Leaf inclination angle in parameter file is different from that documented in https://github.com/ESCOMP/CTSM/issues/807

[olyson]

Brief summary of bug

The leaf inclination values in this parameter file, /glade/p/cesmdata/cseg/inputdata/lnd/clm2/paramdata/ctsm51_params.c210923.nc,

agree with the values implemented in #807

They have different values in the latest parameter file used in the recent coupled simulations,
/glade/campaign/cesm/cesmdata/cseg/inputdata/lnd/clm2/paramdata/ctsm60_params_cal115_c250813.nc

Other leaf and stem optical properties are the same in the two parameter files.

Does anyone have any knowledge of intentional changes in leaf inclination values? Was this maybe a result of PPE tuning?

[wwieder]

@linniahawkins, is this one that's being modified in PPE work?
Also pinging @samsrabin, maybe you know something here?

[linniahawkins]

Is this the xl parameter? Yes, it is being tuned in the PPE. However, we opted not to have albedo as a constraint in the calibration so for the most part it stayed near the default. It looks like it was increased for PFT 4 and 11 by the calibration.

I think it only has a small effect on the C variables we tuned so if it's important to keep this parameter at default I can recalibrate.

[samsrabin]

I got nothing on this, sorry.

[olyson]

Yes, the xl parameter, thanks. I see the following differences in xl between these two parameter files (clm6 - clm5.1):

xl = 0, 0, -0.16, 0, 0.13, 0, 0, 0, -0.36, -0.31, -0.34, -0.34, 0, 0, 0,
-0.1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ;

So, more than just PFTs 4 and 11.
I'm not saying there is anything necessarily wrong with these changes, I just ran across this when I was investigating the RRTMGP surface temperature error, and didn't know the source of these changes. Maybe some of these were changed before the PPE...

[linniahawkins]

Yes, I made these modifications to PFTs 2,4,11. I'm a little surprised by the differences in PFTs 8,9,and 10 but I can confirm that I made these changes (the default parameter set I used at the start of the PPE did not have these changes). I'll dig into where this happened in my workflow and we can back out these changes if appropriate.

[linniahawkins]

Okay I've traced these differences to a step where I un-normalized the default parameter settings. I'll correct this in the next iteration of calibration.