A new global grid model for the determination of atmospheric weighted mean temperature in GPS precipitable water vapor

In ground-based global positioning system (GPS) meteorology, atmospheric weighted mean temperature, $$T_\mathrm{m}$$ , plays a very important role in the progress of retrieving precipitable water vapor (PWV) from the zenith wet delay of the GPS. Generally, most of the existing $$T_\mathrm{m} $$ models only take either latitude or altitude into account in modeling. However, a great number of studies have shown that $$T_\mathrm{m} $$ is highly correlated with both latitude and altitude. In this study, a new global grid empirical $$T_\mathrm{m} $$ model, named as GGTm, was established by a sliding window algorithm using global gridded $$T_\mathrm{m} $$ data over an 8-year period from 2007 to 2014 provided by TU Vienna, where both latitude and altitude variations are considered in modeling. And the performance of GGTm was assessed by comparing with the Bevis formula and the GPT2w model, where the high-precision global gridded $$T_\mathrm{m} $$ data as provided by TU Vienna and the radiosonde data from 2015 are used as reference values. The results show the significant performance of the new GGTm model against other models when compared with gridded $$T_\mathrm{m} $$ data and radiosonde data, especially in the areas with great undulating terrain. Additionally, GGTm has the global mean $$\hbox {RMS}_{\mathrm{PWV}} $$ and $$\hbox {RMS}_{\mathrm{PWV}} /\hbox {PWV}$$ values of 0.26 mm and 1.28%, respectively. The GGTm model, fed only by the day of the year and the station coordinates, could provide a reliable and accurate $$T_\mathrm{m} $$ value, which shows the possible potential application in real-time GPS meteorology, especially for the application of low-latitude areas and western China.