Opened 5 years ago

Closed 3 years ago

#182 closed enhancement (fixed)

Update DGVM

Reported by: jgipsl Owned by: jgipsl
Priority: major Milestone: IPSLCM6.v1
Component: Biogeochemical processes Version: trunc
Keywords: Cc:

Description (last modified by jgipsl)

Dan Zhu has been working with the DGVM and has done modifications in the ORCHIDEE-MICT version. This will now be integrated in the trunk.


Description of the modifications:

See attachment further below written by Dan Zhu.

The ticket #181 describes also modifications needed for the DGVM.

Integration in the trunk:

Black carbon in lpj_fire has been removed completely.

The default value for LPJ_GAP_CONST_MORT has been changed to depend on the parameter STOMATE_OK_DGVM. If STOMATE_OK_DGVM=y then the default value is LPJ_GAP_CONST_MORT=n.

In stomate_vmax.f90, IF (ok_dgvm) is added to protect following modification done,
in ORCHIDDEE-MICT :

       IF (j.EQ.4 .OR. j.EQ.7) THEN
         vcmax(:,j) = Vcmax25(j)
       ELSE

in trunk (pheno_type and leaf_tab is used to be independent of the number of the pfts):

       IF (ok_dgvm .AND. pheno_type(j)==1 .AND. leaf_tab(j)==2) THEN
          ! pheno_typ=evergreen and leaf_tab=needleleaf
          vcmax(:,j) = Vcmax25(j)
       ELSE

After above change in stomate_vmax is done, the modifications in the code do not change the results of the model if dgvm is not activated and if FIRE_DISABLE=y. While activating the fire, the results changes.

Commit in the trunk to come soon...

Attachments (1)

dgvm_modifications_DZhu.pdf (50.4 KB) - added by jgipsl 5 years ago.
Brief explication of modifications for DGVM by Dan Zhu

Download all attachments as: .zip

Change History (9)

comment:1 Changed 5 years ago by jgipsl

Integration done in the trunk rev [2668]. Results are identical as before if setting in run.def:

STOMATE_OK_DGVM= n
LPJ_GAP_CONST_MORT = y
FIRE_DISABLE = y

Still to be done:

  • externalization needed in lpj_gap and lpj_constraints
  • add xios_send for the new variables in lpj_cover, stomate_lpj
  • some more comments needed for the new variables

comment:2 Changed 5 years ago by jgipsl

  • Description modified (diff)

Changed 5 years ago by jgipsl

Brief explication of modifications for DGVM by Dan Zhu

comment:3 Changed 5 years ago by jgipsl

  • Description modified (diff)

comment:4 Changed 5 years ago by jgipsl

Removed special case with restart file rev [2749]

comment:5 Changed 5 years ago by jgipsl

Cleaning done in rev [2944]
One correction in lpj_establish where before:

SUM(veget_max(:,2:9),dim=2)

now changed into:

    ! Calculate the sum of the vegetation over the tree pft's and the number of pft's which are trees
    veget_max_tree(:) = 0.0
    nbtree=0
    DO j = 1, nvm
       IF (is_tree(j)) THEN
          veget_max_tree(:) = veget_max_tree(:) + veget_max(:,j)
          nbtree = nbtree + 1
       END IF
    END DO

Still to be done : add xios_send for the new variables in lpj_cover, stomate_lpj

comment:6 Changed 5 years ago by jgipsl

Rev [2947] :

  • xios_send added for the variables in stomate_lpj. The variables in lpj_cover will not be added using xios.
  • memory optimization on the variable onset_date. Removed one dimension and use julian_diff as the counter for the day in the year.

comment:7 Changed 5 years ago by jgipsl

Bug correction done in [3038] see ticket #215

comment:8 Changed 3 years ago by jgipsl

  • Resolution set to fixed
  • Status changed from new to closed

ORCHIDEE Day 2017: Nicolas Viovy has checked that everything in the ticket has been done.

Note: See TracTickets for help on using tickets.