Quantifying baseline length and height-difference dependent errors from commercially available software

Commercial software are usually refered to in national surveying practice and local deformation studies. Since their working environment is user friendly and implementation is easy, they could be prefered by many surveying practitioners or even researchers. However their usage is usually limited to 20-30 km due mainly to their crude ambiguity resolution algorithms and  the fact that they usually use broadcast ephemeris and standard troposphere models.  Since usualy the tropospheric zenith delay is not estimated but obtained from a standard troposphere model, the accuracy of the vertical component would be affected as the height difference between baseline points grows. As the baseline length becomes >20-30 km the tropospheric error would be coupled with orbital errors. Results based on large height difference would affect positioning solutions as well as local geoid determination studies. Monitoring local deformation such as landslides would also be affected if there is large height difference between the crown and the toe. The level of baseline dependent error is usually well covered in surveying standards manual however the effect of large height difference is generally ignored. In this study, we made an attempt to quantify vertical positioning error levels both considering large height difference between baseline points and longer baseline lengths. We used the data of CORS stations in the western US for the simulation of the observing session duration. TOPCON’s software MAGNET (Ver 4.0.1) was used to process the GNSS data. It appears that every 10 km increase in the baseline length and every 100 m increase in the height difference would cause 2.59mm and 1.24 mm vertical positioning error respectively.