Numerical weather prediction models: application and evaluation on absolute and network-based positioning

Abstract

Spatial technologies have been widely used in positioning methods, the main applied technology is the GNSS (Global Navigation Satellite System). Among GNSS positioning systems one can cite GPS, GLONASS, GALILEO, BEIDOU/COMPASS, besides augmentation systems. In order to obtain high accuracy (centimeters) in GNSS positioning, it is indispensable to accomplish the atmospheric modeling (ionosphere and troposphere). Related to the troposphere, its effects on GNSS signals are caused by two components: the hydrostatic component that represents about 90% of the effect; and the wet component that has a minor effect but its temporal variation is greater, which difficult modeling. Saastamoinen and Hopfield models can be used to minimize the troposphere effects on GNSS signals. But, depending on intended accuracy, they cannot be adequate to the Brazilian reality. In order to solve this limitation, emerged the Numerical Weather Prediction Models (NWP). In this research we intend, mainly, to accomplish a rigorous evaluation concerning the use of PNT models in point positioning and network-based positioning. Besides, to evaluate in which regions of Brazil, in special Sao Paulo State, the atmospheric humidity (ZWD - Zenithal Wet Delay) can impact on geodesic positioning quality to the user.(AU)