Comparative Studies for the Assessment of the Quality of Near-Real-Time GPS-Derived Atmospheric Parameters

Abstract

Accurate and frequent sampling of atmospheric parameters, such as water vapor, is important for enabling reliable weather forecasts and global climate studies over a wide range of spatial and temporal scales. Recent developments in global positioning system data processing have allowed the estimation of zenith total delay (ZTD), the delay of the neutral atmosphere, with a high degree of accuracy using continuously operating GPS networks. From this delay integrated water vapor can be derived by means of additional meteorological information, in particular observed pressure or numerical weather prediction model pressure. Comparisons with other independent techniques must be performed to evaluate the quality of atmospheric parameters directly estimated or retrieved from the GPS system. In this work the accuracy of GPS atmospheric parameter, namely, zenith total delay, delivered in near?real time from a European ground-based network of permanent GPS receivers has been assessed. It is compared to other GPS solutions, radiosonde profiles, and High-Resolution Limited-Area Model (HIRLAM)-derived ZTD. Intercomparisons between results from different GPS analysis centers in the framework of the Targeting Optimal Use of GPS Humidity Measurements in Meteorology (TOUGH) project show a mean ZTD station bias at the level of ±6 mm with a related standard deviation of about 7?8 mm. In the comparison with radiosondes, an overall ZTD bias of about 7 mm with a standard deviation of 9 mm is detected. Finally, the comparison of ZTD near?real time against the HIRLAM models has an average bias of about ?4.8 mm and a standard deviation of 11.5 mm.