Numerical Weather Models for Tropospheric Mitigation in Marine Kinematic GPS: a Daylong Analysis

F.G. Nievinski

Abstract:	It has been recommended that, “in precise [static] applications where millimetre accuracy is desired, the delay must be estimated with the other geodetic quantities of interest” [McCarthy and Petit, 2004, p. 100]. While that recommendation is common practice in static positioning, tropospheric delay remains as one of the main error sources in medium to long-distance kinematic positioning. Its mitigation is more challenging in kinematic applications because its strong correlation with the vertical coordinate is aggravated by the need to estimate the rover position every epoch. In this paper we report one further step in our investigation on the use of Numerical Weather Models for predicting tropospheric delays, aiming at improvements in kinematic applications. We analyze a daylong session. Our results show that NWM yields a slight improvement in height bias, with no improvement in horizontal bias. Regarding ambiguity resolution, NWM shows similar performance as UNB3m. Observation residuals show no significant change. We have show that NWM have only marginal improvement on 70 km kinematic baselines over wellestablished tropospheric delay prediction models (Saastamoinen, UNB3m). As ray-tracing in NWM is far more complex and computationally more expensive than those simpler models, they should be preferred until one demonstrates that the impact of NWM tropospheric delay predictions is, indeed, far superior.
Published in:	Proceedings of the 19th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2006) September 26 - 29, 2006 Fort Worth Convention Center Fort Worth, TX
Pages:	2017 - 2026
Cite this article:	Nievinski, F.G., "Numerical Weather Models for Tropospheric Mitigation in Marine Kinematic GPS: a Daylong Analysis," Proceedings of the 19th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2006), Fort Worth, TX, September 2006, pp. 2017-2026.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In