Adding more diagnostics for help with cloud fraction scheme(s) #283
Conversation
…igned with Thompson-MP for GFSv17-prototype8
There was a problem hiding this comment.
The changes add the required variables to the GFS data structure and the CCPP metadata, but I am wondering if there are additional steps needed to output these variables in the single column model? Or has this not been done, because you were using the UFS in your tests?
| @@ -6893,7 +6911,7 @@ subroutine diag_phys_zero (Diag, Model, linit, iauwindow_center) | |||
| Diag%dku = zero | |||
|
|
|||
| ! max hourly diagnostics | |||
| Diag%refl_10cm = zero | |||
| Diag%refl_10cm = -35. | |||
There was a problem hiding this comment.
This change makes sense to me, but I want to check with @grantfirl if there was a particular reason to reset Diag%refl_10cm to zero instead of -35.
There was a problem hiding this comment.
As FYI to this point. Many people think dbz=zero means no precip. That is not true. Very light drizzle occurs all the time with dbz of about -5 to -10 values. The -35 value is considered (approximately) the lowest sensitivity level of the USA's NEXRAD radars. Cloud radars are sensitive down to even smaller values, but the frequency/wavelength of NEXRAD is 10cm, hence why I was so specific in its name.
There was a problem hiding this comment.
No, @gthompsnWRF reasoning makes perfect sense. This was only zero in the SCM version of GFS_typedefs to match FV3.
That's correct. I added these 6 new variables to GFS_diagnostics.F90 to get them for UFS and not really sure what/where to place them in SCM - unless they go in scm_output.F90. |
|
The new variables need to be added to scm_output.F90 only. I can do this for you @gthompsnWRF (I'll branch off of your PR branch, make modifications, and submit a PR back into your branch so that it will be reflected here once accepted by you). It's pretty straightforward, but there could be a complication or two depending on what is being held in these variables. The Diag DDT in GFS_typedefs holds variables that can be instantaneous, accumulated, etc. and that needs to be accounted for in how they're written out in scm_output.F90. |
…able output on radiation timesteps; fix rt.sh not correctly reporting differing output values
Update NCAR#283 code to latest main + output of new variables
|
This PR is superceded by #285, so closing this one. |
Wrapper for #283, fix RT bug, add radiation output interval
Update #283 code to latest main + output of new variables
Wrapper for #283, fix RT bug, add radiation output interval
Six new variables - 2D only - were added to support cloud fraction schemes to diagnose impacts of explicit cloud condensate versus impacts from cloud fraction. There is a corresponding branch with the same name in the ccpp-physics repo to include the new calculations together with a new/updated cloud fraction scheme.