CAM produces a series of NetCDF format history files containing atmospheric gridpoint data generated during the course of a run. It also produces a series of NetCDF format restart files necessary to continue a run once it has terminated successfully and a series of initial conditions files that may be used to initialize new simulations. The contents of these datasets are described below.
History files contain model data values written at specified frequencies during a run. Options are also available to record averaged, instantaneous, maximum, or minimum values on a field-by-field basis. If the user wishes to see a field written at more than one time frequency (e.g. daily, hourly), additional history files must be declared. This functionality is available via setting namelist variables.
History files may be visualized using various commercial or freely available tools. Examples include the the NCAR Graphics package, FERRET, ncview, MATLAB, AVS, IDL, and Yorick. For a list of software tools for interacting with NetCDF files, view the link Software for Manipulating or Displaying NetCDF Data.
CAM is set up by default to output a set of fields to a single monthly average history file. There is a much larger set of available fields, known as the "master field list," from which the user can choose fields of interest to add to the history file via namelist settings. Both the set of default fields and the master field list depend on how CAM is configured. Due to the large number of fields we have chosen to make lists of fields for some standard configuration available via linked documents rather than to inline the lists here. Each of the field list documents is comprised of tables containing the lists of fields that are output by default as well as the master field list.
Note: The master field list tables may contain some fields that are not actually available for output. The presence of a field in the master field list is a necessary, but not sufficient condition that the corresponding field in the history file will contain valid data. This is because in some instances fields are added to the master field list (this is done in the source code) even though that field may not be computed in the configuration that is built (specified via the arguments to configure). When adding non-default fields to the history file it's important to check that the fields contain reasonable data before doing a long run.
The following links provide tables of default and master field lists for
some standard model configurations which are characterized by the values of
-chem arguments to configure.