CCPP Framework Meeting Minutes 2022 08 30

Agenda

• GitHub issues
• Other business

---

GitHub issues

• None new, see "Other business" discussion

Other business

• Discussed issue priority list at last week's DTC CCPP meeting, in no particular order:
  • "Blocked" data structures (#314)
  • Implement debug switch (#325)
  • Implement "active" attribute (#328)
  • Limit cap name and subroutine name length (#337)
    • Needs some work on the implementation side: decide exactly how to do a robust truncation
    • Mike Kavulich will work on coming up with a strategy
  • Fix error in parsing Doxygen comments in Fortran (#369)
    • Already done!
  • DDT metadata setup (#377)
    • Chunked data is not yet implemented in capgen, but need to solve this before it can be implemented.
• This is an incomplete list, but hopefully can be used for setting priorities if/when time/resources are available for capgen development

Discussion notes

• Discussed issues above, assigned some work
• Hopefully not too much bad news about leadership leaving
• SIMA review: Can we contribute to cost estimates?
  • Questions:
    • How much for implementation?
    • How much saved by implementation?
      • Estimated ~1 FTE constant
  • We need funding/hours to do the work needed to come up with these estimates
    • Need to break down into parts:
      • Physics conversion
      • Full system testing/implementation
    • Ligia believes it's possible to make a rough estimate
      • Obviously huge uncertainties based on who exactly has time/resources to undertake this effort
      • Estimates based on initial UFS effort may be biased high, since we now have experience and learned lessons
      • Schemes already in the model are (maybe?) easier than brand new schemes
    • Dustin used 0.5 FTE for 2 years for putting RRTMGP into the CCPP, but this funding also covered more than just that effort
  • Question from Hanli (relayed by Lulin) For WACCM-X, the thermosphere physics and ionospheric chemistry are mostly column based, and can be organized in a CCPP Suite Groups. However, the new ionospheric electrodynamics and ionospheric transport are solved along geomagnetic field lines, which cut through the geographic columns. Is there a way to organize them in geomagnetic field line “columns”? If not, can the host model (CESM/WACCM-X) be partially CCPP compliant and skip over the ionosphere?
    • Need more discussion to decide whether it makes sense to adapt into CCPP compliance
• Will meet again in 2 weeks