Abstract:
Precipitable water vapour (PWV) is an important component of the atmosphere and significantly influences many atmospheric processes. It is a physical parameter that is difficult to measure with adequate spatial and time resolution under all weather conditions. In this study, radiosonde data for the city of Riyadh in central Saudi Arabia (24430 N; 46400 E, 764 m a.s.l.) from 1985 to 2007 were used to model the precipitable water vapour (PWV) from the measurements of vapour pressure and air temperature. A multilinear model was essentially unbiased with a correlation coefficient (R2 ) of 0.81, a mean bias error (MBE) of 0.06 mm, and a root mean square error (RMSE) of approximately 2.80 mm. The performance of the model was tested against two independent datasets. The first dataset was four and half years of radiosonde measurements of PWV for the period 2008–March 2012, and the second set was two years of PWV data obtained from sun photometer measurements at 940 nm. For both datasets, the predictability of the model was excellent, with MBE values of less than 1 mm. The RMSE was 3.21 mm for the first set and 2.50 mm for the second set. The predictive powers of 15 empirical models from the literature were tested against the measured PWV for the period 1985 to 2007. The performances of these models varied. The proposed multilinear model from this study outperforms the overall performance of these 15 models. ª
