Estimating Station Specific Zenith Tropospheric Delay in a Local GPS Network from Observed Surface Meteorology

Abstract

The total zenith tropospheric delay (ZTD) is an important parameter of the atmosphere which directly or indirectly give reflection of the atmospheric condition in a local GPS network. The use of the global tropospheric models such as the Saastamoinen model, Hopfield model, Neil model etc in estimating the tropospheric effects at the local level leaves much to be desired. These models are derived using data from available radiosonde obtained from Europe and North America continents. The global atmospheric condition used as constants in these models provides a broad approximation of the tropospheric conditions, but ignores the actual atmospheric conditions on a given location i.e. do not take into account the latitudinal and seasonal variations in the atmosphere. It is necessary to assess the effects of the troposphere on geodetic measurements based on ground meteorological measurements including temperature, pressure, and relative humidity. Daily RINEX GPS data from eight (8) Malaysian Real Time Kinematic GPS Network (MyRTKnet) stations in Southern Malaysia from year 2006 to June 2008 thus; covering 2½ years were processed. A computer program for modelling 2½ years of local meteorology within the network was developed and called MetMOD, similarly a program, call SaastroMOD was developed to estimate the local zenith hydrostatic delay. The results showed that, the estimated local tropospheric delay considered the temporal and spatial variation of the network thus; giving true reflection of the tropospheric delay in a local GPS network as against the standard atmosphere which ignores the actual local condition with a standard deviation of 0.18 m having a maximum zenith tropospheric delay (ZTD) of about 2.6 m. The best zenith hydrostatic delay (ZHD) comes from station JHJY with a standard deviation of 0.353 cm, while the best zenith wet delay (ZWD) comes from station TGRH with a standard deviation of 5.943 cm. The models show that, while parameters evolved for ZHD are constant at different locations, however, those parameters evolved for ZWD show significant variations from one location to the other. Hence then use of the local meteorological parameter in modelling tropospheric delay improves GPS positional accuracy.