Comparison of Tropospheric Decorrelation Errors in the Presence of Severe Weather Conditions in Different Areas and Over Different Baseline Lengths

J. Huang, F. van Graas

Abstract:	In the development of accurate and robust differential GPS (DGPS)-based navigation systems, one of the remaining challenges is to mitigate the tropospheric spatial decorrelation errors (TSDE) observed by two GPS receivers. Under nominal tropospheric conditions, this error is generally small and can be removed effectively through the use of tropospheric propagation delay models. However, during times of severe weather conditions, tropospheric propagation delay models are no longer accurate due to the lack of knowledge of the entire tropospheric profile that affects the two GPS receivers. The resulting decorrelation errors are identical on the pseudorange (PR) and accumulated Doppler frequency shift (AD) measurements, as well as on the different GPS frequencies. In this paper, a new method is introduced to detect and characterize TSDE. Mathematical derivations are provided to identify TSDE on a per satellite basis through the use of single and dual-frequency AD measurements. It will be shown that the error in the observed TSDEs is less than 0.17 m with a probability of less than 10 due to all error sources, including ionospheric decorrelation, satellite orbit error, noise and multipath. Thus, this measurement technique can identify tropospheric decorrelation errors with a variation largerthan 0.17 m with high levels of confidence. The magnitude of TSDE is dependent on the difference in the tropospheric conditions observed along the line-ofsights to a particular satellite from two GPS receiver locations. To characterize the TSDE, continuous DGPS measurement data were analyzed for different baseline lengths over long data periods of one year to capture most weather conditions for a particular geographical area. Measurement data were obtained from three pairs of GPS receivers that are part of the Continuously Operating Reference Stations (CORS) network. The receiver pairs are located in the following areas: 1) Athens, OH, with a baseline length of 16.4 km; 2) Raleigh, NC, with a baseline length of 6 km; and 3) Seattle, WA, with a baseline length of 5.4 km. These three locations represent areas with different weather patterns. All data from these locations were analyzed for TSDEs. Next, to identify severe weather conditions, weather data were obtained from Ohio University’s Scalia Laboratory and from the National Climatic Data Center (NCDC). Weather data of interest for this study includes temperature, pressure, humidity, rainfall, and radar images. Close correlation was found between large TSDEs and severe weather conditions. Statistical analyses are included that show the observed TSDEs for all three locations during the one-year period of time. Furthermore, the modified Hopfield model (MHM) was applied to the data in an attempt to reduce the TSDE. It was found that the model was ineffective in the presence of severe weather conditions due to the lack of knowledge of the tropospheric profile along the line-of-sights between the two GPS receivers and a particular satellite. For the shorter baseline of 5.4 km, TSDEs up to 0.23 m were found, while the longer baseline of 16 km resulted in TSDEs up to 0.4 m.
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:	2769 - 2787
Cite this article:	Updated citation: Published in NAVIGATION: Journal of the Institute of Navigation
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In