Modify the code and simulate with BSSP585

[zjulin]

Hello, I encounter a problem when I try to simulate.

I would like to modify part of the POP module code (e.g. modifying the parameterised model) to carry out a simulation with BSSP585 compset. My idea is to do simultaneous modification of BHIST and BSSP585, and then use the last month result of BHIST output as the restart file of BSSP585. Is my idea correct? If correct, would my BHIST have to be simulated from 1850 to the end of 2014?

I sincerely hope to get your advice. Thanks a lot

 

[mlevy]

I'm sorry, but I'm not following -- could you explain your experiment a little more? It sounds like you want to modify the ocean model and study the effect this has in an SSP-5.85 simulation. If that's the case, I would recommend running a BSSP585 case without your modifications as a control run, and then comparing to a BSSP585 case with your source mods. I don't think you need to run the BHIST at all, you can start a BSSP585 case using initial conditions generated on the NCAR machine.

 

[zjulin]

I'm sorry, but I'm not following -- could you explain your experiment a little more? It sounds like you want to modify the ocean model and study the effect this has in an SSP-5.85 simulation. If that's the case, I would recommend running a BSSP585 case without your modifications as a control run, and then comparing to a BSSP585 case with your source mods. I don't think you need to run the BHIST at all, you can start a BSSP585 case using initial conditions generated on the NCAR machine.

Thank you for your answer. I apologize for not being clear enough.
I actually want to modify the Marine Biogeochemistry Library code in POP, not just some initial field files. So I think I should run BHIST first, and then use the last month of BHIST data as the initial conditions to start BSSP585. Just like you said, BSSP585 needs an initial conditions, and if I modify the code parameters and don't run it from BHIST, then I can't provide my own BSSP585 with the initial conditions I need.
Thanks for your advice!

 

[mlevy]

I'm going to move this conversation to the ocean BGC forum; that seems like a better place for this discussion.

Have you tried running a BSSP585 simulation without any modifications? I believe CESM will download all the necessary initial condition files (which should just be restart files from the end of a BHIST control run done at NCAR). If your MARBL modifications require changes to the initial condition file, then you'll need to make those same changes whether you are running BSSP585 or BHIST so I don't see the benefit of doing the BHIST simulation.