Sensing Atmospheric Water Vapor Using Radio Waves: Studies of the 2, 3 and 4-D Structure of the Atmospheric Water Vapor Using Ground-based Radio Techniques Comprising the Global Positioning System, Microwave Radiometry and Very Long Baseline Interferometry

Doctoral thesis, 2002

The radio wave propagation time, the primary observable in the geodetic techniques Global Positioning System (GPS) and Very Long Baseline Interferometry (VLBI) is affected by the presence of the atmospheric gases. By using appropriate modeling in the post-processing of the recorded phase measurements, together with local meteorological information, it is possible to estimate the total amount of atmospheric water vapor. Another technique able to independently provide water vapor estimates through sensing the atmospheric emission is microwave radiometry (MR). Due to the capabilities of GPS to estimate the integrated amounts of water vapor - a major greenhouse gas - it is potentially valuable for meteorology and climate studies, while VLBI and MR could be seen as independent validation techniques. Knowledge of the water vapor behaviour is of importance for: a) improved wet delay modeling in the precise geodetic techniques, b) the real-time GPS navigation, and c) weather forecasting and climate studies. In this thesis we present techniques, methods and results of measurements of the temporal variability of integrated amounts of atmospheric water vapor (2-D), its spatial horizontal gradients (3-D) as well as its 3-D spatial structure (4-D). We study the local structures of water vapor from spatial scales of a few kilometers on temporal scales of some minutes, presenting studies of the local horizontal gradients using data from GPS, VLBI and MR, to spatial scales of some thousands of kilometers, studying the integrated water vapor trends over 9 years in Scandinavia using time series from GPS. Verification of the estimated structures using intercomparisons between the techniques is included. We also present results on the studies of the 4-D structure using a small-scale GPS network where we first use the different sensitivity to gradients of a single GPS receiver compared to a network of receivers to probe the vertical structure of the atmosphere. Then we present the tomog raphic technique and suggest methods for derivation of the 4-D water vapor structure applied to different permanent as well as temporary established GPS networks. We study the quality of the input slant delay estimates and the GPS tomography capabilities and limitations, using simulated and real data. We also suggest methods to improve the technique.