GFDL fast processes #81
Conversation
|
@kellylittleblackdog @grantfirl @llpcarson @ligiabernardet Don, did you have a look at the code changes by Man? Most of the issues related to the public variables that you mentioned last time are covered here, i.e. the qs_init is moved inside the physics. @mzhangw now that we have the capability to run the _init routines, qs_init should become part of the _init routine of the GFDL MP scheme? |
|
@mzhangw I started looking at this. I will need to make some changes, then I would like to create a pull request against your branch mzhangw:master-gfdl-fast so that this pull request here gets updated automatically. Notes: since you didn't add the source files to the ccpp prebuild configs for FV3 and/or SCM, it is to be expected that it passes w/o errors. Adding it in I get errors about missing variables (e.g. top_layer_index_for_gfdl_mp and more) on the host model side. I will add those missing variables. Also, I will have to combine the two files again, because qs_init is actually fv_sat_adj_init. |
|
Thanks a lot.
I did it a while ago, so I am not surprised that there are loosing parts:)
…On Wed, Jun 27, 2018 at 9:57 AM Dom Heinzeller ***@***.***> wrote:
@mzhangw <https://github.com/mzhangw> I started looking at this. I will
need to make some changes, then I would like to create a pull request
against your branch mzhangw:master-gfdl-fast so that this pull request here
gets updated automatically.
Notes: since you didn't add the source files to the ccpp prebuild configs
for FV3 and/or SCM, it is to be expected that it passes w/o errors. Adding
it in I get errors about missing variables (e.g.
top_layer_index_for_gfdl_mp and more) on the host model side. I will add
those missing variables.
Also, I will have to combine the two files again, because qs_init is
actually fv_sat_adj_init.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#81 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/Ab87amQMlAk6GA7GW0YgEoWe6obEt5Lyks5uA6tZgaJpZM4UFBMk>
.
|
|
The changes for making this work in FV3 are rather substantial, I think I will create a new PR instead of trying to update yours. I will make sure to credit you for your work and add you as a reviewer, though ;-) |
|
This PR is superseded by #114 |
Add diagnostic tendencies for all tracers. Move all d*3dt variables into a 4D sparse array stored as `dtend(i,k,dtidx(tracer,process))` to reduce memory usage.
fv3atm:
fractional landmask changes (from @shansun6)
ccpp-framework:
bugfixes for handling optional arguments (no impact on current setup)
bugfix for debug messages (no impact on current setup)
ccpp-physics:
GFSv16 physics updates
add moninedmf_hafs
scientific documentation for NoahMP and satmedmfvdifq
bugfix for threading (avoid dead lock) when calculating Thompson MP lookup tables
addition of radar reflectivity reset switch for Thompson MP, bugfix for non-aerosol Thompson runs (from @ericaligo-NOAA)
fractional landmask changes (from @shansun6)
CCPP version of RAS, updates to shoc and MG (from @SMoorthi-emc)
bugfixes when levr < levs (from @SMoorthi-emc)
Rollback changes to rain evaporation
GFDL fast processes is added. it has passed run ccpp_prebuild.py using gmtb-fv3 release version on cheyenne. see
/glade/p/work/manzhang/_gmtb-fv3-May2018/gmtb-fv3/FV3/gfsphysics