Subject: missing fields in CCM3.6
From: Jerry Olson (olson@helgeland.cgd.ucar.edu)
Date: Fri Sep 04 1998 - 16:18:44 MDT
Ian P. Kraucunas <ian@atmos.washington.edu> writes:
> I am running CCM3.6 on a Sun workstation in dom/eul mode, and I would like
> to add the following fields to the history file: "HA01", the horizontal
> advection tendency of water vapor, and "VA01", the vertical advection
> tendency of water vapor. Both of these fields are ostensibly contained in
> the Master Field List. However, I cannot find the "outfld" calls for these
> particular variables anywhere in the source code to uncomment them (as
> discussed in the CCM User's Guide, Section 4.2). Thus "HA01" and "VA01"
> are included on my history file, but they are blank.
>
> Apparently there is not any code included with the model for the
> computation of "hadvnam(1)" or "vadvnam(1)" (these are the variable names
> associated with the fields "HA01" and "VA01", respectively). Does anyone
> have any code written to perform this calculation?
Code to perform these computations has indeed been removed from CCM3.6
since, in some future CCM version, SLT advection will not be split into 2-D
(horizontal) and 1-D (vertical) pieces. Rather, the advection algorithm
will be truly 3-D.
However, code does exist in CCM2 to perform the diagnostics you
want. Modifying CCM3.6 code using relevant CCM2 code should not be
too difficult. If you do not have CCM2 code available, you can get it
via ftp using the following recipe:
ftp ftp.ucar.edu
Name: anonymous
password: username@domain
cd ccm/oldccm
bin
get ccm2.tar
quit
"grep" for the variables "hqfcst" and "vqfcst" in the CCM2 code to
determine the routines that use them. In particular, "hqfcst" is
passed into SLTB1 (via SCANSLT) to store the horizontal tendency term.
The vertical tendency, "vqfcst" is then computed as a residual in
LINEMS and both are output to the history file from the routine
LINEMS (aka LINEMSBC in CCM3.6). Note also that "hqfcst" must be
written to the restart file in order for the model to restart
successfully.
-- Jerry Olson NCAR Boulder, Colorado
This archive was generated by hypermail 2b27 : Thu Jun 01 2000 - 09:06:49 MDT