Assessment on atmospheric parameters at co-location sites

Atmospheric ties are affected by the differences of atmospheric parameters of space geodetic techniques at co-location sites. Similar to local ties, they could be applied along with local ties for a combination of space geodetic techniques to improve the realization of terrestrial reference frames (TRF). Theoretically, atmospheric ties are affected by the height differences between antennas at the same site and meteorological conditions. Therefore, atmospheric ties could be determined by analytical equation based on meteorological information from in situ measurements or weather model. However, there is often a discrepancy between the expected zenith delay differences and those estimated from geodetic analysis, thus potentially degrading a combined atmospheric ties solution. In this study, we analyse the time series of zenith delays from co-located GNSS antennas at Wettzell (height differences below 3 meters), for 11 years (2008–2018). GNSS observations were analyzed with Bernese GNSS software version 5.2 with double-differencing technique and relative tropospheric delay and gradients were estimated with L1, L2, and the ionosphere-free (L3) linear combination thereof. Atmospheric ties were derived analytically employing meteorological data from Global Pressure and Temperature model 3 (GPT3) and ERA5 reanalysis, as well as corrections derived from ray tracing (Potsdam Mapping Functions, PMF). The comparison shows that zenith delay differences are dominated by equipment changes. The discrepancies between atmospheric ties and estimated zenith delay differences are frequency dependent, with the L1 solutions being the least biased. For these small vertical differences, seasonal signals are not significant for all frequencies.