Consistency and Homogeneity of ZTD Estimates from IGS Repro3

In 2019-2020, the International GNSS Service (IGS) coordinated its third reanalysis of the complete history of GNSS data collected by the IGS global network since 1994. Ten analysis centers (ACs) participated in this so-called “Repro3” effort, using the latest models and methodologies. Several of them provided, among other products, zenith tropospheric delay (ZTD) estimates.

In this study, we analyze the quality of ZTD estimates from four ACs (COD, GFZ, GRG, and TUG) along three different aspects: 1) the number and size of ZTD outliers, 2) the mean ZTD differences (or biases), and the standard deviation of ZTD differences between ACs as well as with respect to the ERA5 reanalysis, and 3) the homogeneity of the ZTD time series at individual sites. Overall statistics and case studies are presented for 200 sites.

We find an overall agreement between ACs at the sub-millimeter (bias) and sub-centimeter (standard deviation) level. However, a notable number of outliers and inhomogeneities are observed at individual sites. These are attributed to differences in metadata, particularly antenna information, and, to a lesser extent, to differences in processing strategies and software-specific features (e.g., outlier editing). Controlling the quality of metadata and optimizing the processing strategy are two major pathways for improving the quality of GNSS ZTD estimates for use in climate analysis.

This work is part of the activities of a Joint Working Group of the IAG Inter-Commission Committee on “Geodesy for Climate Research” (ICCC), in cooperation with the IGS ACs.