user/flavoni/libIGCM/iom_put – NEMO

Context Navigation

Version 33 (modified by mafoipsl, 14 years ago) (diff)
--

Integrating IOM_PUT

1. Integrating IOM_PUT

Integrating IOM_PUT

Working tests:

test with iom_put (compiled with key_iomput):

work/cont003/flavoni/LIBIGCM_NEMO/LIKE-COUPLED/test_iom_lib14_rev1661_WR/modipsl/config/ORCA2_LIM/IGCM_lim14_io

KO: plantage au tout début : **** 90 Fatal exception PROG=mod_event_client.event__set_time ELN=164(4002dc900)

same test but without iom_put (in opa9.card io_xml=n):

work/cont003/flavoni/LIBIGCM_NEMO/LIKE-COUPLED/test_lib14_rev1661_WR/modipsl/config/ORCA2_LIM/IGCM_lim14

OK. Voir atlas et monitoring là : http://dods.extra.cea.fr/data/flavoni/ORCA2_LIM2/lib14if

today 30/10/2009 if you download ./model NEMO_IGCM you can't use iom_put.

If you want activate iom_put

If you want to activate iom_put you have to:

download model NEMO_IGCM:
```
cd modipsl/util 
./model NEMO_IGCM 
```

comment lines in dommsk:

       ij0 = 115   ;   ij1 = 115 
        ii0 = 145   ;   ii1 = 146   ;   fmask( mi0(ii0):mi1(ii1) , mj0(ij0):mj1(ij1) , 1:jpk ) = 4.0e0 
        ij0 = 116   ;   ij1 = 116 
        ii0 = 145   ;   ii1 = 146   ;   fmask( mi0(ii0):mi1(ii1) , mj0(ij0):mj1(ij1) , 1:jpk ) = 4.0e0

put in namelist
```
 rn_bfri1    =    0.e-4 
```

Add key in modipsl/config/ORCA2_LIM/scripts/BB_make.ldef
```
key_iomput
```

Do an ins_make :
```
./modipsl/util/ins_make
```

Go in ORCA2_LIM directory :
```
cd modipsl/config/ORCA2_LIM
```

Compile :
```
gmake
```

you can copy modified files (opa9.driver, opa9.card, lim2.card, POST directory) that you find in:

brodie: /workdir/rech/eee/reee605/LIBIGCM_NEMO/LIKE-COUPLED/test_claire/  
mercure: /work/cont003/flavoni/LIBIGCM_NEMO/LIKE-COUPLED/test_iom_lib14_rev1661_WR/modipsl/config/ORCA2_LIM/IGCM_lim14_io/COMP

Change in opa9.card:
```
io_xml='''y'''  
```

My job: modifications done till 30/10/2009 to prepare files for using iom_put:

Modifications done till now were necessary if not it was not possible to have good file for iodef.xml

Changed revision for nemo in modispl/util/mod.def in NEMO_IGCM:

#-H- NEMO_IGCM NEMO trunk revision 1661
(revision without bottom friction bug)

in lim2.card

changed PREFIX name for output .nc file
[OutputFiles?]
List= (${PREFIX_NEMO}_icemod.nc, ${R_OUT_ICE_NWRITE}/${PREFIX}_${WF1}_icemod.nc, Post_1M_icemod)

removed deptht (for monitoring, no good to do ncrcat in 3 dimensions for ice)
[Post_1M_icemod]
GatherWithInternal? = (nav_lon, nav_lat, time_counter)

removed ileadfra for the moment
TimeSeriesVars? = (iicethic, isnowthi, iicevelu, iicevelv)

