith Tropospheric Delay (ZTD) data with continuous but biased ERA5 ZTD grids, significantly improving the precision and mitigating systematic biases of regional ZTD products in the Netherlands.

Objective

• To develop and validate a multi-source data fusion framework, the Residual Correction Kriging method for ZTD (RK ZTD), that integrates high-precision GNSS ZTD and spatially continuous ERA5 ZTD grids to reconstruct a refined, high-precision regional ZTD product, thereby addressing the limitations of sparse GNSS networks and systematic biases in reanalysis products.

Study Configuration

• Spatial Scale: Regional (The Netherlands), utilizing 36 GNSS stations.
• Temporal Scale: Observations collected in 2023.

Methodology and Data

• Models used: Residual Correction Kriging method for ZTD (RK ZTD), Kriging interpolation algorithm.
• Data sources:
  • High-precision GNSS ZTD estimates from 36 stations in the Netherlands (10 modeling, 26 validation).
  • ERA5-derived ZTD grids provided by the European Centre for Medium-Range Weather Forecasts (ECMWF).

Main Results

• The proposed RK ZTD model provides spatially robust and high-precision ZTD products.
• RK ZTD achieved a Root Mean Square Error (RMSE) of 5.70 mm.
• This represents an improvement of 58.4% compared to the original ERA5 ZTD (RMSE of 13.69 mm).
• This represents an improvement of 35.4% compared to the GNSS-Kriging ZTD (RMSE of 8.82 mm).
• The absolute bias was reduced to 0.41 mm, significantly lower than the 5.15 mm for the original ERA5 ZTD, indicating effective mitigation of systematic biases.
• The method demonstrated stable performance across all seasons and significantly alleviated interpolation inaccuracies caused by sparse GNSS stations, even under extreme weather conditions like Storm Ciarán.

Contributions

• Introduces a novel Residual Correction Kriging (RK ZTD) method for multi-source data fusion, effectively combining the high precision of GNSS ZTD with the spatiotemporal continuity of ERA5 ZTD.
• Significantly enhances the precision and mitigates systematic biases of regional ZTD products, outperforming both original ERA5 and standard GNSS-Kriging methods.
• Provides a robust solution for generating high-accuracy regional atmospheric water vapor information, addressing the limitations of sparse observation networks and reanalysis product biases.
• Demonstrates consistent and stable performance across various seasonal and extreme weather conditions, proving its value for advanced Earth environmental science applications.

Funding

Funding information is not explicitly provided in the paper text.

Citation

@article{Cai2026FusionBased,
  author = {Cai, Yang and Ma, Hongyang and Wang, Zheng and Jia, Shuaishuai and Duan, Xin and Shi, Ge and Chen, Chuang},
  title = {Fusion-Based Regional ZTD Modeling Using ERA5 and GNSS via Residual Correction Kriging},
  journal = {Remote Sensing},
  year = {2026},
  doi = {10.3390/rs18060963},
  url = {https://doi.org/10.3390/rs18060963}
}