EuVeM2022: a new European GNSS velocity model

The European continent is divided into several tectonic plates and velocity variations appear along plate boundaries. However, velocity changes inside a tectonic plate can also occur due to local effects or other geodynamic processes, which is of interest for researchers trying to understand intraplate deformations in the horizontal and vertical directions. These changes can be observed by a dense network of GNSS (Global Navigation Satellite System) stations or more recently by the usage of InSAR (Interferometric Synthetic Aperture Radar). However, a dense GNSS network cannot be maintained over large areas due to, e.g., high costs and topographical obstacles, thus a regional velocity model to study intraplate deformation has to be obtained via an interpolation of scattered GNSS station velocities. The obtained velocity models can be used to estimate strain rates, which can be compared to existing seismic hazard models. In addition, in areas with a limited amount of seismic information, strain rates obtained from GNSS velocity models can provide a useful input for seismic hazard models.

The increased availability of GNSS station velocities in Europe via the EUREF Permanent Network Densification (EPND) project (https://epnd.sgo-penc.hu) allows to obtain a complete picture of the horizontal and vertical deformation in Europe via an interpolation. Here, we apply a new interpolation technique to a velocity field solution from EPND. The homogenized and quality-checked velocity field is interpolated via a least-square collocation including the knowledge of existing plate boundaries to avoid a smoothing of nearby velocities on different tectonic plates. We also apply a moving variance approach to avoid effects of non-stationarity, which arise due to the variable station densities. The new 3D GNSS velocity model EuVeM2022 is used to estimate the strain rates and a comparison to seismic risk maps shows a clear correlation. However, in some areas increased shear strain rates as well as anomalies in the velocity model are visible where peak ground acceleration is not increased for example in Serbia.