Custom Query (964 matches)

Dear ROMS-Team, Dear Hernan Arango,

IMHO the subroutine ./ROMS/Utility/checkvars.F needs a patch to ensure the reading of passive tracers after a restart.

>> gdiff -uN checkvars.F checkvars.F.NEW 

--- checkvars.F 2008-01-28 15:52:49.307101000 +0100
+++ checkvars.F.NEW     2008-01-28 15:52:49.296606000 +0100
@@ -117,11 +117,19 @@
 # endif
 #endif
 #ifdef SOLVE3D
-# if defined T_PASSIVE && !defined ANA_PASSIVE
+# if defined T_PASSIVE
+#  if defined ANA_PASSIVE
+      IF (nrrec(ng).ne.0) THEN
+        DO itrc=1,NPT
+          get_var(idTvar(inert(itrc)))=.TRUE.
+        END DO
+      END IF
+#  else
       DO itrc=1,NPT
         get_var(idTvar(inert(itrc)))=.TRUE.
       END DO
-# endif
+#  endif ANA_PASSIVE
+# endif T_PASSIVE
 # ifdef SEDIMENT
 #  ifdef SED_MORPH
       IF (nrrec(ng).ne.0) THEN

Thanks and have a nice day,

Christian

Corrected the comments about the unit for shortwave radiation in ana_srflux.h. The fluxes inside ROMS kernel are kinematic so the unit for shortwave radiation is degC m/s. Notice that if the shortwave radiation is read from input NetCDF file it is converted to degC m/s using the Fscale in varinfo.dat. Usually, shortwave radiation is in Watt/m2 so:

       Fscale=1.0_r8/(rho0*Cp).

Many thanks to Laura Bianucci for reporting this typo.

I added the capability to save mass fluxes (DU_avg1, DU_avg2, DV_avg1, DV_avg2) for coupling 3D momentum into the time-averaged NetCDF file. In the code these fields are averaged over all the barotropic time-steps. Here, they are also averaged over the specified baroclinic steps. This is only done for output purpuses (AVERAGES and FORWARD_WRITE) so the average NetCDF can be used as the basic state to linearize the tangent linear and adjoint models. This is only relevant in adaptive sampling and optimal observations.