INTERNATIONAL SYMPOSIUM ON GIS APPLICATIONS IN GEOGRAPHY & GEOSCIENCES, Çanakkale, Türkiye, 18 Ekim 2017 - 21 Ağustos 2019, ss.117
Global Navigation Satellite System (GNSS) is actively being used for various purposes such as
engineering studies, deformation measurements, meteorological predictions, navigation. The method
would be used depends on work to be applied. Relative and absolute positioning techniques are the
two main methods that distinguish positioning with GNSS as the working methodology. Precise Point
Positioning (PPP) which is an absolute positioning technique has stood out among these techniques in
recent years and is becoming more popular, has become an important alternative to relative
positioning technique in evaluating GNSS data (postprocessing) in terms of position accuracy it
provides. This improvement in accuracy relies on modelling of the errors in post-process phase. There
are many error sources those have an effect on positioning accuracy. Atmospheric error sources are
ionospheric delay and tropospheric delay. While ionospheric delay error can be eliminated with using
L1-L2 phase and also code observation differences, tropospheric delay needs to be modelled because
it is not predictable due to constantly changing water vapour in the layer. In this study, 3 IGS station
couples those have horizontal distance less than 160 km from each other with significant height
difference were selected. Coordinates were obtained for 7 days from Scripps Orbit and Permanent
Array Center (SOPAC) to perform repeatability analysis. At first, these datas were processed in
GIPSY-OASIS 6.4 (GNSS-Inferred Positioning System and Orbit Analysis Simulation Software) with
Kinematic PPP approach using 5 minutes intervals while Zenith Tropospheric Wet Delay (ZWD)
modelling on, and then secondly off. The results were compared to absolute position of the sites to
determine effects of ZWD modelling on coordinate values as in RMS values. The purpose of the study
was to see if there was a correlation between height difference and RMS improvement with usage of
ZWD modelling.