cases script for GSM


runs/expscr/cases.in is the template for cases script created by 'configure-scr cases'

Initial year, month, day, hour of the first ensemble member

YYYYSTART=1999
YYYYEND=1999
LENMON=1
MMSS=07
DDSS=01
HHSS=00

Number of members in an ensemble

NENS=1

Ensemble initial condition time interval

INCENS=12

Initial conditions and sst sea_ice and snow

Atmospheric Initial condition [ IF INITIAL FILES ARE ON LOCAL DISK]

INITIAL_SIG_PLATFORM=[LEAVE THIS BLANK]
INITIAL_SIG_PROTOCOL=ln
INITIAL_SIG=$DIRECTORY_FOR_INITIAL_FILES/\$YYYYS\$MMS/ncep/anl/sig.anl.\$YYYYS\$MMS\$DDS\$HHS.ieee
INITIAL_SFC=$DIRECTORY_FOR_INITIAL_FILES/\$YYYYS\$MMS/ncep/anl/sfc.anl.\$YYYYS\$MMS\$DDS\$HHS.ieee
INPUT_RESOLUTION=t62k28
INITIAL_SFC_FILE_TYPE=osu1

sst [IF SST FILES ARE ON LOCAL DISK; EACH SST FILE CONTAINS DAILY SST VALUES FOR THE YEAR]

SST_FTP=ln
SST_PLATFORM=[LEAVE THIS BLANK]
SST_ANL_T0=$DIRECTORY_FOR_SST_FILES/\$YYYY/sst.grib
SST_ANL=$DIRECTORY_FOR_SST_FILES/\$YYYY/sst.grib

if you want to use observed or persistent sea ice. ice is climatology in this example

ICE_PLATFORM=[LEAVE THIS BLANK]
ICE_FTP=ln
ICE_ANL_T0=
ICE_ANL=@CON_DIR@/clim.ice.grib

snow

SNO_PLATFORM=
SNO_FTP=
SNO_ANL_T0=
SNO_ANL=
SCV_ANL_T0=
SCV_ANL=

Intervals of sfc execution, sigma output, sfc output and flx output

INTSFCX=24
INTSIG=24
INTSFC=24
INTPOST=24
INTRESTART=24
INCHOUR=384

Discussions

khideki

running the model from mid month

khideki 19 December 2006 17:29:46

A question from sreejith:

Now the model crash at fcst stage.The model stop like this,
___

                              • nnday of year = 121**
0from heatl3 jdnmc etc 2444725 0.50 0.00 121.00 121.00
0 forecast date 1 may 1981 at 0 hrs 0.00 mins
julian day 2444725 plus 0.500000
radius vector 1.0076085
right ascension of sun 2.5418509 hrs, or 2 hrs 32 mins 30.7 secs
declination of the sun 14.9874304 degs, or 14 degs 59 mins 14.7 secs
equation of time 2.8543465 mins, or 171.26 secs, or 0.012489 radians
solar constant 1.9664131
1 ozone climatology for month,day= 5 1
aerosol, o2, co2, h2o, o3 = 1 0 0 1 1
global checked speed maxima for all layers
spdmx(01:10)= 27. 28. 31. 33. 35. 36. 34. 34. 35. 38.
spdmx(11:20)= 42. 45. 50. 54. 61. 68. 74. 80. 74. 60.
spdmx(21:30)= 46. 41. 44. 49. 55. 61. 73. 98.
forward step: kdt in gsmstep= 2
global checked speed maxima for all layers
spdmx(01:10)= 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
spdmx(11:20)= 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
spdmx(21:30)= 0. 0. 0. 0. 0. 0. 0. 0.
run failure. spdmax=0.
Aborted
___

I gave the inputs sfc and sig file for may 15th 1981 and daily sst data file for year 1981 ( when I create ctl file for ploting sst value using grib2ctl.pl utility, that showing varible name "pvmwsfc" surface Potential vorticity, but values looks like sst in Kelvin).
Another problem I notice is while using case script, that's starting forecast hour from 1st may 1981 insted of 15th may 1981. When I use gsm script it is showing time 15th may 1981.
Please let me know what mistake I am doing ?.
I would like to run the model for may 15th to october 15th with observed SST. What sst I should use monthly SST or Daily?.

with regards,
sreejith

khideki
khideki 19 December 2006 17:43:06

1. first of all, run gsm script (without any modification) and make sure your system is able to run the model correctly.

2. gsm script if for test purpose and can be run from 1990 March 9th only, using sample initial data in libs/con.

3. see this page for pvmw question
SST%20data%20files.html#130881

4. see tha main cases script page for more instructions.

Hideki

khideki

surface merge

khideki 20 February 2007 12:30:48

A question from Sreejith:

Could you please tell me what is surface merging and whether it is required for seasonal forecast?.

mkanamitsu
mkanamitsu 20 February 2007 15:57:52

surface merge should be always on for all the run script to work. It was introduced long time ago to test merging sufrace program into forecast model. Please see src/def/regacy.h for other variables no more in use. These variable definitions should not be changed.