Icepack #750
Icepack #750
Conversation
cice single colum model work from: Molly Wieringa Chris Riedel Cecilia Bitz This reverts commit 0932d47.
Sea Ice Single Column Model
…register_module, adjusting style and organization to be more readable and consistent
…t that were replaced with the corresponding ones from default_model_mod
…_module, adjusting style to be more readable and consistent
…ubroutines and unused variables
…ables; use len=vtablenamelength in definition for state variable_table instead of len=NF90_MAX_NAME
…default ; removed assimilation_period_days, assimilation_period_seconds from &model_nml Set default namelist item ‘debug’ value to 1 to not print all the additional info
…tions to use the icepack model_mod. Added a example test case that uses perturb_single_instance=.true. to the repo
…etcdf variable names for the corresponding QTYs. Note that this commit is more for bookeeping since this subroutine will be removed from the code in the following commit. Other model_mods do not include this subroutine and the default is typically hardcoded in the model_state_variables item of the &model_nml
…comment. The rest of the model_mods do not use a subroutine like this. The default is provided in the input namelist. Added check in place of the subroutine calll to error out if the model_stat_variables nml entry is completely empty (model_stat_variables = ‘ ’)
|
Hi @mjs2369 not looked at this, but don't commit binary files and test data. Can you remove from the repo. |
yep, on it Also meant to change the dir name but slipped my mind. I'll do that as well |
… dir from the repo and include the model_nml item grid_oi. Updating source variable in model_mod
… from model_interpolate
| @@ -457,6 +457,7 @@ References | |||
| models/gitm/readme | |||
| models/gitm/netcdf_to_gitm_blocks | |||
| models/gitm/gitm_blocks_to_netcdf | |||
| models/icepack/readme | |||
There was a problem hiding this comment.
also need to add icepack to
https://github.com/NCAR/DART/blob/icepack/models/README.rst
|
|
||
| module model_mod | ||
|
|
||
| ! Modules that are absolutely required for use are listed |
There was a problem hiding this comment.
| ! Modules that are absolutely required for use are listed |
| use time_manager_mod, only : time_type, set_calendar_type, get_time, set_date, get_date | ||
| use location_mod, only : location_type, get_close_type, get_close_obs, get_dist, & | ||
| convert_vertical_obs, convert_vertical_state, & | ||
| set_location, set_location_missing, VERTISLEVEL, & |
There was a problem hiding this comment.
set_location_missing isn't used, register_module isn't used. Have a check to see what is actually used in this module.
| nc_write_model_atts, & | ||
| init_time, & | ||
| init_conditions, & | ||
| check_sfctemp_var |
There was a problem hiding this comment.
'check_sfctemp_var' is not a 'required by DART' routine. Is check_sfctemp_var used anywhere?
| read(iunit, nml = model_nml, iostat = io) | ||
| call check_namelist_read(iunit, io, "model_nml") | ||
|
|
||
| call error_handler(E_MSG,'static_init_model','model_nml values are',' ',' ',' ') |
There was a problem hiding this comment.
what's the goal of this message? It is going to print:
'static_init_model','model_nml values are',' ',' ',' '
| model_state_variables = 'aicen', 'QTY_SEAICE_CONCENTR', 'UPDATE', 'vicen', | ||
| 'QTY_SEAICE_VOLUME', 'UPDATE', 'vsnon', 'QTY_SEAICE_SNOWVOLUME', | ||
| 'UPDATE' |
There was a problem hiding this comment.
ditto: more readable as a table
| | ``analysis_mean.nc`` and ``analysis_sd`` - the mean and standard deviation of the state of all ensemble members after the assimilation | ||
| | ``obs_seq.final`` - the ensemble members' estimate of the observations. | ||
| | ``dart_log.out`` - detailed log file for the execution of filter | ||
|
|
There was a problem hiding this comment.
There is no mention of cice_to_dart in the documentation.
| endif | ||
| enddo | ||
|
|
||
| call nc_check( nf90_open(trim(original_cice_input_file), NF90_WRITE, ncid), & |
There was a problem hiding this comment.
I think you could use netcdf_utilites_mod for a lot of these netcf calls.
| 'get_var '//trim(msgstring)) | ||
|
|
||
|
|
||
| var(:) = holder(gridpt_oi,:) |
There was a problem hiding this comment.
is gridpt_oi a time slice?
| phi ! ice liquid fraction | ||
|
|
||
| ! from shr_const_mod.F90 | ||
| real(r8),parameter :: SHR_CONST_CPSW = 3.996e3_R8 ! specific heat of sea water ~ J/kg/K |
There was a problem hiding this comment.
there's a mixture of r8 and R8 in this section
Description:
Adds the DART interface to Icepack, the CICE Single Column Model
This new model code is due to the work done by @mollymwieringa and @criedel40
This PR is on draft because there are a few things that still need worked out. Namely, updating the model_interpolate lon_lat_interpolate to use state_structure_mod routines instead and editing the commit history to give Molly and Chris authorship on the initial commit
This PR will not include any scripting to assist our users with running the assimilations, these will be added in a separate PR. The same goes for new observation converters for sea ice observations
Here is the original specification
Fixes issue
#475
Types of changes
Documentation changes needed?
Tests
Compiled and ran the Icepack code with full debugging flags with ifort and gfortran
Tested to be bitwise identical with original icepack code on the branch cice-scm
Checklist for merging
Checklist for release
Testing Datasets
Test cases can be found here:
/glade/derecho/scratch/masmith/DART_latest/DART/models/icepack/tests
Information on them can be found in the README file in their respective directories
Note that for these tests, the assimilation is only performed over a single time window and therefore the model is never advanced.