A physical algorithm for precipitable water vapour retrieval over land using passive microwave observations

Precipitable water vapour (PWV) plays an important role in weather prediction, hydrological cycles and climate change. In this study, a physics-based algorithm was proposed to retrieve PWV over land using satellite passive microwave observations near 165.50 GHz and 183.31 GHz, which have higher spatial resolution and lower sensitivity to land surface emissivity than low-frequency channels. The algorithm uses a perturbation formulation of simplified radiative transfer equation to derive PWV and at-surface brightness temperature (BT) in 165.50 GHz simultaneously. It directly derives the at-surface BT instead of land surface temperature, so the algorithm needs no prior information of surface emissivity. Sensitivity analysis and retrieval experiments from simulated data were carried out for Advanced Technology Microwave Sounder (ATMS) onboard the National Polar-orbiting Operational Environmental Satellite System (NPOESS) Preparatory Project (NPP) satellite. The results showed that the algorithm using ATMS channels 17, 18 and 19 measurements can significantly reduce the PWV retrieval errors caused by surface emissivity uncertainty. The coefficient of determination (R-2), root-mean-square error (RMSE), and bias of the retrieved PWV from ATMS data by the proposed algorithm and SuomiNet Global Positioning System (GPS) PWV were 0.90, 0.43 cm and -0.02 cm, respectively. In addition, the Visible Infrared Imaging Radiometer Suite (VIIRS) cloud products were used to evaluate the effects of cloud on the PWV retrievals. The results showed that the presence of cloud does not decrease the accuracy of the PWV retrievals.