Open adamrher opened 3 weeks ago
@PeterHjortLauritzen @adamrher Is this a bug that needs correcting in prior versions of the code, or did it creep in recently?
Hi Dan, this bug came in when implementing sponge layer diffusion on the physics side, in cam6_3_151 https://github.com/ESCOMP/CAM/pull/968.
What happened?
The sponge layer diffusion coefficients for momentum set in src/physics/cam/vertical_diffusion.F90 are not being applied to the sponge layer. This bug impacts mid-top, WACCM and WACCM-X configurations. In the following loop adding
kvm_sponge
tokvm
:kvm
has a hard coded level index of 1, where it should instead be k.A separate issue with this code is that the
kvm
array is on the pverp grid, meaning level 1 is the top of the model, and are therefore of no consequence because thefin_vol_lu_decomp
diffusion solver code uses a fixed or zero flux boundary condition. This implies the values ofkvm_sponge(1)
are also not of any consequence, even after fixing the bug. Running two simulations and varyingkvm(:ncol,1)
, the answers are bfb, confirming that the model topkvm
values are not used.What are the steps to reproduce the bug?
Run FMTHIST or a WACCM/-X compset
What CAM tag were you using?
cam6_4_038
What machine were you running CAM on?
CISL machine (e.g. cheyenne)
What compiler were you using?
Intel
Path to a case directory, if applicable
No response
Will you be addressing this bug yourself?
Yes
Extra info
@PeterHjortLauritzen