[Dart-dev] DART/branches Revision: 11584

[dart at ucar.edu]

thoar at ucar.edu
2017-05-02 14:42:43 -0600 (Tue, 02 May 2017)
128
Moving the unfinished scripts to a separate directory.
The model_mod.html now tells people where to get the 
example datasets.




Modified: DART/branches/rma_trunk/models/cm1/model_mod.html
===================================================================
--- DART/branches/rma_trunk/models/cm1/model_mod.html	2017-05-02 19:58:25 UTC (rev 11583)
+++ DART/branches/rma_trunk/models/cm1/model_mod.html	2017-05-02 20:42:43 UTC (rev 11584)
@@ -88,8 +88,10 @@
 <P>There are two sets of scripts in the <em class=file>shell_scripts</em> directory.
 Luke contributed a set written in python, and the DART team had a set written in csh.
 The csh scripts have not been tested in quite some time, so use with the understanding
-that they will need work.
-<P>
+that they will need work. Those csh scripts and some unfinished python scripts reside
+in a <em class=file>shell_scripts/unfinished</em> directory and should be used with the
+understanding that they will need work.
+</P>
 <H3>Strategy and Instructions for using the python scripts.</H3>
 <H4>There are some prerequisites:</H4>
 <ol><li>CM1 is required to use netCDF restart files.</li>
@@ -116,6 +118,14 @@
 run the experiment, the working directory. {centraldir} should have a lot of capacity, 
 as ensemble data assimilation will require lots of disk.
 The term {dart_dir} will refer to the location of the DART source code.
+<br />
+<br />
+The data referenced in the directories (the initial ensemble, etc.) are provided
+as a compressed tar file 
+<a href="http://www.image.ucar.edu/pub/DART/CM1/cm1r18_3member_example_data.tar.gz">cm1r18_3member_example_data.tar.gz</a>
+You will have to download the tar file, uncompress it, and modify the scripts to 
+use these directories instead of the example directories in the scripts.
+You will also have to compile your own cm1 executable.
 </P>
 
 <ol>

Deleted: DART/branches/rma_trunk/models/cm1/shell_scripts/advance_model.csh
===================================================================
--- DART/branches/rma_trunk/models/cm1/shell_scripts/advance_model.csh	2017-05-02 19:58:25 UTC (rev 11583)
+++ DART/branches/rma_trunk/models/cm1/shell_scripts/advance_model.csh	2017-05-02 20:42:43 UTC (rev 11584)
@@ -1,215 +0,0 @@
-#!/bin/csh
-#
-# DART software - Copyright UCAR. This open source software is provided
-# by UCAR, "as is", without charge, subject to all terms of use at
-# http://www.image.ucar.edu/DAReS/DART/DART_download
-#
-# DART $Id$
-
-#-----------------------------------------------------------------------------
-#
-# Script for use in assimilation applications
-# where the model advance is executed as a separate process.
-#
-# Arguments are (created by 'filter' or 'perfect_model_obs' and include):
-# 1) the process number of caller,
-# 2) the number of ensemble members/state copies belonging to that process, and
-# 3) the name of the control_file for that process.
-#
-# If this script finishes and the 'control_file' still exists, it is
-# an ERROR CONDITION and means one or more of the ensemble members did
-# not advance properly. Despite our best attempts to trap on this
-# condition, some MPI installations simply hang, some properly terminate.
-#
-# This script loops over all the entries in the control_file to advance
-# any/all of the ensemble members.  The number of trips through the
-# loop is the second argument to this script. The control_file contains
-# the information about which ensemble members are to be advanced by THIS TASK.
-# Sometimes it may be just one ensemble member, sometimes all of them.
-# Read DART/doc/html/filter_async_modes.html and the mpi_intro.html
-# for an overview.
-#
-# This script has 4 logical 'blocks':
-# 1) determine how many ensemble members (num_states) this process 
-#    will need to advance.
-# 2) convey the forecast length to the model
-# 3) run the model (make the forecast)
-# 4) determine if there are more ensemble members to advance
-
-set      process = $1
-set   num_states = $2
-set control_file = $3
-
-echo "advance_model: process $process"
-echo "advance_model: number of states to advance $num_states"
-echo "advance_model: control file name is $control_file"
-
-set CENTRALDIR = `pwd`
-
-# Loop through each model state
-
-set ensemble_member_line = 1
-set      input_file_line = 2
-set     output_file_line = 3
-set    current_time_line = 4
-set advance_to_time_line = 5
-
-set  state_copy = 1
-
-while($state_copy <= $num_states)
-
-   #-------------------------------------------------------------------
-   # Block 1: Parse the information from the control file
-   #