Generation of 3D water vapour tomography using voxel-based approach in the Himalayan region

Along with regular position-related applications, observations from navigation satellite constellations can also be utilised for atmospheric studies. The present study deals with the 3D water vapour tomography carried out for the first time using Indian GPS stations set up in the Himalayan region. A voxel-based approach is applied to retrieve the vertical profile of precipitable water vapour (PWV). Horizontal 4×4 grids have been formed between 29°–31° latitude and 77.5°–79.5° longitude with 0.5° spatial resolution. To construct a tomography model, voxels are formed over these grids up to 10 km with 0.5 km vertical resolution. Water vapour density (WVD) is estimated for different days to observe its variations under different atmospheric conditions. The results show that the estimated vertical variation of water vapour closely matches the AIRS WVD profile. To compare the results, AIRS WVD observations are interpolated at the same altitude where the tomographic estimations are available. The average root mean square (RMS) error, standard deviation (SD), and mean absolute error (MAE) between estimated water vapour density and AIRS WVD are observed to be 1.52, 1.32 and 2.02 g/m³, respectively. The coefficient of determination (R²) value is observed to be close to 0.9, reflecting an excellent linearity between estimated and observed values. The results also show that the WVD is high up to the 5 km voxel and it is between 0 and 5 g/m³ above 5 km altitude. A good estimation of vertical water vapour profile has been obtained using CORS observations and can be used for further atmospheric analyses in this region.