Scheduled Downtime
On Tuesday 24 October 2023 @ 5pm MT the forums will be in read only mode in preparation for the downtime. On Wednesday 25 October 2023 @ 5am MT, this website will be down for maintenance and expected to return online later in the morning.
Normal Operations
The forums are back online with normal operations. If you notice any issues or errors related to the forums, please reach out to help@ucar.edu

Not able to create monthly and daily history files simultaneously

Hello everyone,

Myself, Sukanta recently started working in “Climate Change Studies using Coupled Models”.

we have installed CCSM3 (Community Climate System Model) successfully on the SGI Linux-Altrix i64 server, which has maximum 32 duel core processors. We are able to run all the three types of run viz. startup, branch and hybrid with default component set (i.e. B == cam, clm, pop, csim, cpl) and resolutions T42_gx1v3 as well as T85_gx1v3. All these test runs create the monthly history files for each component with the restart files as well.

But we are facing some problems (listed below) when trying to change the history file option for the atmospheric component CAM. The model completes the run without any error massage but the changes are not reflected correctly in the history files. It is creating only default history tapes. Also for some name-list configurations, we have received error massages.

1. Not creating the secondary History File for CAM:

From the literature, we came to know that for the component model ‘CAM’, we can create maximum 6 types of history files, the first one is default monthly or daily primary history file. The files 2 to 6 are additional history files that may contain same or different fields as compared with the first history file with different time-steps and different time-samples. Some important name-list variables for creating additional history files are NHTFRQ(6), MFILT(6), NDENS(6), FINCL1-6 and FIXCL1-6.
We have tried to save a primary monthly history file (the naming convention is *.h0.*) and a secondary daily history file (the naming convention is *.h1.*) with specific field variables. The following are the details to set up the T42_gx1v3 startup run with fully active model components:
env_conf:
RUNTYPE: startup
COMPONENT SET: B (cam, clm, pop, csim, cpl)
CSIM_MODE: prognostic
GRID: T42_gx1v3

env_run:
RESUBMIT: 0
CONTINUE_RUN: FALSE
STOP_OPTION: nmonths
STOP_N: 12

Buildnml_Prestage/cam.buildnml_prestage.csh:
cat >! atm.stdin
 
I'm experiencing a similar problem with WACCM, which combines CAM dynamical core with MOZART chemistry and TIME-GCM for upper atmospheric dynamics.

I'm trying to get two history-files, or even one, with specified fields. WACCM doesn't seem to like my Namelist, and writes out "default" history-files with all possible variables.

here is my namelist:

