New land_ocean_seaice experiment.

[MJHarrison-GFDL]

• new example directory for LM3
• land_ice_ocean_LM3_SIS2/OM_360x320_C180
• requires update to atmos_null (dev/master on FMS github)
• INPUT files need to be moved to a centralized location.
• Just needs first 10days+1hr from the atmos data.

[adcroft]

I'm working on pulling this but once I'm done we will need to retroactively install a preprocessing directory.

Some questions:

1. What are the linked runscript and XML for?
2. README says "See ./gather_lm3_data.sh for details." which isn't present.
3. I see you are renaming files in the links, e.g.
   q10.nc -> /lustre/f1/pdata/gfdl_O/datasets/reanalysis/MERRA/q10/q10_merra_1992.nc
   1. Doesn't the XML have trouble with this?
   2. How do you plan to use a year other than 1992? Shouldn't we link all the MERRA files here (they're only links) and put the full filename (which includes year) in data_table?
4. Directory is called OM_360x320_C180, diag_table has OMz_360x320_C180. Which do you want? Or OM4_360x320_C180?
5. field_table has "atmos_mod" entries?

[adcroft]

Just noticed there's no ice_model.res.nc in INPUT/ but there is in /lustre/f1/unswept/Matthew.Harrison/MOM6_datasets/C180_OM360x320.071315. Should it be linked?

[adcroft]

Matt, dev/master doesn't exist for atmos_null? Send a commit id/hash...

[MJHarrison-GFDL]

git@github.com/NOAA-GFDL/atmos_null/commit/a7b75f3bcbe1ef4efe625edc71843466117ef7c3

On Thu, Nov 5, 2015 at 9:04 PM, Alistair Adcroft notifications@github.com
wrote:

Matt, dev/master doesn't exist for atmos_null? Send a commit id/hash...

—
Reply to this email directly or view it on GitHub
#60 (comment)
.

[MJHarrison-GFDL]

Removed. Cold starting the ice.

On Thu, Nov 5, 2015 at 8:36 PM, Alistair Adcroft notifications@github.com
wrote:

Just noticed there's no ice_model.res.nc in INPUT/ but there is in
/lustre/f1/unswept/Matthew.Harrison/MOM6_datasets/C180_OM360x320.071315.
Should it be linked?

—
Reply to this email directly or view it on GitHub
#60 (comment)
.

[MJHarrison-GFDL]

1,2 : Need to consult with @SergeyMalyshev and @elenashevliakova . Need to
commit LM3 spin-up.

1. Resetting the link every 1 year segment in production runs. I have XML
   for this. I obviously need to add a preprocessing directory!

Part of it is here

https://github.com/NOAA-GFDL/MOM6-examples/tree/user/mjh/FLOR2/tools/python/FLOR2gridGeneration

Needs work to add MERRA download scripts.

1. Done
2. Need to allocate for spec. humidity . Not sure about cld amt.

On Thu, Nov 5, 2015 at 8:32 PM, Alistair Adcroft notifications@github.com
wrote:

I'm working on pulling this but once I'm done we will need to
retroactively install a preprocessing directory.

Some questions:

1. What are the linked runscript and XML for?
2. README says "See ./gather_lm3_data.sh for details." which isn't
   present.
3. I see you are renaming files in the links, e.g. q10.nc ->
   /lustre/f1/pdata/gfdl_O/datasets/reanalysis/MERRA/q10/q10_merra_1992.nc
   1. Doesn't the XML have trouble with this?
   2. How do you plan to use a year other than 1992? Shouldn't we link
      all the MERRA files here (they're only links) and put the full filename
      (which includes year) in data_table?
4. Directory is called OM_360x320_C180, diag_table has OMz_360x320_C180.
   Which do you want? Or OM4_360x320_C180?
5. field_table has "atmos_mod" entries?

—
Reply to this email directly or view it on GitHub
#60 (comment)
.