Abstract
Abstract. The determination of the distribution of water vapor in the atmosphere plays an important role in the atmospheric monitoring. Global Navigation Satellite Systems (GNSS) tomography can be used to construct 3-D distribution of water vapor over the field covered by a GNSS network with high temporal and spatial resolutions. In current tomographic approaches, a pre-set fixed rectangular field that roughly covers the area of the distribution of the GNSS signals on the top plane of the tomographic field is commonly used for all tomographic epochs. Due to too many unknown parameters needing to be estimated, the accuracy of the tomographic solution degrades. Another issue of these approaches is their unsuitability for GNSS networks with a low number of stations, as the shape of the field covered by the GNSS signals is, in fact, roughly that of an upside-down cone rather than the rectangular cube as the pre-set. In this study, a new approach for determination of tomographic fields fitting the real distribution of GNSS signals on different tomographic planes at different tomographic epochs and also for discretization of the tomographic fields based on the perimeter of the tomographic boundary on the plane and meshing techniques is proposed. The new approach was tested using three stations from the Hong Kong GNSS network and validated by comparing the tomographic results against radiosonde data from King's Park Meteorological Station (HKKP) during the one month period of May 2015. Results indicated that the new approach is feasible for a three-station GNSS network tomography. This is significant due to the fact that the conventional approaches cannot even solve a network tomography from a few stations.
Highlights
Information of the distribution and variation of atmospheric water vapor is essential for meteorological applications
Where mw(e) is a wet mapping function and the VMF1 mapping function was used in this study; GwN and GwE are the wet delay gradients in the north–south and east–west directions, respectively; R is the post-fit residuals and in one satellitereceiver, the residuals exceeding 2.5× the standard deviation were removed and computed means were subtracted from the remaining residuals to clean observation from systematic effects; ZWD is the zenith wet delay of the Global Navigation Satellite Systems (GNSS) station, which can be obtained by subtracting the zenith hydrostatic delay (ZHD) from the zenith total delay (ZTD)
In order to find the reason for the large difference between the worst and best results, the tomographic field, the distribution of the signals and the nodes at these four epochs are given in Fig. 9, where Fig. 9a and b correspond to the best results at 00:00 and 12:00 UTC, respectively, both of which show uniform distributions of the GNSS signals
Summary
Information of the distribution and variation of atmospheric water vapor is essential for meteorological applications. With the development of Global Navigation Satellite Systems (GNSS), using GNSS measurements to remotely sense water vapor in the atmosphere has attracted significant attention due to their 24 h availability, global coverage and low cost. Based on GNSS measurements collected from a regional or global GNSS reference network, a regional or a global tomographic model, which is three-dimensional (3-D), can be constructed. Using the slant wet delays (SWDs) estimated from the GNSS signals of a GNSS network to construct a tomographic model is called GNSS tomography. Flores et al (2000) built the first GNSS tomographic model using 4 × 4 × 40 voxels and developed Local Tropospheric Tomography Software (LOTTOS) for simulation and processing of GNSS data. Gradinarsky (2002) developed the wet refractivity Kalman filter (WeRKaF) for tomographic inversion of GNSS data and the filter mainly focused on the initialization of the to-
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
