ccpp-physics
ccpp-physics copied to clipboard
NCAR
Update Tiedtke scheme with MPASv8.0.0 updates
The updates include: introducing scale dependency factors (Wang 2022, WAF), updated entrainment coefficients, vertical indexing fix, and horizontal wind tendency term (pgf_u, pg_v) calculation also for shallow convection not just deep and mid- convection, cleaner fields declarations.
Issue https://github.com/NCAR/ccpp-physics/issues/1024
@matusmartini Have these changes been tested and evaluated for scientific impact in any UFS application?
@lisa-bengtsson This scheme is orphaned in CCPP, and it sounds as if there is an interest to try it with FV3? We tested with NEPTUNE at various configurations and found improved skill especially in lower troposphere. More info and discussion on this is in #1024
According to the discussion in the 8/2/2023 CCPP Physics Code Mgmt meeting, we will delay the final review of this PR until Wei Wang (who did the Tiedtke updates for MPAS v8) is back in mid September 2023.
According to the discussion in the 8/2/2023 CCPP Physics Code Mgmt meeting, we will delay the final review of this PR until Wei Wang (who did the Tiedtke updates for MPAS v8) is back in mid September 2023.
Thank you @ligiabernardet for the discussion yesterday and @grantfirl for all of the above comments. The goal of my initial commit was to point out the differences between orphaned Tiedtke scheme in CCPP and seemingly cleaner version in MPAS with the scientific updates without imposing sense of urgency. These changes simply attempt to sync as much as possible with MPAS version while leaving MPAS-specifics out (for example units of latent and sensible heat, units of pressure, humidity).
If the ultimate goal is to have ONE Tiedtke scheme across MPAS and CCPP then some of @grantfirl suggestions (such as the one about assumed shape of argument variable arrays) would have to propagate into MPAS, bridging the two... In addition, it would be nice to avoid duplicate efforts with MPAS shared physics driver @dustinswales going forward.
In the interim, proceeding with these changes (after review) would bring the CCPP version closer to the MPAS version. Alternatively, I could just bring the very minimal set of updates and these could just be the scientific ones and bugfixes. Perhaps that would be a better start, when @weiwangncar is available, and would also make the review much easier if there is preference for that. @yangfanglin
