- Home
- Scientific program
- Proceedings
- List of participants
- Photos
- Venue/Lodging
- SOC/LOC
- Links
- Contact



Benedikt Soja
Type of contribution: oral

Subdaily station motions from Kalman filtering VLBI data

Benedikt Soja, GFZ German Research Centre for Geosciences Maria Karbon, GFZ German Research Centre for Geosciences Tobias Nilsson, GFZ German Research Centre for Geosciences Kyriakos Balidakis, GFZ German Research Centre for Geosciences Susanne Glaser, Technische Universit{"a}t Berlin Li Liu, GFZ German Research Centre for Geosciences Zhiguo Deng, GFZ German Research Centre for Geosciences Robert Heinkelmann, GFZ German Research Centre for Geosciences Harald Schuh, GFZ German Research Centre for Geosciences

Station coordinates are one of the most important parameters in geodetic VLBI analysis as they are necessary for global terrestrial reference system realization. It is common that the coordinates are estimated as a constant offset per session. However, subdaily changes in position occur, most notably due to tides and loading effects. While the majority of the geophysical signals can be removed by applying state-of-the-art models, some seem to remain due to deficiencies of the models. We have used the Kalman filter in the GFZ version of the Vienna VLBI Software (VieVS@GFZ) to estimate subdaily station coordinate time series. The state in the Kalman filter is updated for every VLBI observation, resulting in a temporal resolution of just a few minutes. Several models for tidal and non-tidal ocean and atmosphere loading, and combinations thereof have been applied to investigate the extent of the subdaily motion reduction. The residual coordinate variations may be due to shortcomings of the geophysical models, correlations of the station coordinates with other parameters, and other unmodeled effects. In order to test whether some part of the motions are specific to the VLBI technique, the coordinates have been compared to ones from a geophysically consistent GNSS solution derived at GFZ. Finally, empirical models of the subdaily motions of various stations, e.g. containing diurnal and semidiurnal signals, have been estimated based on the coordinate time series derived from Kalman filtering. Tests regarding the significance of the model parameters and the impact on coordinate and baseline length repeatabilities have been performed as well.