8.3 History Files

History files contain gridded data values written at specified times during a model run. By default, the history files will be written to the directory history_dir defined in the namelist. The netCDF file names are prepended by the character string given by history_file in the ice_nml namelist. This character string has been set according to CCSM Output Filename Requirements. If history_file is not set in the namelist, the default character string 'iceh' is used. The user can specify the frequency at which the data are written. Options are also available to record averaged or instantaneous data. The form of the history file names are as follows:

Yearly averaged: $CASE.csim.h.yyyy.nc
Monthly averaged: $CASE.csim.h.yyyy-mm.nc
Weekly averaged: $CASE.csim.hw.yyyy-mm-dd-sssss.nc
Daily averaged: $CASE.csim.h.yyyy-mm-dd.nc
Instantaneous (histfreq = 'y', 'm', 'w' or 'd'): $CASE.csim.h.yyyy-mm-dd-sssss.nc
Instantaneous (written every dt, histfreq = 1): $CASE.csim.h.yyyy-mm-dd-sssss.nc

$CASE is set in the main setup script. All history files are written to /hist in the executable directory. Changes to the frequency and averaging will affect all output fields. The best description of the history data comes from the file itself using the netCDF command ncdump -h filename.nc. Variables containing grid information are written to every file and are listed in Table 10. In addition to the history files, a netCDF file containing a snapshot of the initial ice state is created at the start of each run. The file name is $CASE.csim.i.yyyy-mm-dd-sssss.nc and is written to /init in the executable directory.

8.3.1 Caveats Regarding Averaged Fields

In computing the monthly averages for output to the history files, most arrays are zeroed out before being filled with data. These zeros are included in the monthly averages where there is no ice. For some fileds, this is not a problem, for example, ice thickness and ice area. For other fields, this will result in values that are not representative of the field when ice is present. Some of the fields affected are:

• Flat, Fsens - latent and sensible heat fluxes
• albsni - snow/ice broadband albedo
• evap - evaporative water flux
• Fhnet - ice/ocn net heat flux
• Fswabs - snow/ice/ocn absorbed solar flux
• strairx, strairy - zonal and meridional atm/ice stress
• strcorx, strcory - zonal and meridional coriolis stress

For some fields, a non-zero value is set where there is no ice. For example, Tsfc has the freezing point averaged in, and Flwout has $\sigma T_f^4$ averaged in. At lower latitudes, these values can be erroneous.

To aid in the interpretation of the fields, a field called ice_present is written to the history file. It contains information on the fraction of the time-averaging interval when any ice was present in the grid cell during the time-averaging interval in the history file. This will give an idea of how many zeros were included in the average.

The second caveat results from the coupler multiplying fluxes it receives from the ice model by the ice area. Before sending fluxes to the coupler, they are divided by the ice area in the ice model. These are the fluxes that are written to the history files, they are not what affects the ice, ocean or atmosphere, nor are they useful for calculating budgets. The division by the ice area also creates large values of the fluxes at the ice edge. The affected fields are:

• Flat, Fsens - latent and sensible heat fluxes
• Flwout - outgoing longwave
• evap - evaporative water flux
• Fresh - ice/ocn fresh water flux
• Fhnet - ice/ocn net heat flux
• Fswabs - snow/ice/ocn absorbed solar flux

When applicable, two of the above fields will be written to the history file: the value of the field that is sent to the coupler (divided by ice area) and a value of the flux that has been multiplied by ice area (what affects the ice). Fluxes multiplied by ice area will have the suffix _aice appended to the variable names in the history files. Fluxes sent to the coupler will have "sent to coupler" appended to the long_name. Fields of rainfall and snowfall multiplied by ice area are written to the history file, since the values are valid everywhere and represent the precipitation rate on the ice cover.

8.3.2 Changing Frequency and Averaging

The frequency at which data are written to a history file as well as the interval over which the time average is to be performed is controlled by the namelist variable histfreq. Data averaging is invoked by the namelist variable hist_avg. The averages are constructed by accumulating the running sums of all variables in memory at each timestep. The options for both of these variables are described in Table 5. If hist_avg is true, and histfreq is set to monthly, for example, monthly averaged data is written out on the last day of the month.

8.3.3 Changing Content

The second namelist in the setup script controls what variables are written to the history file. To remove a field from this list, add the name of the logical variable associated with that field to the &icefields_nml namelist in csim.buildnml_prestage.csh (coupled) or csim.setup.csh (uncoupled) and assign it a value of .false.. For example, to remove ice thickness and snow cover from the history file, add

&icefields_nml
    f_hi   =  .false.
  , f_hs   =  .false.
/

to the namelist.

Table 10: Time and Grid Information Written to History File

Field	Description	Units
time	model time	days
time_bounds	boundaries for time-averaging interval	days
TLON	T grid center longitude	degrees
TLAT	T grid center latitude	degrees
ULON	U grid center longitude	degrees
ULAT	U grid center latitude	degrees
tmask	ocean grid mask (0=land, 1=ocean)
tarea	T grid cell area	m$^{2}$
uarea	U grid cell area	m$^{2}$
dxt	T cell width through middle	m
dyt	T cell height through middle	m
dxu	U cell width through middle	m
dyu	U cell height through middle	m
HTN	T cell width North side	m
HTE	T cell width East side	m
ANGLET	angle grid makes with latitude line on T grid	radians
ANGLE	angle grid makes with latitude line on U grid	radians
ice_present	fraction of time-averaging interval that any ice is present

