Statistical Comparison Between Different Approaches to GNSS Single-Frequency Data Processing for Meteorological Applications

The use of low-cost GNSS stations for meteorological applications requires the modeling of ionospheric errors. Although low-cost dual-frequency receivers are now available, current ones are still missing the L2 frequency and this prevents the availability of iono-free observations. This second frequency can be predicted by exploiting dual-frequency data collected by existent geodetic receivers according to different techniques. This paper presents a quality assessment of three different algorithms for synthetic L2 observations reconstruction, evaluating the impact on the Zenith Total Delay estimation. The three algorithms were applied to reconstruct the L2 frequency of a target geodetic receiver of EUREF Network. The differences between the ZTDs obtained from dual frequency observations and those obtained from L1 observations and synthetic L2 observations derived from the three different methods are computed to assess the quality of the reconstruction algorithms. The differences show good performances of the three methods with an overall accuracy ranging between 0.1 cm and 1 cm when the corrections are computed from geodetic stations at distances up to 65 km from the target one. Some considerations on the advantages and limits of the three compared techniques conclude the paper