Comparison of Numerical Weather Models in Estimating the GNSS Signals Slant Tropospheric Delay
作者单位：Nottingham Geospatial Institute University of Nottingham
会议名称：《第四届中国卫星导航学术年会》
会议日期：2013年
学科分类：07[理学] 08[工学] 070601[理学-气象学] 080401[工学-精密仪器及机械] 081105[工学-导航、制导与控制] 0804[工学-仪器科学与技术] 0706[理学-大气科学] 0825[工学-航空宇航科学与技术] 0811[工学-控制科学与工程]
关 键 词：GNSS Tropospheric Delay NWM PPP
摘 要：For high accuracy Global Navigation Satellite Systems （GNSS） positioning, such as Precise Point Positioning （PPP）, the troposphere delay remains one of the major error sources. The PPP convergence time and vertical positioning accuracy are influenced by the applied tropospheric model and mapping function. Due to the lack of real time atmosphere information, the accuracies of both the model and mapping function are limited. In the past decades, Numerical Weather Modelling （NWM） has experienced a period of rapid development. With the support of high performance supercomputers, NWM uses a 3D grid model to describe or forecast the atmospheric physical conditions, such as pressure, temperature, humidity, etc. These information could enable high accuracy direct estimation of the slant tropospheric delay （STD）. In contrast to the traditional mapping function approach, direct estimation on individual STDs can avoid the assumption that the site local weather condition is a balanced profile, thus the troposphere delays are estimated on both signal elevation and azimuth. This paper will review the technique of NWM based direct STD estimation, and investigate its resolution requirement as well as the real time applicability. Various products from the Unified Model （UM）, provided by the United Kingdom Met Office, is used in this study. In the resolution test, the troposphere delay estimation based on the UM North Atlantic and European （NAE） regional product （12-km horizontal resolution） is compared with estimation based on the UM global product （40-km horizontal resolution）. The high resolution product could provide a better estimation accuracy, but with the cost of a larger data volume. This test is aiming to initially investigate the estimation accuracy from different atmospheric information resolutions. In the real time applicability test, the troposphere delay estimation based on the UM forecast field product is compared with estimation based on the analyzed field. Although