Hitting runtime fail in canopy resistance error check.

[rosiealice]

Hi all,

In the process of tuning the arctic shrubs, broadly, when the growth of the shrubs becomes more marginal, either for parametric reasons or at e.g. 1850 CO2, I am hitting this error:

Combined canopy resistance was somehow smaller than
15: its boundary layer resistance component
15: r_sb_leaves [s/m]: 22.7443527131291
15: bc_in(s)%rb_pa(ifp) [s/m]: 47.4918907756017
15: ENDRUN:

So, questions:

• Any thoughts on why this might happen? My suspicion is that something is up with the layering system during the course of slow slow death process?
• Do we think it is a total disaster if this happens occasionally? If I turn off the endrun statement it happens maybe 100 time in the course of a year of 4x5 simulation, which is a very small fraction of the total.

Maybe we can discuss it a bit on the FATES call (Oct 13)?

[glemieux]

@rgknox notes during the software meeting call today that the stomatal conductance would have to be negative. He suggested looking at individual leaf level stomatal conductances.

See for the relationship between bc_in(s)%rb_pa (her rb):

fates/biogeophys/FatesPlantRespPhotosynthMod.F90

Line 1287 in a5c3491

g_sb_laweight = g_sb_laweight + 1.0_r8/(rs_llz(il)+rb) * cohort_layer_eleaf_area

@rgknox reviewing medlyn code to make sure we can't get negative conductance.