Retrieving tropospheric parameters using predicted multi-GNSS orbit and clock

During the last decade the stability of GNSS clocks has increased dramatically. New generation GNSS satellites are equipped with highly precise and stable clocks and the clock parameters can be predicted with even picoseconds accuracy for several hours. In this work we determined and predicted 90 days precise orbits and clocks of up to 115 satellites from GPS, GLO, GAL, BDS2/3 and QZSS. Based on the calculated and predicted orbit and clock products (SP3) we processed data from about 140 globally distributed stations using PPP in 24 hours static mode. The first 22 hours part uses the calculated satellite products and the last two hours part uses the predicted satellite products. The estimated parameters are daily station coordinates and 30 min tropospheric parameters (ZTD). To validate the last 2-hours of ZTD we generate a reference solution based on 24-hour calculated SP3 products. We also performed a statistical comparison with ECMWF weather model data which yields a root mean square deviation of about 12 mm. This initial comparison indicates that the ZTD estimated from predicted satellite orbit and clocks are sufficiently accurate for time critical meteorological applications.