Tightly coupled water vapor tomography algorithm for combining GNSS and MODIS signals

Due to several significant advantages, including high-precision observation data, high spatio temporal resolution, and all-weather availability, GNSS tomography technology has become one of the most potential technologies for sensing the atmospheric water vapor. Currently, fusion of multi-source atmospheric remote sensing data has gradually become a research hotspot to make up for the geometric defects of GNSS signal in the tomography model. In this paper, the disadvantage of traditional model including the MODIS signals is analyzed at first. An improved tomography method, combining the GNSS and MODIS signals, based on the voxel node model is proposed, which introduces high-resolution MODIS PWV into the tomographic model in the form of three-dimensional signals. To assess the validity of the proposed algorithm,three experimental schemes are carried out using the 15 MODIS images and the simultaneous GNSS data derived from five GNSS stations over Xuzhou region.The experimental results show that the average number of effective signals is increased by 34.15% and the mean root mean square error(RMSE) of tomography results is decreased by 25.10% with the proposed tomography approach. Furthermore, the water vapor profiles retrieved from the three schemes are assessed using the reference profiles from the radiosonde data close to the acquisition time.It is found that in the lower layers from 0 to 2 km, the improved method retrieves better 3D distribution of water vapor than the traditional approach, which highlights that the reconstruction quality of 3D water vapor field near the surface can be optimized by including the MODIS signals.