Changeset 4071 for trunk/LMDZ.COMMON/libf/evolution/clim_state_rec.F90

Timestamp:

Feb 16, 2026, 10:28:56 AM (12 days ago)

Author:

jbclement

Message:

PEM:

• Making the computation of ice tendencies more reliable by doing it after 'read_startpem' to know exactly where perennial ice is (no matter if there is a "startpem.nc" or not). Moreover, when there is no "startpem.nc", location of perennial ice depends now on 'watercaptag' and on huge amount of frost. This prevents negative ice tendency while there is no ice which can happen with weird PCM inputs (i.e. 'watercaptag' = F & 'watercap' /= 0 & no "stratpem.nc").
• Few small safeguards throughout the code.

JBC

File:

• trunk/LMDZ.COMMON/libf/evolution/clim_state_rec.F90 (modified) (3 diffs)

trunk/LMDZ.COMMON/libf/evolution/clim_state_rec.F90

@@ -137 +137 @@
 ! DEPENDENCIES
 ! ------------
-use geometry,   only: nlayer, nslope
+use geometry,   only: nlayer
 use atmosphere, only: teta_PCM, u_PCM, v_PCM
 use tracers,    only: nq, qnames
@@ -155 +155 @@
 ! LOCAL VARIABLES
 ! ---------------
-integer(di) :: funit, ierr, i, j, l
+integer(di) :: funit, ierr, i, l
 
 ! CODE
@@ -244 +244 @@
 
 ! Variables that have been modified
-call put_var_nc('watercaptag',merge(1.,0.,is_h2o_perice))
+call put_var_nc('watercaptag',merge(1._dp,0._dp,is_h2o_perice))
 call put_var_nc('watercap',h2o_ice4PCM,1)
 call put_var_nc('h2o_ice',h2o_frost4PCM,1)