NOTE: for this moment ileadfra removed because in iomdef.xml there are new variables' names (like IPPC requested) and ileadfra there is no more in output from nemo's code, waiting for that nemo put in output 1-soicecov (=ileadfra)

in opa9.card

changed PREFIX name for output .nc file
[OutputFiles?]
List= (${PREFIX_NEMO}'_grid_T.nc, ${R_OUT_OCE_NWRITE}/${PREFIX}_${WF1}_grid_T.nc, Post_1M_grid_T),\
(${PREFIX_NEMO}_grid_U.nc, ${R_OUT_OCE_NWRITE}/${PREFIX}_${WF1}_grid_U.nc, Post_1M_grid_U),\ [BR]] (${PREFIX_NEMO}_grid_V.nc, ${R_OUT_OCE_NWRITE}/${PREFIX}_${WF1}_grid_V.nc, Post_1M_grid_V),\
(${PREFIX_NEMO}_grid_W.nc, ${R_OUT_OCE_NWRITE}/${PREFIX}_${WF1}_grid_W.nc, Post_1M_grid_W),\

(in opa9.driver ${PREFIX_NEMO} is:
${PREFIX_NEMO}=${PREFIX}_{WF1} if we use iomput
${PREFIX_NEMO}=${PREFIX_NWRITE}_${DATE_OPA} if not )

in opa9.driver

possibility to use io_server (but not testing now)
##--Variables used for XMLIO_SERVER configuration file
# using_server: .TRUE.(.FALSE.) to use(bypass) the io_server
PAT_SERVER=$( supergrep using_server ${SUBMIT_DIR}/PARAM/xmlio_server.def )

Added output_frequency in days, if not xml file is not correct
(( output_frequency_oce = $WriteFrequencyInDays? * 86400 ))
(( output_frequency_ice = $WriteFrequencyInDays? * 86400 ))

replaced output filenames in xml file
sed -e "s/<file_name_grid_T>/ \"${PREFIX_NEMO}_grid_T.nc\"/" \
-e "s/<file_name_grid_U>/ \"${PREFIX_NEMO}_grid_U.nc\"/" \
-e "s/<file_name_grid_V>/ \"${PREFIX_NEMO}_grid_V.nc\"/" \
-e "s/<file_name_grid_W>/ \"${PREFIX_NEMO}_grid_W.nc\"/" \
-e "s/<file_name_icemod>/ \"${PREFIX_NEMO}_icemod.nc\"/" \
-e "s/<output_frequency_oce>/ \"${output_frequency_oce}\"/g" \
-e "s/<output_frequency_ice>/ \"${output_frequency_ice}\"/g" \
iodef.xml > iodef.xml.tmp

IGCM_sys_Mv iodef.xml.tmp iodef.xml

determine number of day(s) in WriteFrequency? :
case ${config_UserChoices_OCE_WriteFrquency} in

1Y|1y)

(( WriteFrequencyInDays? = $( IGCM_date_DaysInYear $year ) )) ;;

1M|1m)

(( WriteFrequencyInDays? = $( IGCM_date_DaysInMonth $year $month ) )) ;;

5D|5d)

(( WriteFrequencyInDays?=5 )) ;;

1D|1d)

(( WriteFrequencyInDays?=1 )) ;;

*)

IGCM_debug_Exit "IGCM_config_OCE_WriteFrquency " ${config_UserChoices_OCE_WriteFrquency} " invalid write frquency length : choose in 1Y, 1M, 5D, 1D." IGCM_debug_Verif_Exit ;;

esac

Created POST directory under $SUBMIT_DIR:
In POST directrory there are:
monitoring01_lim2_ORCA2.cfg
monitoring01_opa9_ORCA2.cfg

In monitoring01_lim2_ORCA2.cfg I removed ileadfra because now this variable does not exists, and if I don't use "dynamic monitoring" I can't have ICE monitoring (monitoring don't create monitoring images if there is one or more variables that does not exist).

see https://forge.ipsl.jussieu.fr/nemo/wiki/private/libIGCM/postprocessingpage

At this point: open problems:

Don't have ATLAS for "ice cover": because new variable is named soicecov and in atlas is written ileadfra
to do: calcul for 2 frequencies (waiting for a new iodef.xml)

Tests on vargas

Directory :

