Go to the bottom of this page. See the search engine and sub-section links.
Go to next page Go to previous page Go to top of this section Go to top page Go to table of contents

Previous Section Headers

User's Guide to NCAR CAM2.0

2. Using CAM2.0

2.3 Model Input Datasets

CAM2.0 is a combination of atmosphere, land, ocean and sea-ice components. In what follows we discuss the input datasets required by each of these components. CSMDATA refers to the root directory where the distribution datasets have been untarred by the user.

2.3.1 Atmosphere Component Datasets

Input datasets needed for the atmospheric component provide initial state data, ozone boundary data and water vapor absorptivity/emissivity data. Only the ozone dataset contains time-variant input data which is based on a 365 day year with no leap years. All initial and boundary datasets are in NetCDF format.

In general input datasets following the following convention

<type><simulation date><resolution><Spectral truncation> <Vertical levels><creation date> Where Initial Conditions Dataset

The initial conditions, dataset is specified by namelist variable NCDATA. This dataset contains initial values of the prognostic variables UVTQPSCWATTSICESNOWHICE  and TS1 through TS4, the surface geo-potential field PHIS, the grid-box land fraction LANDFRAC, the land ocean transition mask LANDM, and the standard deviation of geo-potential height SGH. When running flux-coupled the fields: TSICE, SNOWHICE, LANDFRAC and TS1 through TS4 are not included on the dataset. Required initial fields are outlined in the table below.  By default, the model periodically writes an instantaneous history file containing these fields for possible use as initial datasets in other runs. The frequency with which these instantaneous history datasets are written is controlled by the namelist variable INITHIST.

Table 2.6: Atmospheric Component Initial Dataset Fields
History file 
Field Name
Variable Name
Multi-Level Fields
T T temperature  (K)
U u zonal wind component (m/sec)
V v meridional wind component  (m/sec)
Q q water vapor specific humidity (KgH2O/Kgair)
CWAT q(:,:,,ixcldw,:) cloud water mass mixing ratio (Kgcloud/Kgair)
Single Level Fields
PHIS surface geo-potential (m2/s2)
PS Ps surface pressure  (Pa)
SGH SGH standard deviation of geo-potential height  (m)
LANDM landm Land Ocean transition mask:
= 0   ocean 
= 1   continent 
= 0-1 transition
TSICE Tice CSIM sea-ice model snow/ice surface temperature (not stored on CCSM flux-coupled simulations).
SNOWHICE   Snow depth over ice
LANDFRAC   Land fraction
TS1,TS2,TS3,TS4 Ts Four CSIM sea-ice subsurface temperature levels (not stored on CCSM flux-coupled simulations) (K) 

Finite-volume dynamics uses the same set of prognostic variables as shown above, except that U and V are on a staggered grid and are identified with names US and VS.  When running with finite-volume dynamics, the initial dataset must contain US and VS; U and V are ignored.

All fields are instantaneous values.  In addition to the fields listed above, the initial dataset contains information on the model date, the dimensionality of the fields, the spectral truncation, and the latitudes, longitudes, and vertical levels of the data.

A T42 (64 latitudes x 128 longitudes grid resolution) 26-level initial dataset is provided with the CAM2.0 distribution:


Other initial datasets at different model resolutions can also be found in the directories



CSMDATA/atm/cam2/inic/fv Ozone Dataset

The ozone boundary dataset contains ozone volume mixing ratios which are constant for a given latitude. This dataset is defined on a pressure grid (unlike the hybrid grid of the model). These values are interpolated to the model vertical levels at each model grid point. The ozone dataset is in NetCDF format and contains the fields PS and OZONE. PS is a constant 1000 mb pressure field needed for interpolation.  The ozone dataset is specified by namelist variable BNDTVO. Unlike the initial conditions file, CAM is capable of interpolating the ozone dataset to any vertical and or horizontal resolution so the user need not specify new ozone datasets when changing model resolution.  The default ozone dataset provided with the distribution is

CSMDATA/atm/cam2/ozone/noaao3.1990.21999.nc. Water vapor absorptivity/emissivity dataset

This is a lookup table for water vapor absorption. It is specified by setting the namelist variable, ABSEMS_DATA. The default dataset provided with the dataset distribution can be found in


2.3.2 Ocean Component Datasets

The ocean component is a data ocean model which reads in input sea surface temperature (SST) data. Running the data ocean component requires a time-variant SST dataset. The SST dataset must be at the model resolution and is specified by the namelist variable BNDTVS.

The standard SST dataset is a climatological dataset containing 12 monthly time samples. The model can also read multi-year SST datasets.  If a multi-year SST dataset is used, the namelist variable, SST_CYC must be set to .false. (it's default value is .true.).

An SST dataset on a (64 X 128) Gaussian grid can be found in


Other SST datasets, at different model resolutions, can also be found in this directory.

2.3.3 Sea-Ice Component Datasets

The sea-ice component in CAM2.0 is a simplified version of the CCSM CSIM4 code. This component requires ice coverage data. The ice coverage data is contained in the same file as the SST data required by the ocean component. For the standard configuration, ice-coverage data is found in


2.3.4 Land Component Datasets

For a full discussion of CLM2.0 input datasets see the CLM2.0 User's Guide at
CLM2.0 always requires a dataset providing plant functional type physiological constants. This dataset is specified via the CLM2.0 namelist variable FPFTCON.

A time-invariant CLM2.0 surface dataset will be generated at run time if the CLM2.0 namelist variable FSURDAT is set to blank. If this is the case, additional settings must be provided for MKSRF_FGLACIER, MKSRF_FLAI, MKSRF_FLANWAT, MKSRF_FSOICOL, MKSRF_FSOITEX, MKSRF_FURBAN, MKSRF_FVEGTYP. If a surface dataset at the model resolution already exists, then FSURDAT should be set appropriately. The default land surface dataset provided with the distribution is


Other provided surface datasets at other model resolutions can also be found in this directory.

CLM2.0 initial conditions will be generated at run time if the CLM2.0 namelist variable FINIDAT is blank (the default setting). If spun up values exist for a model run, then FINIDAT should be correspondingly set. The default initial dataset provided with the distribution is:


This initial dataset should only be used in conjunction with the default CLM2.0 surface dataset given above.

Sub Sections

    2.3.1 Atmosphere Component Datasets Initial Conditions Dataset

    2.3.2 Ocean Component Datasets

    2.3.3 Sea-Ice Component Datasets

    2.3.4 Land Component Datasets

 Go to the top of this page. See links to previous section headers.
Go to next page Go to previous page Go to top of this section Go to top page Go to table of contents

 Search for keywords in the CAM2.0 Users GuideSearch page

Questions on these pages can be sent to... erik@ucar.edu .

$Name: $ $Revision: 1.1 $ $Date: 2004/06/08 02:57:21 $ $Author: jmccaa $