New 3D velocity model of Estonia from GNSS measurements
DOI | 10.3176/earth.2021.08 |
---|---|
Year | 2021 |
Journal | Estonian Journal of Earth Sciences |
Volume | 70 |
Number | 2 |
Pages | 107-125 |
Type | article in journal |
OpenAccess | |
Litsents | CC BY 4.0 |
Estonian author | |
Language | English |
Id | 32637 |
Abstract
The aim of this study was to create a 3D crustal deformation model for Estonia, based on dense Global Navigation Satellite System (GNSS) data (geodetic points with velocities) and validate the existing models of horizontal and vertical crustal deformations with velocities from Estonian GNSS measurements. The observations performed for at least eight years at Estonian GNSS permanent stations and during the GNSS campaign measurements of 1997, 2008 and 2017 on the Estonian 1st-order geodetic reference network were used as input data. Coordinates of the geodetic points were calculated in the ITRF2008 reference frame using the Precise Point Positioning method. Horizontal and vertical velocities (in the North, East and Up directions) were calculated for a total of 22 GNSS points. Models for horizontal and vertical velocities were calculated using the remove–compute–restore method. The model of glacial isostatic adjustment (GIA) of the Nordic Geodetic Commission NKG2016GIA was used as a reference model. Residual velocities of GNSS points showed a good fit with respect to the reference model. The residual velocities were analysed by geostatistical methods and the prediction surfaces of the residual velocities were modelled. After adding the surface of the residual velocities back to the reference model NKG2016GIA, the modelled surface EST2020VEL was obtained. The obtained model was compared with the up-to-date intraplate deformation model NKG_RF17VEL. It was found that recent Fennoscandian intraplate deformation models NKG2016LU and NKG_RF17VEL fitted well with the Estonian GNSS data. However, both models are systematically shifted with respect to the Estonian GNSS data. For applications in Estonia, it is better to use the fitted model EST2020VEL. The uncertainty of the model is estimated to be lower than ±0.5 mm/a.