ESCOMP
A new physically based dust emission scheme with more aeolian physics
Description of changes
This is a scheme that builds upon issue #1230 and #1604 which switched CESM2's default dust emission scheme (Zender et al., 2003) to a more physical and less empirical one (Kok et al., 2014). Based on #1230 's edits, #1604 's modifications add new aeolian physics to the Kok's scheme, most notably, by adding the roughness effect (or called drag partition effect) which discounts surface soil erosion by winds due to the presence of local-scale land-surface roughness elements (mostly plants and rocks). We use a hybrid approach to account for both roughness from rocks (with a 2-D time-invariant dataset we provide) and roughness from plants (time-varying, as a function of CLM's LAI). We further include the dust emission intermittency effects due to boundary-layer turbulence.
Changes in files mainly include: clm/src/biogeochem/DUSTMod.F90 clm/src/biogeophys/SoilStateType.F90 clm/src/biogeophys/SoilStateInitTimeConstMod.F90 clm/src/main/controlMod.F90 clm/src/main/clm_varctl.F90 clm/bld/namelist_files/namelist_definition_clm4_5.xml user_nl_cam user_nl_clm
Specific notes
-
One interesting thing is that CAM applies a source function (for Charlie Zender’s scheme) to the CLM dust emissions. In CESM, the dust emission calculation isn’t finished in CLM and is further processed in CAM before CAM calculates the transport. Because of this, when changing from Zender's scheme to Kok’s scheme one also needs to comment out the code block in CAM relevant to the source function. The relevant changes should be made in this file: cam/src/chemistry/modal_aero/dust_model.F90
-
Two files below were changed when coupling to the atm model as a trick to avoid the very occasional occurrence of some negative downwelling LW radiation. The two files should ultimately not be merged with the master branch but are put here for documentation: clm/src/main/lnd2atmMod.F90 clm/src/cpl/lnd_import_export.F90
Contributors other than yourself, if any: @ekluzek
CTSM Issues Fixed (include github issue #): Fixes #1604 Are answers expected to change (and if so in what way)? Only
Any User Interface Changes (namelist or namelist defaults changes)? Yes. For now, mainly clm/bld/namelist_files/namelist_definition_clm4_5.xml
Testing performed, if any: The changes above were used to simulate an F case (F2000climo) coupling atm and lnd. They have not been tested following the guideline yet.
The surface roughness file used is here on cheyenne:
/glade/u/home/dleung/CESM2/myinput/nas_cleared/F_eff_r_static_mean_allLAI-09x125-12172021.nc
This is based on the 2005 Prigent data which was at a quarter degree and interpolated using area weighted mean. The 2012 Prigent data is used in the Ronny Meir work and is at 6km resolution.
@ekluzek here are my thoughts (italics) to some of your questions (in bold) from email: Should the Prigent data be handled as a stream file or on the surface dataset?
- Which is easier to maintain? Assuming this is a static input to the model, are these typically handled by surface data (vs. streams which I think of as time evolving).
Which Prigent data should we use for the end result: Meier, Leung, or original Prigent?
- I'd say the original, assuming Danny's satisfied with the results.
Should we bring in Prigent dataset all at once with the Dust work, or as an incremental improvement after the resolution dependent version is brought in?
- Which option will let Danny keep making progress more quickly AND end up be less work for you?
I'll let CAM and CAM-CHEM people respond to questions about what parameterizations and name list options we want to maintain, especially for paleo configurations.
We have a new branch for this work based off the latest development version of the model ctsm5.1.dev106 (dustemis_dev). I moved the changes here to it by doing the following...
git diff release-clm5.0.34 > dust.diff
$EDITOR dust.diff # To correct the paths for namelist_definition and lnd_import_export.F90...
git apply -3 dust.diff
$EDITOR src/biogeochem/DUSTMod.F90 src/biogeophys/SoilStateInitTimeConstMod.F90 src/cpl/mct/lnd_import_export.F90 src/main/controlMod.F90. # Fix the conflicts
After we get that branch working and in place for a PR, we'll close this one.
@dmleung has successfully run a case with CAM for the new development version branch, and shown that the results are similar to the previous clm5.0 case. This will allow us to open a new PR for that branch, and close this one. The new branch is very close to the very latest development version based off of ctsm5.1.dev106 which is only 7 tags behind the latest. This is all very good news.
Here's the plot he gave me...
