Investigation of the effect of tropospheric delay on the quality of GNSS time series in a volcanological context

Global Navigation Satellite System (GNSS) observations are widely used to monitor volcanic deformation through continuous measurements of surface displacements. Detecting subtle volcano-tectonic signals relies on stable and repeatable position estimates over long time periods, which can be difficult to achieve in complex atmospheric environments.

In geodetic GNSS processing, station coordinates are estimated jointly with tropospheric parameters, including zenith delays and, in many processing strategies, horizontal tropospheric gradients. These parameters are known to be coupled with the vertical component of the position and to reflect both large-scale and locally anisotropic atmospheric conditions. At volcanic summits, tropospheric variability is often enhanced by strong orographic effects, frequent cloud formation, intense precipitation, and, in tropical regions, persistently high atmospheric humidity.

In this study, we investigate the behaviour and temporal variability of tropospheric parameters estimated for GNSS stations installed near the summits of La Soufrière de Guadeloupe and Montagne Pelée in Martinique, two active volcanoes located in the tropical Lesser Antilles. The analysis focuses on zenith tropospheric delays, horizontal gradients, and their consistency over time, as well as on diagnostic indicators derived from GNSS phase residuals.

GNSS data are processed using two independent geodetic software packages, GINS and GipsyX, enabling a comparative assessment of tropospheric estimates and residual patterns obtained under different processing strategies. The GNSS-derived tropospheric parameters are examined in conjunction with observations from nearby meteorological stations and with the ECMWF ERA5 reanalysis, providing an external reference for the observed atmospheric variability.

This work presents an initial investigation of tropospheric modelling at volcanic summits in tropical environments. It discusses possible implications for the stability and interpretation of GNSS position time series used in volcanic deformation monitoring.