Determination of weighted mean temperature (Tm) lapse rate and assessment of its impact on Tm calculation

Abstract

In Global Navigation Satellite System (GNSS) meteorology, improved understanding of weighted mean temperature (Tm) variation and estimation is imperative because Tm is crucial to the quantification of precipitable water, which is an important parameter in numerical weather prediction systems. The commonly used methods for determining Tm use the relationship between surface temperature Ts and Tm, and use blind models developed from atmospheric reanalysis products. Since the Ts recorded in the sensors or the sample points of the reanalysis data sets are usually not at the same height with the GNSS station, it is always necessary to vertically adjust the estimated Tm, which requires Tm lapse rate. In this study, the globally distributed radiosonde data were collected to calculate the Tm lapse rate, and the relationship between the Tm lapse rate and location, seasonal changes was shown in detail. To verify the importance of the Tm lapse rate in Tm calculation, the Tm lapse rate obtained in this study was applied in the Global pressure and temperature 2 wet (GPT2w) model. Numerical results show that the root mean square error (RMSE) of Tm estimated with the consideration of Tm lapse rate is improved by 0.64 K in average. Moreover, the impact of Tm on GNSS-PWV was analyzed, showing that the improvement of RMSEPWV and RMSEPWV/PWV are 0.05 mm and 0.23%, respectively.