| description abstract | A ground network of global positioning system (GPS) receivers is a powerful source of continuous integrated precipitable water vapor (IPWV) data that can be suitable for weather prediction and climate studies. The technical features of two different GPS data analysis strategies, one for climatological and the other for meteorological applications, are thoroughly described together with the achieved accuracy of the related atmospheric products. The Centro di Geodesia Spaziale (CGS) of the Italian Space Agency in Matera, Italy, has been processing, since January 1999, GPS data from a network covering the central Mediterranean area, providing GPS postprocessed zenith total delay (ZTD) with a latency of 15 days, suitable for climate applications only. In June 2001 a near-real-time (NRT) datastream was set up to deliver, on an hourly basis and with 1 h, 45 min latency, GPS ZTD useful for assimilation into numerical weather prediction models. The GPS internal validation of the postprocessed ZTD performed considering two different 2.5-yr-long time series, shows a ±5 mm bias (equivalent to ±1 mm of IPWV) and a standard deviation ranging from 4 to 15 mm. The validation of the NRT data, if compared with postprocessed ZTD estimates based on 16 months (June 2001?September 2002) of routine activities, shows a bias ranging from ?6 to 10 mm and a standard deviation from 20 to 5 mm significantly decreasing during the considered period, due to processing tuning. Annual and semiannual signatures are detected in the ZTD residuals time series that can be related to similar signals found in the GPS coordinate time series. These signatures have to be taken into account when the terrestrial reference frame is fixed in the meteorological solutions together with the linear trend of the site velocities. Problems and issues related to the routine operation, which could be useful, whenever an operational stage of ground-based GPS meteorology is planned, are discussed. | |
