How to get continuous tropospheric delay when processing a long time series of GNSS data?

[FFFFFG-IGG]

Hi,
We're now processing a long time series of regional GNSS data, but we found that the output tropospheric delay is not continuous between days. For example, results of year 2022 doy 001 and year 2022 doy 002 show significant inconsistency:
year 2022 doy 001(end of that day)

year 2022 doy 001(begining of that day)

All products are obtained from COD and we run a standard PPP procedure described by GROOPS documentation (without estimating VTEC).

We just wonder if there is someway to avoid this inconsistency or someone else also find this issue? Thanks in advance.
Regards Fan

[tmayerguerr]

Hi Fan,

First of all, if you disable VTEC, you should also disable tecBiases and set sigmaSTEC=0 in ionosphereSTEC.

As the solutions are determined independently on a daily basis, there is no guarantee that the solutions are continuous. Rather, the deviations reflect the accuracy of the solution. If the deviations are too large, you should also check the quality of the other parameters (e.g. station coordinates) to find possible errors in the settings. Another hint are the statistics of the residuals in the screen output.

best regards
Torsten