composite [ parameter=value ... ]
[ inputfile ... outputfile ]
Parameters are: include_vars, method, min_good_vals.
composite combines several input datasets element-by-element, according to a user-specified rule. Data that has been assigned to bad_value is not included in the composite. The mean, median, first good value, last good value, minimum value, or maximum value can be used as the selection rule. Up to 255 inputs can be combined to produce one output. (The operating system may impose a limit on the number of open files that reduces the effective limit of the number of inputs.)
Each variable to be combined must exist in all input datasets. Variable dimensions must be the same across input datasets.
The earth transform from the first dataset is copied to the output. If an earth transform does not exist for the first dataset, the output dataset will not have an earth transform.
List of variables to composite. If list is prefixed with a minus sign, the listed variables are excluded from the composite. Wildcards * and ? are allowed.
The default is to composite all variables.
This is the rule used to compute the element-by-element variable composite. The composite method can be either the mean, median, first good value, last good value, minimum or maximum value. If there is no good data for a given element, that element is assigned a bad (missing) value.
Valid responses are [mean, median, first, last, min, or max]. The default is mean.
OPTIONAL. This specifies the minimum number of good values needed to evaluate an aggregate function at a given point. This number must be between 1 and the number of input datasets being processed.
The default is 1.
Combine all variables.
%composite in/out files : char(255) ? Southern.Cal So_Cal.com include_vars : char(255) ? [] * method : char( 6) ? [mean] median
For any given variable, if its datatype and scaling are the same across all input datasets, then the corresponding output variable will have the same datatype and scaling.
For any given variable, if its datatype and/or scaling vary across input datasets, the corresponding output variable will be converted to either float or double datatype, with unit scaling. (The output datatype will be double only if any of the corresponding input datatypes is double.)
composite assumes that units for a given variable are the same across all input datasets. This is not checked. Variable units from the first input dataset are carried over to the output dataset.
Last Update: $Date: 1998/05/28 19:42:06 $