&camexp
absems_data = '/home/bsc29/bsc29637/WACCM/inputdata/rad/abs_ems_factors_fastvx.c030508.nc'
aeroptics = '/home/bsc29/bsc29637/WACCM/inputdata/rad/AerosolOptics_c040105.nc'
airpl_emis_file = '/home/bsc29/bsc29637/WACCM/inputdata/emis/emis.air.nc'
avgflag_pertape = 'A','A'
bnd_topo = '/home/bsc29/bsc29637/WACCM/inputdata/topo/USGS-gtopo30_1.9x2.5_remap_c050602.nc'
bndtvaer = '/home/bsc29/bsc29637/WACCM/inputdata/rad/AerosolMass_V_1.9x2.5_clim_c040811.nc'
bndtvg = '/home/bsc29/bsc29637/WACCM/inputdata/ggas/noaamisc.r8.nc'
bndtvo = '/home/bsc29/bsc29637/WACCM/inputdata/ozone/pcmdio3.r8.64x1_L60_clim_c970515.nc'
bndtvs = '/home/bsc29/bsc29637/WACCM/inputdata/sst/sst_HadOIBl_bc_1.9x2.5_clim_c050526.nc'
caseid = 'waccm3_3'
cftgcm = '/home/bsc29/bsc29637/WACCM/inputdata/tgcm/waccm-bnd-P-150-20.nc'
ch2o_emis_file = '/home/bsc29/bsc29637/WACCM/inputdata/emis/emis.ch2o.nc'
co_emis_file = '/home/bsc29/bsc29637/WACCM/inputdata/emis/emis.co.nc'
depvel_file = '/home/bsc29/bsc29637/WACCM/inputdata/dvel/depvel_monthly.nc'
dtime = 1800
efield_hflux_file = '/home/bsc29/bsc29637/WACCM/inputdata/efld/coeff_hflux.dat'
efield_lflux_file = '/home/bsc29/bsc29637/WACCM/inputdata/efld/coeff_lflux.dat'
efield_wei96_file = '/home/bsc29/bsc29637/WACCM/inputdata/efld/wei96.cofcnts'
electron_file = '/home/bsc29/bsc29637/WACCM/inputdata/phot/electron.dat'
euvac_file = '/home/bsc29/bsc29637/WACCM/inputdata/phot/euvac_v1.2.nc'
euvacdat_file = '/home/bsc29/bsc29637/WACCM/inputdata/phot/euvac_v1.2.dat'
fincl1 = 'PS','Q','U','V','T'
fincl2 = 'PS','Q','U','V','T'
inithist = 'YEARLY'
itgcmcyc = 1
iyear_ad = 1950
lbc_file = '/home/bsc29/bsc29637/WACCM/inputdata/lbc/LBC_Scen=A1b_1950-2050_1.9x2.5.nc'
lbc_fixed_date = 19950101
lbc_scenario = 'FIXED'
mfilt = 1,1
mss_irt = 0
ncdata = '/home/bsc29/bsc29637/WACCM/inputdata/ic/wa319x2_smn9.2.cam2.i.1965-01-01-00000.nc'
neckel_file = '/home/bsc29/bsc29637/WACCM/inputdata/phot/neckel.flx'
nelapse = 10
nhtfrq = 10,10
nox_emis_file = '/home/bsc29/bsc29637/WACCM/inputdata/emis/emis.nox.nc'
nsrest = 0
photon_file = '/home/bsc29/bsc29637/WACCM/inputdata/phot/photon.dat'
print_energy_errors = .false.
rsf_file = '/home/bsc29/bsc29637/WACCM/inputdata/phot/RSF_GT200nm.v050316.nc'
sad_file = '/home/bsc29/bsc29637/WACCM/inputdata/sad/SAD_SAGE_1.9x2.5.c050627.nc'
snoe_ubc_file = '/home/bsc29/bsc29637/WACCM/inputdata/ubc/snoe_eof.nc'
solar_parms_file = '/home/bsc29/bsc29637/WACCM/inputdata/phot/wa_smin.nc'
sulf_file = '/home/bsc29/bsc29637/WACCM/inputdata/sulf/sulfate.M1.nc'
tgcm_ubc_file = '/home/bsc29/bsc29637/WACCM/inputdata/ubc/tgcm_ubc.nc'
woods_etf_file = '/home/bsc29/bsc29637/WACCM/inputdata/phot/woods.nc'
xs_coef_file = '/home/bsc29/bsc29637/WACCM/inputdata/phot/effxstex.txt'
xs_long_file = '/home/bsc29/bsc29637/WACCM/inputdata/phot/temp_prs_GT200nm.nc'
xs_short_file = '/home/bsc29/bsc29637/WACCM/inputdata/phot/xs_short.v050316.nc'
/
&clmexp
finidat = '/home/bsc29/bsc29637/WACCM/inputdata/ic/wa319x2_smn9.2.clm2.i.1965-01-01-00000.nc'
fpftcon = '/home/bsc29/bsc29637/WACCM/inputdata/clm/pft-physiology'
fsurdat = '/home/bsc29/bsc29637/WACCM/inputdata/clm/surface-data.144x096_06302005.nc'
/

Some variables are the same ones used in CAM, such as mfilt and nhtfrq.

Is this history-file problem a bug in NCAR models?

Kind regards
 

rneale

Rich Neale
CAM Project Scientist
Staff member
For the output error try specifying the mfilt and nhtfrq options separately e.g
mfilt(1) = ..
mfilt(2) =..

For the threading issue you may have insufficient latitudes to decompose the dynamics on 24x2 threads i.e. if you try 12x2 then that should work.
 
Top