Changeset 1930 for trunk/LMDZ.MARS/libf/phymars

Timestamp:

May 31, 2018, 7:35:36 AM (6 years ago)

Author:

aslmd

Message:

LMDZ.MARS: some compilers complain about argument mismatch when calling updaterice_microco2. This prevents the MESOSCALE model for MARS to compile with ifort for instance. This is a hack that is harmless to GCM compile in double precision, that fixes the problem in MESOSCALE compile in single precision, but that does not solve entirely the problem.

File:

• trunk/LMDZ.MARS/libf/phymars/improvedCO2clouds.F (modified) (2 diffs)

trunk/LMDZ.MARS/libf/phymars/improvedCO2clouds.F

@@ -113 +113 @@
       DOUBLE PRECISION mem_Mh2o_co2(ngrid,nlay) ! Memory of H2O mass integred into CO2 crystal
       DOUBLE PRECISION mem_Nccn_co2(ngrid,nlay) ! Memory of CCN number of H2O and dust used by CO2
-      
+      DOUBLE PRECISION interm1,interm2,interm3     
+ 
 !     Radius used by the microphysical scheme (m)
       DOUBLE PRECISION n_aer(nbinco2_cld) ! number concentration volume-1 of particle/each size bin
@@ -527 +528 @@
            No   = zq(ig,l,igcm_ccnco2_number)* tauscaling(ig)+1.e-30
            IF (No .ge. 1)THEN   ! we trigger crystal growth
-              call updaterice_microco2(zq(ig,l,igcm_co2_ice),
-     &            zq(ig,l,igcm_ccnco2_mass),zq(ig,l,igcm_ccnco2_number),
+              interm1 = DBLE(zq(ig,l,igcm_co2_ice))
+              interm2 = DBLE(zq(ig,l,igcm_ccnco2_mass))
+              interm3 = DBLE(zq(ig,l,igcm_ccnco2_number))
+              call updaterice_microco2(interm1,
+     &            interm2,interm3,
      &            tauscaling(ig),riceco2(ig,l),rhocloudco2(ig,l))