MULTI-GNSS ERROR CHARACTERISTICS AND BENEFITS TO LONG-TERM MONITORING APPLICATIONS IN GEOSCIENCES

Global Navigation Satellite System (GNSS)-derived position solutions are used
for crustal deformations for long-term monitoring studies such as correcting sea-level
records for vertical land movements and to determine present-day surface-mass changes.
In all these studies scientists rely heavily on precise International GNSS Service (IGS)
products. In recent years the IGS products have partly been generated from a rigorous
combination of GNSS, such as Global Positioning System (GPS) and Globalnaya
Navigatsionnaya Sputnikovaya Sistema (GLONASS) observations. Although combined
solutions from two or more GNSS benefit from the diversity and redundancy of having
more than one GNSS, the solutions are also subjected to system-specific systematic
errors. Applications which demand high-accuracy products, therefore, would profit
from evaluations of the benefits and error characteristics of combined GNSS solutions.
In response to the increased availability of multi-GNSS observations from a truly
global ground network of receivers, the goal of this thesis is to investigate their overall
impacts on the derived products. Primarily, the impacts of combined GNSS data
processing for stations in a constrained environment with a potential for signal obstructions,
is investigated. The effects of signal obstructions on derived parameter
time series and station velocity estimates are assessed. The benefits of combined solutions
are evaluated for stations in constrained environments.
Moreover, the study of the impacts of combined solutions on satellite orbits and
station parameters contributes to the understanding of the error characteristics of combined
GNSS data processing on derived products. The consistency of the parameters,
noise analysis and system-specific periodic errors are assessed. Dominant system specific periodic errors and the impact of combined solutions on reducing the effects
are addressed. Unmodelled or insufficiently modelled (sub-)daily errors propagate to
longer periods and appear in high-end products coinciding with other longer periods,
which in turn may lead to misleading interpretations of the latter. The propagation
mechanism mainly depends, among other factors, on data sampling deficiencies and
GNSS ground repeat periods. Here, the results of this study show that combined solutions
not only reduce system-specific effects but also provide a means to identifying
the sources from other compatible elements.