Hi @Paul.Gregory,
As you are aware, @Leena and I are working on PGW simulations for the SHEP project with ACCESS-rAM3. We are perturbing ERA5 data for atmosphere initial and boundary conditions, however we’ll also want to manually adjust the SSTs. I understanding we can do this after running the OSTIA ancillary suite. Is there a good way to manually apply a “Delta” to the SSTs in the completed OSTIA ancillary file before returning it back to run in the model?
Cheers,
Sam
Hi Sam, Paul is on leave 
for the rest of this week.
@cbengel may be able to advise, as she did all the work to get rAM3 running with OSTIA SSTs. But I think it’s likely that she will say that modifying model input files (which are fields file format) once they have been created is a tricky process.
Since you’re applying a delta, could you apply it to a local copy of the OSTIA input data before running the OSTIA ancillary suite, and then point the suite to read the modified input data?
Hi @bethanwhite and @sdahl,
yes, making a modified version of the raw OSTIA files is probably the simplest solution for many different reasons.
Let me know if you need help with paths and configs, etc.
Best regards,
thanks @cbengel and @bethanwhite!
I see that SSTs are included in the GRIB files created inside dm_ec_ic_lbc_dir when running the model. Am I correct in assuming that at this stage, I can simply perturb the SSTS here like I would the atmospheric variables? In other words, once the model gets to this stage, the OSTIA files are not used anymore? That would make things a lot easier!
No. There are SSTs in the GRIB files but they are from ERA5.
The OSTIA data goes in through the USE_OSTIA pathway.
For the BARRA/SHEP work though, I have another option for you that I will discuss with you and @Leena separately.