Changeset 659

Timestamp:

May 15, 2012, 2:17:33 PM (13 years ago)

Author:

acolaitis

Message:

Thermals: minor changes to some parameters, according to latest high domain size simulations. This really does not change the result when using yamada4 for diffusion anyway, but is consistent with the most precise LES observations. On the conceptual level, one of the important change is that the mass-flux compensation for updrafts and downdrafts is now a subsidence and no more an uplift, as in terrestrial modeling.

File:

• trunk/LMDZ.MARS/libf/phymars/thermcell_main_mars.F90 (modified) (5 diffs)

trunk/LMDZ.MARS/libf/phymars/thermcell_main_mars.F90

@@ -381 +381 @@
       a1=1.4716 ; b1=0.0005698 ; ae=0.03683 ; be = 0.57421
       ad = 0.00048088  ; bd = -0.6697
-      fdfu = -1.3
+!      fdfu = -1.3
+      fdfu=-0.8
       a1inv=a1
       b1inv=b1
@@ -404 +405 @@
 
 ! Best config for norad 222 levels:
-
-       omega=0.06
+! and more specifically to C.large :
+
+!       omega=0.06
+!       omega=0.
+       omega=-0.03
 !       omega=0.
        a1=1.
@@ -413 +417 @@
        be=1.1*be
        ad = 0.0004
-       adalim=0.
-
-       b1inv=0.00025
-!       b1inv=b1
+!       ad=0.0003
+!       adalim=0.
+
+!       b1inv=0.00025
+       b1inv=b1
 
 !       b1inv = 0.0003
@@ -478 +483 @@
 !        do l=2,4
          do ig=1,ngridmx
-           if (ztv(ig,l)>(ztv(ig,l+1)+0.) .and. ztv(ig,1)>=ztv(ig,l) .and. (alim_star(ig,l-1) .ne. 0.)) then ! .and. (zlev(ig,l+1) .lt. 1000.)) then
+           if (ztv(ig,l)>(ztv(ig,l+1 + 0.)) .and. ztv(ig,1)>=ztv(ig,l) .and. (alim_star(ig,l-1).ne. 0.)) then ! .and. (zlev(ig,l+1) .lt. 1000.)) then
                alim_star(ig,l)=MAX((ztv(ig,l)-ztv(ig,l+1)),0.)  &
 !     &                       *log(1.+zlev(ig,l+1))
@@ -1279 +1284 @@
               if(zlay(ig,l) .le. zmax(ig)) then
               fm_down(ig,l) =fm(ig,l)* &
-     &      max(fdfu,-3*max(0.,(zlay(ig,l)/zmax(ig)))-0.9)
+     &      max(fdfu,-4*max(0.,(zlay(ig,l)/zmax(ig)))-0.6)
               endif