NCAR
Scientific documentation for CCPP-SCM v7.0.0
In some places (e.g., RRTMG documentation), there is a section on CCPP Physics Updates for v6.0.0. I suggest incorporating all v6 and previous documentation into the main body of the documentation. Then, possibly call out any v7 updates (as you have for, e.g., Thompson).
What is the overall strategy for defining acronyms throughout the documentation?
RRTMGP: The description looks good to me, on par with the RRTMG documentation. My only recommendation is to add reference(s). cc @dustinswales
@grantfirl The GF documentations has the blurb below. However, I believe some (or all) of these innovations have not made it to the NCAR/main either because 1) they were meant just for RRFS, and/or 2) they are not in the UFS fork yet. Can you let us know if any of these innovations is appropriate to list her
The Implementation of GF in RRFS prototypes
- Updates for aerosol-awareness (experimental)
- Scale-awareness is turned off when explicit microphysics is not active anywhere in the column
- GF is completely turned off at grid points when MYNN produces shallow convection at that point
- Radar reflectivity considers mass flux PDF as well as whether scale-awareness is turned on at the grid point in equation.
I do not understand why so many schemes in the SciDoc have "GFS" in front of their name, such as "GFS NoahMP LSM". Noah MP is not used in GFS and is a target to all UFS applications. Can we reduce the presence of "GFS" to communicate the message that the CCPP is interoperable?
@grantfirl The GF documentations has the blurb below. However, I believe some (or all) of these innovations have not made it to the NCAR/main either because 1) they were meant just for RRFS, and/or 2) they are not in the UFS fork yet. Can you let us know if any of these innovations is appropriate to list her
The Implementation of GF in RRFS prototypes
- Updates for aerosol-awareness (experimental)
- Scale-awareness is turned off when explicit microphysics is not active anywhere in the column
- GF is completely turned off at grid points when MYNN produces shallow convection at that point
- Radar reflectivity considers mass flux PDF as well as whether scale-awareness is turned on at the grid point in equation.
@ligiabernardet I think that all of those made it into ufs/dev (and main). The only PR related to GF that got into the RRFSv1 release branch but not ufs/dev/main is: https://github.com/ufs-community/ccpp-physics/pull/176. It doesn't look like any of those changes correspond to anything listed. They reinstated scale-awareness for the first hour of the simulation for in that PR, but otherwise, it looks like the bullets should still apply. I did not go back and find specific commits/PRs that correspond to those bullet points, but if there is only one PR that is different between the branches, and it doesn't implement any of those points, the implementation must have occurred before the release branch was cut.
Hi Ming, thanks for all your work on this. I have some suggested edits and comments.
- In v6 we had a "modules" page that listed all of the CCPP modules in one place (https://dtcenter.ucar.edu/GMTB/v6.0.0/sci_doc/modules.html). This is linked from the technical doc, so can we generate that for v7 as well?
- One thing I think is sorely missing from the scientific documentation is linking from the module/subroutine name to the page describing the associated physics scheme. For example, if I look at a CCPP suite definition file and see that it contains
sfc_diff, I can put that into the scientific documentation and it brings me to this page with a very brief description of that specific routine, but no link to the physics scheme. In this case it is possible to search for "GFS surface layer scheme" to get more information, but I have to know to look for that, and this page is in the minority: many CCPP routines have no description at all! I understand that this might be too much work for this update, but I wanted to bring it up for the future.- I noticed something odd for one specific scheme: the scientific documentation for v4.0 contains a detailed description and some nice diagrams for the GFS_NSST scheme, yet those seem to be lost in later versions. Can we bring that back for v7?
Mike, I really appreciate your suggestions.
- At some point, it was hidden from the display since we decided to focus on the high level summary for each scheme after consulting with CCPP-physics management group. But let's bring modules page back, since we will refer to it in techdoc.
- For this release, we can work on the description of entry-level routines of physics schemes. I am not sure if it is realistic for any routines, i.e., interstitials and non entry-level physics routines.
- For the GFS NSST, could you specify which feature is missing in [v7]?
Hi Man, please lmk when these items have been addressed and the website refreshed for final review. Tks
- In some places (e.g., RRTMG documentation), there is a section on CCPP Physics Updates for v6.0.0. I suggest incorporating all v6 and previous documentation into the main body of the documentation. Then, possibly call out any v7 updates (as you have for, e.g., Thompson).
- What is the overall strategy for defining acronyms throughout the documentation? I can't figure what should be defined where. Do you want to add a page of acronym definitions? If it is too much to do now, perhaps after the release?
- RRTMGP could use reference(s)
- All schemes overview: Instead of "Others", how about" Lake Models"
- Smoke and Dust should be removed from phys updates since they do not function with SCM
- RUC: There is a list of "physics updates". Are those for v7. If so, please state. If they are older, please incorporate into main documentation.
@mzhangw Regarding the GFS_NSST scheme, it seems to me like the v4.0 page has a more detailed history and scientific description of the algorithm used along with a nice scientific visualization, though I don't know if it's out-of-date or anything like that. It's not a big deal, just something I noticed while browsing older drafts.
@ligiabernardet @mkavulich Since you requested changes, please re-review to see if your concerns have been addressed. With https://github.com/NCAR/ccpp-physics/pull/1077/commits/8442afa56f3bbc8ffe624b0563b1d431663c769a, this now builds OK.