Baroclinic wave experiment by Polvani et al. (2004)

A method to perform a baroclinic wave experiment by Polvani et al. (2004) is described.

Overview

This experiment does not require additional data, such as SST data. Thus, the experiment is performed with the following 3 steps:

Preparation of a directory for experiments

Let us move to the top directory of DCPAM5 src tree (assuming dcpam5-YYYYMMDD in the following example). Here, we perform an experiment in ../dcpam5-exp/p04-exp directory. Please create the directory and enter there as follows:

$ mkdir -p ../dcpam5-exp/p04-exp
$ cd ../dcpam5-exp/p04-exp

Then, pleaase create the directories for executable files and configuration files as follows:

$ mkdir bin
$ mkdir conf

Finally, executable files and configuration files are copied as follows:

$ cp ../../dcpam5-YYYYMMDD/src/main/dcpam_main bin
$ cp ../../dcpam5-YYYYMMDD/src/main/dcpam_init_data  bin
$ cp ../../dcpam5-YYYYMMDD/exp_setup_files/*.conf conf

Note that you can perform an experiment in any directory by using executable files and configuration (NAMELIST) files.

Create initial data file

Create initial data file "init_T21L20.nc" using "bin/dcpam_init_data" and "conf/init_data_p04_T21L20.conf"

$ ./bin/dcpam_init_data -N=./conf/init_data_p04_T21L20.conf

 *** MESSAGE [init_data] ***  Run: Initial data generation
 *** MESSAGE [init_data] ***  -- version = $Name:  $$Id: exp-p04.rd,v 1.14 2014/07/07 14:50:00 yot Exp $
 *** MESSAGE [namelist_util] ***  ----- Initialization Messages -----
 *** MESSAGE [namelist_util] ***    MaxNmlArySize = 256
 *** MESSAGE [namelist_util] ***  -- version = $Name:  $$Id: exp-p04.rd,v 1.14 2014/07/07 14:50:00 yot Exp $
 *** MESSAGE [timeset] ***  NAMELIST group "timeset_nml" is loaded from "./conf/init_data_p04_T21L20.conf".
                       : 
 *** MESSAGE [initial_data] ***    QVapAvr = 0.
 *** MESSAGE [initial_data] ***    Ueq     = 0.
 *** MESSAGE [initial_data] ***
 *** MESSAGE [initial_data] ***  -- version = $Name:  $$Id: exp-p04.rd,v 1.14 2014/07/07 14:50:00 yot Exp $
 *** MESSAGE [HistoryClose] ***  "init_T21L20.nc" is closed

 ############## CPU TIME SUMMARY ################
 others                 0.200000E-01
 ------------------------------------------------
        TOTAL TIME =    0.200000E-01

Run the experiment

Using an executable files 'dcpam_main' and a NAMELIST file dcpam_p04_T21L20.conf, execute 'dcpam_main' as follows. This program will be finished in few minutes - tens of minutes.

$ ./bin/dcpam_main -N=./conf/dcpam_p04_T21L20.conf | tee p04.log

 *** MESSAGE [namelist_util] ***  ----- Initialization Messages -----
 *** MESSAGE [namelist_util] ***    MaxNmlArySize = 256
 *** MESSAGE [namelist_util] ***  -- version = $Name:  $$Id: exp-p04.rd,v 1.14 2014/07/07 14:50:00 yot Exp $
 *** MESSAGE [dcpam_main] ***  NAMELIST group "dcpam_main_nml" is loaded from "./conf/dcpam_p04_T21L20.conf".
 &DCPAM_MAIN_NML FLAGDYNAMICS=T,FLAGFULLPHYSICS=F,FLAGHS94=F,FLAGVENUSSIMPLE=F,FLAGJUPITERSIMPLE=F,FLAGRADIATIONDENNOUAGCM=T,FLAGRADIATIONDCPAMEV1=F,FLAGRADIATIONDCPAMMV1=F,FLAGRADIATIONSL09=F,FLAGPHYIMP1LAYMODEL=T,FLAGPHYIMPSOILMODEL=F,FLAGPHYIMPSOILMODELSO=F,FLAGPHYIMPATMONLY=F,FLAGVERTICALFILTER=F/
 *** MESSAGE [dcpam_main] ***  FlagDynamics=<T>.
 *** MESSAGE [dcpam_main] ***  FlagRadiationDennouAGCM=<T>.
                   :
 *** MESSAGE [HistoryClose] ***  "TotEngy.nc" is closed
 *** MESSAGE [HistoryClose] ***  "Enstro.nc" is closed

 ############## CPU TIME SUMMARY ################
 surface_properties     0.904434E-07
 dynamics_hspl_vas83    0.380824E+01
 mass_fixer             0.132011E+00
 check_prog_vars        0.360025E-01
 timefilter_asselin19   0.760067E-01
 others                 0.212009E+00
 ------------------------------------------------
        TOTAL TIME =    0.426427E+01

In this case, 12 days integration is performed. Resolution is T21L20 (Number of grid points of longitude, latitude is 64, 32, and number of vertical layer is 20). History data are output to 'U.nc' and 'Temp.nc' etc., and a restart data is output to 'restart.nc'.

Final remark

If you would like to keep the results of the numerical experiment, please create a sub directory ("data01" for example), move the files, and keep the current directory neatly.

$ mkdir data01
$ mv *.nc *.log data01/

You can output the results of the numerical experimet concurrently with the execution. Please refer to "Change output settings" in Change experimental setup and try to configure "FilePrefix" variable in NAMELIST "&gtool_historyauto_nml".

References


$Id: exp-p04.rd,v 1.14 2014/07/07 14:50:00 yot Exp $