Running gnumake (gmake) in this directory will create an executable named

"interpinic"

PLATFORM:
This tool has been tested on ncar's ibms only.

USAGE:
interpinic -i <input initial data file>.nc  -o <output initial data file>.nc

DESCRIPTION:
Clm's initial files will initialize runs of the same resolution
and the same gridcell/landunit/column/pft configuration. Two runs may use
different resolutions or different gridcell/landunit/column/pft
configurations if the <surface-dataset>.nc files are different. Then
initial files from one run will fail to initialize the other. In that case
the tool interpinic may help.

The tool interpinic maps initial data from one clm initial file (the input
file) to another (the output file) by overwriting the contents of the output
file. The input and output files may be of any spatial resolution and
gridcell/landunit/column/pft configuration.

The input file typically originates from a long simulation with clm (spin up).
The output file is also created by running clm: Start your new
clm simulation with finidat='' in the namelist, ask the model to output
an initial file DAILY, and run for one day. The initial file thus created is
the file that interpinic calls the ouput file. Now run interpinic as shown
under USAGE (see above). When done, go back to your new clm run. Change
the namelist to say finidat='<output file name>.nc', don't ask the model to
output initial files DAILY, and run for as long as you need.

I recommend renaming the output file to something that describes
both what data is in the file and what run the file will be used for.


OUTLINE OF THE PROCESS:
Details may vary in newer versions of CAM and CLM...

Part 1. Make new clm initial file by making slight modifications to your run's namelist:
1) save a copy of your run's namelist
2) set  finidat=''  (i.e., two single quotes) in the clm namelist (bottom part of your namelist)
3) add the line  inithist='DAILY'  to the cam namelist (top part of your namelist)
4) run the model for 2 days and you will get one or two clm initial files from it; if you don't see them, they may be saved in the mass store...

Part 2: Map initial data from one clm initial file to another
1) copy my directory "/fis/cgd/tss/slevis/interpinic"
2) copy one of your newly created clm initial files to the /out directory
3) copy the spun up clm initial file to the /in directory
4) execute this command: "interpinic -i in/<file copied in step 3>.nc -o out/<file copied in step 2>.nc"

Part 3: Start your run
1) go back to your original namelist
2) set  finidat='<the new file after the mapping step>'
3) read DESCRIPTION above for more details...


5 June 2006 Notes:
-----------------

1) I updated interpinic.f90 when I found that the code
included in the clm3 release did not work for very high resolutions
(T170 -> T341) nor for the dgvm. This is in /interpinic3.0.

2) Jerry Olson suggested a correction in the calculation of dx:
--------------------------------------------------------------
Currently this calculation does not account for the jump from 360 to 0 degrees
at Greenwich. So Jerry suggested repeating the calculation of dx three times.
Effectively, once east-to-west, once west-to-east, and once to allow for a
jump from 180 to -180 degrees at the dateline. I think that he is making this
code change in his version.

Jerry may have made additional code changes in his version. Eg, he wanted
pft-level variables to get their data from the *same* pft rather than the
first available pft in the gridcell.

13 Dec 2006 Notes:
-----------------

I updated interpinic.f90 to accomodate new hydrology
variables in the clmi files. This code will be released with clm3.5 and is
in /interpinic3.5.

2 Mar 2007 Notes:
----------------

I modified the code described on 13 Dec 2006 to handle the special case
when the input file contains no dgvm data while the output file does.
For other cases continue to run older versions.
This version is in /myinterpinic

slevis
