Modeling of Residual GNSS Station Motions through Meteorological Data in a Machine Learning Approach

Abstract

Long-term Global Navigation Satellite System (GNSS) height residual time series contain signals that are related to environmental influences. A big part of the residuals can be explained by environmental surface loadings, expressed through physical models. This work aims to find a model that connects raw meteorological parameters with the GNSS residuals. The approach is to train a Temporal Convolutional Network (TCN) on 206 GNSS stations in central Europe, after which the resulting model is applied to 68 test stations in the same area. When comparing the Root Mean Square (RMS) error reduction of the time series reduced by physical models, and, by the TCN model, the latter reduction rate is, on average, 0.8% lower. In a second experiment, the TCN is utilized to further reduce the RMS of the time series, of which the loading models were already subtracted. This yields additional 2.7% of RMS reduction on average, resulting in a mean RMS reduction of 28.6% overall. The results suggests that a TCN, using meteorological features as input data, is able to reconstruct the reductions almost on the same level as physical models. Trained on the residuals, reduced by environmental loadings, the TCN is still able to slightly increase the overall reduction of variations in the GNSS station position time series.

Modelling and prediction of GNSS time series using GBDT, LSTM and SVM machine learning approaches

“…With application of this model, 3D position accuracy of the GNSS receiver is improved by about 20%. Ruttner et al (2021) applied temporal convolutional neural network (TCN) to find connection between raw meteorological parameters and GNSS height residuals. They suggest that the trained TCN can achieve almost the same level with physical models on reduction of GNSS height residuals.…”

Rainfall‐induced ground deformation in southern Africa

“…The Earth surface deforms under several processes involving both deep earth and external envelopes, such as ocean, atmosphere and continental water (Biessy et al, 2011; Cazenave & Feigl, 1994). The dynamics of the external envelopes generate variations in mass recorded by GNSS methods (Ray et al, 2013; Ruttner et al, 2022; White et al, 2022), ground gravimetry (Crossley et al, 2005; Güntner et al, 2017) and spatial gravimetry (Ramillien et al, 2008; Wahr et al, 2004). The regional‐scale hydrology is investigated from GRACE satellite data where gravity variations are converted into an equivalent water thickness and into displacements (Chanard et al, 2018; Llovel et al, 2010; Ramillien et al, 2008; Wahr et al, 2004).…”