Near Real-Time Water Vapor Monitoring ina German GPS Network and Assimilation intoWeather Forecast Model

摘要

In the framework of the “GPS Atmosphere SoundingrnProject” (GASP) of the German Helmholtz Associationrnoperational determination of vertical integrated water vapor (IWV) in near real-time is performed since Mayrn2000 within a dense German GPS network forrnapplications in numerical weather predictions and climaternresearch. The estimation of IWV with a delay of 40rnminutes has demonstrated an accuracy of better than 2rnmm with a standard deviation at the level of ±1 mm. Thisrnaccuracy is sufficient for introducing GPS technologyrnresults into the operational weather forecast. The analysisrnis performed with the GFZ EPOS software in sliding 12-rnhour windows using parallel analysis of station clustersrnwith the Precise Point Positioning strategy. The GASPrnnetwork is steadily increasing and consists presently ofrnapproximately 100 sites. Further densification of the testrnnetwork to more than 100 sites is foreseen in the nextrnfuture. The NRT results are compared with the postprocessedrnestimates and validated using co-locatedrninstruments, namely water vapor radiometer andrnradiosondes. GPS observations have also been comparedrnregularly with the output of the limited-area model of thernGerman Weather Service to check the observation qualityrnand to validate the model output.rnThe comparisons indicate that the quality of the GPSrnresults is acceptable for data assimilation purposes. Somerncases of disagreement between GPS and model IWVrnvalues could be explained as shortcomings of the modelrndue to the poor time resolution of conventionalrnobservations. This confirms the high potential of the GPSrntechnique for the detection of rapidly evolvingrnmeteorological phenomena, e.g. frontal passages. Work isrncurrently being carried out to use the GPS data in thernforecast model and first results from assimilationrnexperiments will be discussed.