Who uses the code in driver/fvGFS/atmosphere.F90
Opened this issue · 4 comments
Is your question related to a problem? Please describe.
I am looking at reorganizing the data storage model for the GFS data types in the UFS / FV3ATM. Doing that requires making changes in driver/fvGFS/atmosphere.F90. I am wondering who else than the UFS or other models that inherited the GFS data types (e.g. because of IPD first, then CCPP) uses this code. In other words, if I am making changes there, do I need to wrap all of them in #ifdeg GFS_TYPES?
If nobody outside the UFS-CCPP world uses this part of the code (somewhat suggested by the subdirectory fvGFS), then my life will be a lot easier and we can also simplify the code quite a bit (less ifdef = better).
Describe what you have tried
n/a
The HAFS moving nest code in the FV3/moving_nest directory uses the IPD structures, especially in moving_nest/fv_moving_nest_physics.F90. I am happy to discuss further with you to coordinate any needed changes here.
@climbfuji - each modeling system will have its own version of atmosphere.F90. If you look at the main branch, you'll see versions for SHiELD, GFDL climate, and a solo models. While the name is still driver/fvGFS, driver/UFS would probably now be more appropriate.
Thanks for your quick feedback @bensonr and @wramstrom. So, if I understand you correctly, as long as I make sure that driver/UFS/* (I am going to rename that as part of the work) works for any of the UFS applications using the FV3 dynamical core, then we are ok; I do not need to worry about other modeling systems so long as I only touch the files in these directories.
@climbfuji - yes, that is correct.