GPS-Tomography for Meteorology: Impact on Operational Weather Forecast

Marc Troller, Alain Geiger, Donat Perler, Hans-Gert Kahle, Daniel Leuenberger, and Elmar Brockmann

Abstract:	This paper presents a) recent developments in ground based GPS-tomography for the determination of atmospheric water vapor content b) the impact of GPS determined humidity profiles on to operational weather forecast and c) the influence of an additional satellite navigation system as Galileo on the quality of the tomographic solution. As the occurrence of severe weather conditions, flooding and storms became more and more frequent in recent times, precise weather forecasting has gained major interest especially in view of life protection and mitigation of damage. One of the key elements in weather dynamics and, therefore, in weather forecast is the water vapour distribution. However, it is difficult to obtain a 3D humidity distribution which could be fed into the numerical forecast process. In most cases ground measurements of meteo parameters and balloon soundings are available. In the recent years it became feasible to assimilate GPS determined zenith path delays in weather models. But still, the time and the spatial resolution remain fair. Operational soundings are made 1, 2 or 4 times a day and launch sites are separated by several hundreds of kilometres. The GPS zenith path delays are given in a high temporal and sometimes in a good spatial resolution, however, they have no height resolution e.g. no 3D resolution. For local weather prediction it is of utmost value to have a humidity field with high temporal (1 h or less) and spatial (few 10 km or less) resolution available. To this end, we developed a tomographic approach to estimate and model the spatial distribution of the tropospheric water vapor using GPS signals. Atmospheric effects causing path delays are limiting factors in high precision satellite navigation applications. The slant path delays can be determined by the GPS measurements themselves. From these (integrated) path delays the 3D humidity distribution can then be extracted by tomographic approaches. A short description of the method will be given. In January 2006, we have set up an operational GPS-tomography procedure based on the permanent GPS network AGNES of swisstopo. 40 humidity profiles over Switzerland are determined on an hourly basis at about 30 km horizontal resolution. So far, a large data set of more than one year is now available for assessing its quality. The main limitation of the accuracy is given by the fact that the adjustment is partly ill-posed in the sense that only a part of the unknowns (water vapor distribution) can be determined. To circumvent this problem, geometric constraints and/or additional data, such as ground meteorological measurements are taken into account to achieve a uniquely defined solution. In this paper the method used is presented and results as well as comparisons with independent humidity observations from balloon soundings and the Swiss numerical weather prediction model analyses are shown. The accuracy and reliability in function of the prevailing weather situation and the ground meteorological parameters are assessed and their impact on weather forecast is discussed.
Published in:	Proceedings of the 20th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2007) September 25 - 28, 2007 Fort Worth Convention Center Fort Worth, TX
Pages:	2861 - 2869
Cite this article:	Troller, Marc, Geiger, Alain, Perler, Donat, Kahle, Hans-Gert, Leuenberger, Daniel, Brockmann, Elmar, "GPS-Tomography for Meteorology: Impact on Operational Weather Forecast," Proceedings of the 20th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2007), Fort Worth, TX, September 2007, pp. 2861-2869.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In