Advances in the GNSS Based Estimation of Atmospheric Water Vapour and its Application in Numerical Weather Prediction in Hungary

Abstract:

In the last decade a CORS network (GNSSnet.hu) has been established in Hungary by FÖMI, incorporating stations from the neighbouring countries as well. The availability of this network enabled us to develop and realize a processing facility to estimate the tropospheric zenith delays from the observations in near real time within the frames of a research project funded by the Hungarian National Research Fund.

An automated near-realtime processing facility was established to collect and process the GNSS observations from 54 GNSSnet.hu stations with spatial resolution of approximately 60 kilometres, and it estimates the zenith wet delays (ZWD) and the precipitable water vapour (PW) and validate these values using radiosonde observations.

Recently we updated the operational Bernese processing software from version 5.0 to 5.2. The estimation of precipitable water vapour from the zenith wet delays was also further refined and new methods were implemented. The GNSS based zenith total delay and precipitable water vapour estimates are transmitted to EUMETNET's E-GVAP project.

Currently the real-time processing of GNSS observations using the PPP technique is experimented to enable the exploitation of GNSS observations for nowcasting applications. The first results and comparisons with the near-realtime estimations are presented on this poster. We contribute to a dedicated project, concentrating around the Transcarpathian region, which targets the real-time monitoring of severe weather events based on GNSS analysis. The first results may expected by the end of this year.

The Hungarian Meteorological Service has also conducted the first tests of the zenith total delay assimilation in their AROME numerical weather prediction model. The data assimilation diagnostics showed that the zenit total delays originated from nearly 70 stations have important contribution to AROME analysis (especially to humidity analysis). The verification results of the first test conducted in January, 2014 show that the assimilation of GNSS based zenith wet delays slightly improved the accuracy of the short-term forecasts of the 2m relative humidity values.