Assessment of spatiotemporal filtering methods towards optimising crustal movement observation network of China (CMONOC) GNSS data processing at different spatial scales

Spatiotemporal filtering can effectively remove the common mode error (CME) which significantly affects the accuracy of the Global Navigation Satellite System (GNSS) coordinate time series. This contribution explores the performance of different spatiotemporal filtering methods applied to GNSS networks at different spatial scales. We selected small-scale (<500 km) and large-scale (>2000 km) GNSS networks from the Crustal Movement Observation Network of China (CMONOC) for the focus of the study. To remove or mitigate CME from the different-scale GNSS networks, principal component analysis (PCA), independent component analysis (ICA) and correlation-weighted spatial filtering (CWSF) are compared. In addition, we investigate the correlations between each of the GNSS station residual time series to examine the effectiveness of the novel CME filter. When compared with PCA and ICA results, we find that CWSF is less intrusive on the data and is more effective in reducing the CME in the different-scale GNSS networks, and thus the preferred the filtering methodology. We conclude that this study could provide an important reference to remove CME from GNSS coordinate time series.

Automated summary

This study explores the performance of different spatiotemporal filtering methods in reducing common mode error (CME) in GNSS networks at different spatial scales and concludes that correlation-weighted spatial filtering (CWSF) is the most effective method.