Logical Variable	Description	Units
f_hi	grid box mean ice thickness	m
f_hs	grid box mean snow thickness	m
f_Tsfc	snow/ice surface temperature	C
f_aice	ice concentration (aggregate)	%
f_aice1	ice concentration (category 1)	%
f_aice2	ice concentration (category 2)	%
f_aice3	ice concentration (category 3)	%
f_aice4	ice concentration (category 4)	%
f_aice5	ice concentration (category 5)	%
f_aice6	ice concentration (category 6)	%
f_aice7	ice concentration (category 7)	%
f_aice8	ice concentration (category 8)	%
f_aice9	ice concentration (category 9)	%
f_aice10	ice concentration (category 10)	%
f_vice1	ice volume (category 1)	m
f_vice2	ice volume (category 2)	m
f_vice3	ice volume (category 3)	m
f_vice4	ice volume (category 4)	m
f_vice5	ice volume (category 5)	m
f_vice6	ice volume (category 6)	m
f_vice7	ice volume (category 7)	m
f_vice8	ice volume (category 8)	m
f_vice9	ice volume (category 9)	m
f_vice10	ice volume (category 10)	m
f_uvel	zonal ice velocity	cm s$^{-1}$
f_vvel	meridional ice velocity	cm s$^{-1}$
f_fswdn	downwelling solar flux	W m$^{-2}$
f_flwdn	downwelling longwave flux	W m$^{-2}$
f_snow	snow fall rate received from coupler	cm day$^{-1}$
f_snow_ai	snow fall rate on ice cover	cm day$^{-1}$
f_rain	rain fall rate received from coupler	cm day$^{-1}$
f_rain_ai	rain fall rate on ice cover	cm day$^{-1}$
f_sst	sea surface temperature	C
f_sss	sea surface salinity	g kg$^{-1}$
f_uocn	zonal ocean current	cm s$^{-1}$
f_vocn	meridional ocean current	cm s$^{-1}$
f_frzmlt	freeze/melt potential	W m$^{-2}$
f_fswabs	absorbed solar flux sent to coupler	W m$^{-2}$
f_fswabs_ai	absorbed solar flux in snow/ocn/ice	W m$^{-2}$
f_albsni	snow/ice broad band albedo	%
f_aldvr	visible direct albedo	%
f_aldvi	near-infrared direct albedo	%
f_flat	latent heat flux sent to coupler	W m$^{-2}$
f_flat_ai	ice/atm latent heat flux	W m$^{-2}$
f_fsens	sensible heat flux sent to coupler	W m$^{-2}$
f_fsens_ai	ice/atm sensible heat flux	W m$^{-2}$
f_flwout	outgoing longwave flux sent to coupler	W m$^{-2}$
f_flwout_ai	ice/atm outgoing longwave flux	W m$^{-2}$
f_evap	evaporative water flux sent to coupler	cm day$^{-1}$
f_evap_ai	ice/atm evaporative water flux	cm day$^{-1}$
f_Tref	2 m reference temperature	C
f_Qref	2 m reference specific humidity	g/kg
f_congel	basal ice growth	cm day$^{-1}$
f_frazil	frazil ice growth	cm day$^{-1}$
f_snoice	snow-ice formation	cm day$^{-1}$
f_meltb	basal ice melt	cm day$^{-1}$
f_meltt	surface ice melt	cm day$^{-1}$
f_meltl	lateral ice melt	cm day$^{-1}$
f_fresh	ice/ocn fresh water flux sent to coupler	cm day$^{-1}$
f_fresh_ai	ice/ocn fresh water flux	cm day$^{-1}$
f_fsalt	ice to ocn salt flux sent to coupler	kg m$^{-2}$ day$^{-1}$
f_fsalt_ai	ice to ocn salt flux	kg m$^{-2}$ day$^{-1}$
f_fhnet	ice/ocn net heat flux sent to coupler	W m$^{-2}$
f_fhnet_ai	ice/ocn net heat flux	W m$^{-2}$
f_fswthru	SW transmitted through ice to ocean sent to coupler	W m$^{-2}$
f_fswthru_ai	SW transmitted through ice to ocean	W m$^{-2}$
f_strairx	zonal atm/ice stress	N m$^{-2}$
f_strairy	meridional atm/ice stress	N m$^{-2}$
f_strtltx	zonal sea surface tilt	m m$^{-1}$
f_strtlty	meridional sea surface tilt	m m$^{-1}$
f_strcorx	zonal coriolis stress	N m$^{-2}$
f_strcory	meridional coriolis stress	N m$^{-2}$
f_strocnx	zonal ocean/ice stress	N m$^{-2}$
f_strocny	meridional ocean/ice stress	N m$^{-2}$
f_strintx	zonal internal ice stress	N m$^{-2}$
f_strinty	meridional internal ice stress	N m$^{-2}$
f_strength	compressive ice strength	N m$^{-1}$
f_divu	velocity divergence	% day$^{-1}$
f_shear	strain rate	% day$^{-1}$
f_opening	lead opening rate	% day$^{-1}$
f_sig1	normalized principal stress component
f_sig2	normalized principal stress component
f_daidtt	area tendency due to thermodynamics	% day$^{-1}$
f_daidtd	area tendency due to dynamics	% day$^{-1}$
f_dvidtt	ice volume tendency due to thermo.	cm day$^{-1}$
f_dvidtd	ice volume tendency due to dynamics	cm day$^{-1}$
f_mlt_onset	melt onset date
f_frz_onset	freeze onset date