GNSS (GNSS - Global Navigation Satellite System) positioning is the most by used technique to obtain a location over terrestrial surface or close to it. After the effects generated by the ionosphere, the refraction that the signal suffers, when it goes through the neutrosphere, can be considered as one of the largest error sources in the signal, in which is generated a neutrospheric delay, also called Zenithal Tropospheric Delay (ZTD). When the ZTD is not modeled, significant errors in the GNSS positioning are observed. Numerical Weather Prediction models are good alternatives to obtain ZTD over South America, because they are fed daily by atmospheric observations and consequently they are able to capture the temporal and spatial oscillations of the ZTD values. In Brazil, a system is operating at CPTEC/INPE for the purpose of generating such forecasts with spatial resolution of 15 km and temporal of 3 hours. In order to assess the impact of the recent improvements in the quality of ZTD predictions, a robust evaluation of the ZTD predictions was performed, taking into consideration several analyses in the Brazilian territory, in which ZTD values obtained from GPS data collected by the RBMC stations were used as reference, considering the evaluation: seasonal, the continentality, altitude and latitude variation.
GNSS Positioning; Zenith Neutrospheric Delay; Numerical Weather Prediction; Data Assimilation
