Variables in CASES97 5-Minute and 30-Minute Data Sets

Most variables will have a "time" dimension, indicating they are time-series variables. Variables with a "station" dimension were sampled at more than one station or site.

Variables with a "sample" dimension are available at a resolution of 1/sample times the resolution of the dataset. For example, in a file of 5 minute statistics, a sample dimension of 5 indicates that the variable is available at 1 minute resolution.

For more information on variable names, see the ISFF Variable Name Convention .

Variables that were slowly sampled (usually once per second, or less) start with an upper case character (P, RH, T, etc). Lower case is used for more rapidly sampled variables, typically those used in eddy correlation, like u,v,w,tc,h2o, etc.

Higher moments are represented by products of variables, separated by single quotes. For example, u'v' is the covariance of u and v.

To save space, third and fourth moments of eddy correlation variables aren't listed in the table, but are available in the datasets.

Standard NetCDF variable names should only contain alphanumeric characters and underscores. Therefore, in Netcdf files the actual variable names will be the variable name shown here, with underscores substituted for periods and quotes. The variable name shown in this table will be saved as the "short_name" attribute of the Netcdf variable.

For some variables, the sensor name is appended, e.g. "v.csat" is the lateral velocity component from the Campbell Scientific sonic anemometer (only at station #8), to distinquish it from "v" which is the same variable measured by the default sensor at each station (either an ATI or Gill sonic anemometer).

5-Minute Variables

Variables in 30-minute dataset