/workgpfs/rech/psl/rpsl003/VARGAS/NEMO/CLIM_IOM/modipsl/config/ORCA2_LIM

Main idea :
- use xml.def to list fields and file for differents types of output (1M, 1D, 1Y, ...)
  - Usefull for grid_[TUVW] and icemod files
  - old fashion for diaptr
- use -1 for 1Y length simulation to have monthly (variable length) outputs
  - OK for xml files
  - nead adjustment for old fashion (diaptr) files
Modifications :
- COMP directory :
  - opa9.driver :
    - use Vxx_enable variable to set information in xml.def file (1M enable for example)
  - opa9.card :
    - add a list of possible files : ORCA2_1m_grid_T.nc ...
    - add a Patch for time counter (to rename the first time axis time_counter, t_ave_01month allowable, works when no tule_counter rename is required)
      - See : /workgpfs/rech/psl/rpsl003/VARGAS/NEMO/CLIM_IOM/modipsl/libIGCM/libIGCM_post/IGCM_Patch_20091030_histcom_time_axis.ksh
    - delete sophtldf variable in monitoring list
  - lim2.card :
    - add a list of possible files : ORCA2_1m_icemod.nc ...
    - add a Patch for time counter (to rename the first time axis time_counter, t_ave_01month allowable, works when no tule_counter rename is required)
    - delete deptht variable in gather list
- PARAM directory :
  - add iodef.xml
  - xmlio_server.def
- POST directory :
  - add monitoring01_lim2.cfg file :
    - a copy (take care : without _ORCA2) of /home/rech/psl/rpsl035/atlas/monitoring01_lim2_ORCA2.cfg
    - without ileadfra_south and ileadfra_north lines

4 tests :

All tests have been done with 1Y atlas/monitoring :

Modifications in config.card for all tests :

JobNumProcTot=16
RebuildFrequency=1Y
DateEnd=2002-12-31
TimeSeriesFrequency=1Y
SeasonalFrequency=1Y
WriteFrequency="1M 1D"
WriteFrequency="1M 1D"

PeriodLength? 1M 1Y
360d NLIO5 / IGCM360MO NLIO7 / IGCM360
leap NLIO20 /IGCM365MO6 NLIO15 / IGCM3656

NLIO5

Status : OK
Modifications in IGCM360MO/config.card
```
JobName=NLIO5
PeriodLength=1M
```
Issues already known :
- No mask in atlas for continents
- ileadfra maps empty
See : http://dodsp.idris.fr/rpsl003/ORCA2_LIM2/NLIO5

NLIO7

Status : OK
Modifications in IGCM360/config.card
```
JobName=NLIO7
```
Issues already known :
- No mask in atlas for continents
- ileadfra maps empty
See : http://dodsp.idris.fr/rpsl003/ORCA2_LIM2/NLIO7

NLIO15

Status : KO
Modifications in IGCM3656/config.card
```
JobName=NLIO15
CalendarType=leap
```
Issues :
- diaptr not saved name : NLIO15_CU*_20010101_20011231_diaptr_0000.nc
  - ZNL atlas empty
  - Monitoring : only ICE variables, only 1 year,
  - will work better when diaptr files will exist
Issues already known :
- No mask in atlas for continents
- ileadfra maps empty
See : http://dodsp.idris.fr/rpsl003/ORCA2_LIM2/NLIO15

NLIO20

Status :

Modifications in IGCM365MO6/config.card

JobName=NLIO20
CalendarType=leap
PeriodLength=1M

Issues already known :
- No mask in atlas for continents
- ileadfra maps empty
See : http://dodsp.idris.fr/rpsl003/ORCA2_LIM2/NLIO20

Open questions

waiting for same test on brodie in mono (Claire?)

Proposition

Add a directory IGCM01 with this COMP, PARAM, POST files here : https://forge.ipsl.jussieu.fr/nemo/browser/branches/libIGCM/ORCA2_LIM

Download in other formats:

Plain Text

PeriodLength?	1M	1Y
360d	NLIO5 / IGCM360MO	NLIO7 / IGCM360
leap	NLIO20 /IGCM365MO6	NLIO15 / IGCM3656

New URL for NEMO forge! http://forge.nemo-ocean.eu

Context Navigation

Integrating IOM_PUT

Integrating IOM_PUT

If you want activate iom_put

My job: modifications done till 30/10/2009 to prepare files for using iom_put:

Tests on vargas

NLIO5

NLIO7

NLIO15

NLIO20

Open questions

Proposition

Download in other formats: