Gridcell mean z0m

[tcraig]

To couple CTSM with WRF, I need to compute the gridcell CTSM momentum roughness length. I have looked thru 2.5. Momentum, Sensible Heat, and Latent Heat Fluxes — ctsm CTSM master documentation as well as the code. From what I can find, some portion of those variables can be found in the frictionvel_type in src/biogeophys/FrictionVelocityMod.F90,

real(r8), pointer, public :: z0mv_patch (:) ! patch roughness length over vegetation, momentum [m]
real(r8), pointer, public :: z0mg_col (:) ! col roughness length over ground, momentum [m]

I know I can aggregate those variables to the gridcell level using the standard CTSM interfaces. My questions are

- Is z0mv_patch just the value over vegetation and z0mg_col just the value over ground? And if so, how do I combine them and are there any other terms or surface types that I need to take into account when aggregating to the gridcell?
- If I do have to combine multiple terms, is that a simple sum, do I need to sum the logs or exp of these values? Do I need to weight them some special way?
- When I aggregate patches or columns to the gridcell level, will the standard CTSM tools (i.e. p2g) work. Again, I'm worried that I may need to aggregate the log or exp of the values and then compute the exp or log of the value to get the proper weighted average over the gridcell.

I'm approaching this as a software engineer and this is not my area of expertise. Thanks!

 

[oleson]

Hi,
I guess I thought this had already been addressed through this issue:

Send roughness length to LILAC · Issue #892 · ESCOMP/CTSM

WRF needs gridcell-average roughness length from CTSM. It looks to me like there are a few different roughness length variables in CTSM, and I'm not completely confident in my interpretation of how...

github.com

Or maybe this is being revisited?

 

[tcraig]

Thanks Keith. I think this is exactly what I need. I have a slightly older version of the code (from March). It looks like this mod was added to the lilac_cap branch in January, do you know when those code mods were merged to the master?

 

[sacks]

ctsm1.0.dev104 (07/06/2020)

Tony, I think this is for the RASM project, where you are coupling CTSM with WRF via cpl7. Is that right?

I partly want to avoid confusion for anyone else reading this thread, because CTSM has now been coupled directly to WRF in changes that will be coming to WRF's develop branch very soon (Coupling CTSM with WRF by negin513 · Pull Request #1275 · wrf-model/WRF).

 

[tcraig]

Thanks Bill and that is correct. My question is associated with a CESM1/RASM coupled model application using cpl7 for coupling. Ultimately, I think we need the same "gridcell average momentum roughness length" in our application for coupling with WRF as you have in the WRF/CTSM Lilac system. But they are completely different applications. I agree it's good to be clear on that point.