Impact of non-tidal loading corrections and processing strategy on Antarctic GNSS vertical time series

Accurate estimation of vertical land motion in Antarctica is crucial for understanding glacial isostatic adjustment (GIA), ice mass change, and sea-level rise. However, Global Navigation Satellite System (GNSS) position time series are affected by non-tidal loading (NTL), which can obscure geophysical signals and bias trend estimates. In this study, we evaluate the performance of 11 NTL model combinations from EOST (École & Observatoire des Sciences de la Terre, Strasbourg) and ESMGFZ (Earth System Modelling Group, GFZ Potsdam) in correcting vertical GNSS time series at three East Antarctic stations in Dronning Maud Land. We analyse five GNSS solutions processed with different strategies, including precise point positioning (PPP), double-difference (DD) network solutions, and a combined product.

Our results show that NTL corrections improve time series quality in PPP-based solutions, reducing root mean square (RMS), coloured noise, and seasonal amplitudes by more than 20 % at some sites. In contrast, network-based and combined solutions exhibit limited improvements, and in some cases, corrections introduced additional variability. Among loading components, non-tidal atmospheric loading (NTAL) consistently produces the largest reductions, while additional non-tidal oceanic (NTOL) and hydrological loading (HYDL) contributions are beneficial mainly in specific GFZ model combinations applied to PPP datasets.

Our findings demonstrate that both GNSS processing strategy and NTL model choice can affect inferred vertical trends, and in some cases even change their sign. Our evaluation provides a regional assessment of widely used NTL products under Antarctic conditions, with direct implications for GIA modelling and reference frame realisation, and supports the development of more robust correction strategies for future Antarctic GNSS studies.