Groops PPP Example with recent GNSS Data (OBS and SSR)

[jiargei]

I tried to reproduce the example for PPP (https://groops-devs.github.io/groops/html/cookbook.gnssPpp.html). By running the epoch 2020-01-01 it works fine and the result (Station Coordinates GRAZ) align with those of the Combined solution from the IGS Datacenters:

I tried no to adapt the XML and instead of SP3, CLK, BIA, etc. from TUG i used COD0MGXFIN for the week 2409 (2026:067 .. 2026:073) Unfortunately, by just changing the input, the coordinate is wrong by several centimeters:

= 2020-001 GRAZ =
TUG0GROOPS_2020001 4194423.5568  1162702.9583  4647245.5664
EUR0OPSSNX_2020001 4194423.5632  1162702.9607  4647245.5676

= 2026-067 GRAZ =
TUG0GROOPS_2026067 4194423.3801  1162702.9798  4647245.6163
EUR0OPSSNX_2026067 4194423.4606  1162703.0739  4647245.6345

Are there some further information, details i have to change?

• TUG -> COD
• timeStart = 61107
• timeEnd = 61113

I'm running GROOPS with MPI on Fedora 44, WSL under Windows.

[jiargei]

One thing I already found in the 02 xml... It uses igsR3 which is some version of igs14! replace it to igs20... the coordinates do still not fit

[jiargei]

IT would also be nice to get a kind of summary for the results, not just a position list. which is probably needed for the plot generation

[tmayerguerr]

Hi,

PPP with ambiguity resolution (PPP-AR) is a little bit tricky, as all the metadata has to match up. In particular, only signals/frequencies for which bias data is provided should be used (GnssProcessing:receiver:useType)

I've put together a scenario based on the COD0MGXFIN, see scenarioGnssPPP_COD0MGXFIN.zip. GLONASS cannot be used because CODE uses a different method to apply the bias than GROOPS. QZSS has not been tested.

Please update your GROOPS from the main branch as I corrected BiasSinex2SignalBias.

Best regards
Torsten

[jiargei]

Dear Torsten!

The updated version and the new bias process delivers very good solutions for GRAZ, PFA3, SBG2 and TRF2 (weekly comparison to EPN combined ~= 3-5 mm!)

You mentioned that CODE handles GLONASS bias different to GROOPS. Is there any other FIN product that aligns with GROOPS' bias handling? (IGS0OPS, IGS0DEM, etc.)

Best regards,
Jürgen F.