Opened 4 years ago

Closed 3 years ago

#604 closed defect (fixed)

Need some cleaning of the variables and keywords for soil properties in thermosoil

Reported by: aducharne Owned by: aducharne
Priority: major Milestone: Not scheduled yet
Component: Anthropogenic processes Version: trunc
Keywords: Cc:


1) In thermosoil, we can find three different variables for the same thing = porosity :

  • mcs, which is the same as in hydrol, and changeable owing to keyword VWC_SAT;
  • smcmax, whic takes the same values as mcs in constantes_soil_var, but is not modified if mcs is changed by VWC_SAT;
  • poros, which is an inheritance of thermosoilc (Choisnel path), and is always equal to 0.41 unless modified by keyword POROS.

2) The WET_SOIL_HEAT_CAPACITY and WET_SOIL_HEAT_COND keywords have no effect in thermosoil, since the wet values of heat conductivity and capacity are calculated based on soil moisture.
It might good to prevent from using them in this case.

3) I think that DRY_SOIL_HEAT_CAPACITY and DRY_SOIL_HEAT_COND have no effect either, although the corresponding variables are used in thermosoil.
These keywords define so_capa_dry and so_cond_dry, which only serve for the last soil layer in (so_capa_dry_ns and so_capa_dry_ns), and so_capa_dry serves to define so_capa_ice in thermosoil_initialize

so_capa_ice = so_capa_dry + poros*capa_ice*rho_ice

The really effective values for these dry parameters are so_capa_dry_ns and so_capa_dry_ns, which are not updated based on the keywords in thermosoil_initialize.

4) The following line in thermosoil_getdiff is useless, as cap_iw is nowhere used:

cap_iw = xx * so_capa_wet + (1.-xx) * so_capa_ice

5) We discussed several times in ORCHIDEE meeting that the quartz content (QZ) could be deduced from the fraction of sand instead of a specific table in constantes_soil_var (QZ_fao or QZ_usda).

The work on the above issues, in away that makes the use of the keywords tractable for optimization would be a good opportunity to code this. Catherine should be involved.

Change History (8)

comment:1 Changed 4 years ago by aducharne

  • Milestone set to ORCHIDEE 3.0

comment:2 Changed 4 years ago by aducharne

Some of these changes (1 and 3) have been done in trunk [6058] : "changed comments and replaced poros and SMCMAX by mcs", as written in ticket #423 (closed).

But the other concerns remain, especially if one want to use the external keywords to define some parameters, for calibration for instance.

Finally, couldn't we propagate these changes in tag 2.1?

comment:3 Changed 4 years ago by jgipsl

4) Done here [6499]

comment:4 Changed 4 years ago by aducharne

  • Milestone changed from ORCHIDEE 3.0 to ORCHIDEE 4.0
  • Owner changed from somebody to aducharne
  • Status changed from new to accepted

Bilan le 28/01/2020:
Point 1: le commit 6058 ne concerne que la variable poros, il faut aussi traiter les liens entre mcs et smcmax (Agnès)
Point 2: Josefine va le faire prochainement
Point 4: fait dans 6499
Point 5: par Agnès, avec vérif par Fabienne

On décide aussi que ces changements devraient être propagés à 2_2, car ils entrainent des bugs si on veut modifier les paramètres thermiques via les keywords externalisés

Last edited 4 years ago by aducharne (previous) (diff)

comment:5 Changed 4 years ago by jgipsl

Point 2 done in [6505]

comment:6 Changed 4 years ago by luyssaert

  • Milestone changed from ORCHIDEE 4.0 to Not scheduled yet
  • Version set to trunc

comment:7 Changed 3 years ago by aducharne

The changes made to the trunk (r6499, r6505, r6508) have been implemented in branch 2.2 by r7199.

The last pending question is about the quartz content definition in thermosoil, transferred to a new ticket #778.

Last edited 3 years ago by aducharne (previous) (diff)

comment:8 Changed 3 years ago by aducharne

  • Resolution set to fixed
  • Status changed from accepted to closed
Note: See TracTickets for help on using tickets.