Changeset 652 for trunk/LMDZ.MARS/libf/phymars/thermcell_main_mars.F90

Timestamp:

May 7, 2012, 4:01:14 PM (13 years ago)

Author:

acolaitis

Message:

Thermals:

• minor changes

Diffusion with Yamada4:

• corrected numerical oscillation in daytime first-level temperature and momentum predictions through the implementation of a sub-timestep (40 sub-steps) of one loop. (cheap in cpu time). This loop is the one that computes the tke at t+dt from the tke at t.
• corrected numerical oscillations in nighttime first-level temperature and momentum predictions through the use of a minimum mixing coefficient, computed as in Holtslag and Boville (93), tuned for Mars.
• yamada4 now takes into account the diffusion of tke (iflag_pbl=9), as it tends to smooth numerical oscillations. Results are very similar to those obtained without diffusion.

File:

• trunk/LMDZ.MARS/libf/phymars/thermcell_main_mars.F90 (modified) (1 diff)

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

@@ -857 +857 @@
             if (l.le.lmax(ig)) then
                 if (zw2(ig,l).lt.0.)then
-                  print*,'pb2 zw2<0'
+!                  print*,'pb2 zw2<0',zw2(ig,l)
+                  zw2(ig,l)=0.
                 endif
                 zw2(ig,l)=sqrt(zw2(ig,l))