In this thesis it is shown how the atmospheric precipitable water content derived from GNSS data can be assimilated within an operational meteorological now-casting system and how Precise Point Positioning results compare to the network solution. Passing weather fronts can be analyzed much better by considering the information provided by GNSS derived tropospheric wet delays because this data is directly influenced by changes in humidity in the free atmosphere, whereas the data at the meteorological ground stations react to these changes with a considerable time delay. This allows to forecast, e.g.
heavy rainfall potentially causing local floodings more reliable and to narrow down the affected region. It is to be expected that the accuracy of the PPP zenith wet delay estimates is worse than a network relative solution due to several effects (satellite clock errors, biases, no ambiguity resolution), but independence from the reference station data will significantly shorten the latency of the results (few min), and provide the regional/national weather service to enhance the prognosis in the numerical forecast model.