Nothnagel, A. G., & Krásná, H. (2021). Reference epochs in VLBI estimations of clock parameters. In R. Haas (Ed.), Proceedings of the 25th European VLBI Group for Geodesy and Astrometry Working Meeting (pp. 124–127). https://doi.org/10.34726/3065
Proceedings of the 25th European VLBI Group for Geodesy and Astrometry Working Meeting
-
ISBN:
978-91-88041-41-8
-
Datum (veröffentlicht):
Dez-2021
-
Veranstaltungsname:
25th European VLBI Group for Geodesy and Astronomy Working Meeting
-
Veranstaltungszeitraum:
14-Mär-2021 - 18-Mär-2021
-
Veranstaltungsort:
Cyberspace, Unbekannt
-
Umfang:
4
-
Keywords:
VLBI clock parameters; epoch of minimal variance
en
Abstract:
By default, the reference epoch of the relative clock offsets has always been the start of the sessions. If we consider a simple first order model for the relative clock parameters, i.e., just an offset and a rate, the formal errors of the parameters improve if the reference epoch is chosen to be at the middle of the session. In Altamimi et al. (2002), this is called the epoch of minimum variance. In a small study we have investigated whether this fact could be exploited in VLBI data analysis. For the CONT17a series of sessions, we ran solutions with the reference epochs of the clocks being the middle of the session and compared the results to those of standard solutions. We found that even for more sophisticated parameterizations of the clocks, i.e., with polynomials and piece-wise linear polygons, the formal errors of all clock parameters did improve significantly. Although the correlation matrices change as well, there, unfortunately, is no improvement for the formal errors of any other,(non-clock) parameter. At the same time, the condition numbers of the solutions did not change significantly either. We conclude that the effects of changing the reference epoch in VLBI estimations of clock parameters are confined to the clock parameter space alone.