Data assimilation of GPS-ZTD into the RAMS model through 3D-Var: preliminary results at the regional scale

Abstract The knowledge of water vapour distribution is a key element in atmospheric modeling and considerable information, also at the local scale, can be derived from the GPS-ZTD (global positioning system-Zenith total delay) data. This paper shows the assimilation of GPS-ZTD data into the RAMS@ISAC (Regional Atmospheric Modeling System at Institute of Atmospheric Sciences and Climate of the National Research Council) to improve the representation of the water vapour in the meteorological model. The data assimilation system is based on 3D-Var (three-dimensional variational assimilation system) and it is applied to a network of 29 receivers located within the Lazio Region, Central Italy. All collected data are processed using the PPP (precise point positioning) method through RTKLIB, an open source program package for GNSS (Global Navigation Satellite Systems) Positioning. Among the GPS receivers, three are single frequency receivers, able to acquire L 1 frequency only, so that it is necessary a preliminary reconstruction of L 2 synthetic observations, which is achieved by a new original ground-based augmentation strategy. Results show remarkably that the single frequency receivers can be used the same way as geodetic receivers. The RAMS@ISAC is run at 4 km horizontal resolution over central Italy and is nested, using one-way nesting, into a 10 km horizontal resolution run of the same model. The experiment was performed along to two months, from 28 July to 28 September 2017. Results show that the GPS-ZTD data, assimilated by 3D-Var, have an important impact on the analysis of the water vapour field and the RMSE of ZTD and IWV (vertically integrated water vapour) is roughly halved for the analysis compared to the background. The impact of the GPS-ZTD data assimilation is also evaluated for the very short term (VSF) forecast (1–3 h), obtaining an improvement of the ZTD and IWV RMSE for all three hours of forecast.