Modelling The Azimuthal Dependence of The Tropospheric Delay on GPS Using Gamit/GLOBK and the Updated Vienna Mapping Function
No Thumbnail Available
Date
2013-04
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Addis Ababa University
Abstract
In order to improve the accuracy of point coordinates obtained by the Global Positioning System
(GPS), it is important to model the atmospheric delay caused by the Troposphere (Neutral
atmosphere). Until now the atmosphere has been considered to be horizontally layered and
azimuthally symmetric. Although for most applications this assumption is appropriate, azimuthal
asymmetry may introduce significant errors in geodetic measurements where high precision is
required. Azimuthal variation of 20% is quite commonly observed in humid areas. The
assumption of azimuthal symmetry may cause significant errors when the local troposphere has
large lateral temperature, pressure or humidity gradient variations. Recent VLBI studies indicate
that azimuth asymmetry effects at 150 Elevation typically produce a Root Mean Square (rms)
variation of 7mm, but at times this effect can be as much as 5times. This research investigates
the effect of azimuth asymmetry on Tropospheric delay modelling using GPS. The dependence
of the Tropospheric Delay on azimuth of the satellites was modelled using GAMIT/GLOBK
software and the application of the updated Vienna Mapping Function (VMF1) in Tropospheric
Delay modelling was analyzed. A combination of different mapping functions (Vienna Mapping
Function (VMF), Niell Mapping Function (NMF) and the Global Mapping Function (GMF))
with the corresponding a priori constraints (U-file (UFL) and Global Pressure Temperature (GPT
50) were used. The combination of VMF and UFL for a different number of gradients estimated
per day was done for North America, Europe and Uganda GPS data. The combinations made
were evaluated using the Repeatability graphs developed using GLOBK. The results were also
presented using tables and line graphs. When the gradients are modelled, the accuracy of fixing
the GPS points increases. The accuracy obtained is very low when the gradient is not estimated.
The accuracy of the East and North offsets increased more than the up offset in this research
since there was no serious storm surge. When the gradient estimated per day is two (2) the
accuracy is sufficient for fixing the East and North offsets well as for the up offset, more
gradients have to be estimated especially when there is a serious storm surge. VMF1 with GPT50
fix well the East and North offsets well as VMF1 and UFL fix the up offset better. When there is
a serious storm surge, gradients more than three should be estimated per day to model the
Tropospheric delay well. Different MF/APRIORI combinations should be varied with the
number of gradients estimated per day. Meteorological Rinex Files (RNX) should be used as a
priori zenith constraints in modelling the Tropospheric delay in the future.
Description
Keywords
Troposphere Modelling, Mapping Function, GPS, GAMIT/GLOBK