A new approach for GNSS time-series analysis with focus on uncertainty and sensitivity

Earth deformation monitoring is an important aspect in the realization and investigation of reference frame. Global Navigation Satellite Systems (GNSS) is well-established technique to estimate the Earth surface velocity field and the deformation rates at the location of permanent stations. We make use of GNSS coordinate time-series to estimate horizontal and vertical velocity fields. However, one of the challenges is the superposition of a variety of signals and the contribution of noise and random processes in the GNSS time-series that adds to the complexities in extraction of the signal of interest in a straightforward manner. Therefore, we need to employ signal processing methods in the analysis of GNSS time-series signal separation or mode decomposition. In addition, the geometry of GNSS positioning and some other effects, e.g., solar radiation pressure-induced draconitic year etc. cannot be neglected. On the other hand, because of unpredictable mechanisms, e.g., earthquakes, coordinate time-series may experience discontinuities. For this reason, we employ mathematical tests to detect, estimate and remove these mechanisms to conclude the velocity field. In addition to signal separation and estimation of a realistic velocity field, the uncertainty estimation is also important for geoscientific applications. For this purpose, we take advantage of the variance and covariance component estimation methods to estimate not only a realistic velocity field, but also for realistic estimation of uncertainties. Studying of uncertainties is of importance and applicable in the sensitivity analysis. In some cases, we would like to know the sensitivity of the GNSS time-series to specific signals, e.g., solid Earth and mantle signals etc. In this study, we will present our work-in-progress approach for GNSS time-series analysis by combining data-driven and analytical approaches to have a better understanding of GNSS time-series with focus on the velocity field of our study area in central Europe.