-
Notifications
You must be signed in to change notification settings - Fork 64
CCPP Framework Meeting Minutes 2024 12 19
Courtney Peverley edited this page Dec 19, 2024
·
4 revisions
Attendees: Michael Kavulich, Cheryl Craig, Dustin Swales, Jesse Nusbaumer, Dom Heinzeller, Courtney Peverley, Michael Waxmonsky, RhaeSung Kim, Jordan Powers, Soren Rasmussen
CCPP Framework (issues, PRs, discussions)
Standard names (issues, PRs, discussions)
-
Update StandardNamesRules.rst: add _due_to_planetary_boundary_layer #80
- Merged (with updated commit message)
-
Add/modify skin temperature at surface variables. #84
- Pinged Steve Vahl to ask the status
-
- Had some conversation and updated suggested changes, waiting to hear back
-
- PR to
release/v1
branch
- PR to
Offer suggestions/comments on development workflow rules:
CCPP Framework
- Bufix for variables in multiple groups (#618)
- Dustin to review
- Testing refactor (#612)
- Michael W to keep targeted to develop branch; get up to date
- Still draft, but would appreciate informal feedback on design
Standard Names
- Add due_to_PBL (#80)
- merged
- add/modify skin temperature at surface variables (#84)
- Change sea => ocean; haven’t heard from Steve Vahl
- Add emissions names (#86)
- Michael K iterating with Jerome
- Jesse - another one of us can chime in if there’s more push back and Michael K needs backup
- Cheryl - would like to get feedback from NCAR chemistry group (ACOM); pinging Matt Dawson
- Second rules update; update write_standard_name_table.py to allow for subsections (#87)
- People to take a look
- Two identical quantities from different schemes/processes (issue #79)
- Dom added suggestion from Cory
- Jesse ok with whole atmosphere verbiage
- Dom to ping UFS’s gravity wave drag guy
- Dustin - weird to distinguish when they’re both operating on the same vertical grid
- Settled on cumulative_change_of_x_wind_due_to_convective_whole_atmosphere_gravity_wave_drag
Discussion
- Prebuild doesn’t have variables defined by framework; have to track variables in host model that are purely output. Capgen can handle those, but…
- Won’t work with multi-instance capgen (workaround is to define the variables on the host side)
- Related issue: #561; added to “capgen unification” project
- Standard names repo name change (discussion/announcement #88)
- Jesse suggests adding a deadline / when we will make the change
- We want a “speak now or forever hold your peace” situation
- Possible deadline of 1/23
- Michael K to create issue (include instructions on how to update .gitmodules, etc)
- CCPPStandardNames => ESMStandardNames
- Dom - would recommend putting standards into workflow documentation about when a PR is draft vs not - document
- Dom - agree on a time frame for PR review & responding to PR review requests
- Michael K - thinks we previously decided 1 week for PR reviews
- Jesse - maybe we have 1 week for high priority bugfixes and 2 weeks for other PRs (or something of that ilk); or maybe that’s just unofficial
- Timeline
- UFS
- Dustin working on contiguous memory in MPAS
- Some schemes have different bounds (not just (:,:) for example)
- Needs fix
- Need to document changes & have time to implement changes in framework
- Hoping to get close next month
- Might go into GFS17
- Acronym clarity: GFS is like a UFS configuration (global configuration of UFS)
- NEPTUNE
- End of september is the goal
- UFS