The plot operation is based on the concept of displaying Dependent Variables as functions of the Independent Variables. In each data file, the variables are conceptually divided into Independent and Dependent Variables. The Dependent Variables are further divided into Auxiliary and Primary Variables. The Auxiliary Variables are functions of only one Independent Variable, called the Independent Variable Mark (IVM), whereas the Primary Variables are functions of all Independent Variables.
The IVM is the Independent Variable with the largest index, i.e., X1 if there is only one Independent Variable, X2 if there are two, etc. The IVM is always explicitly recorded in the data files. Numeric Auxiliary Variables can only be plotted as functions of a numeric IVM.
For one file format, FFI 2160, the IVM is type character, and the file may contain Auxiliary Variables which are also type character. For those files, values of the character string IVM can be selected, but any Auxiliary Variables which are type character are ignored by the plot operation.
For all files, each Primary Variable can be plotted as a function of any of the numeric Independent Variables. For files with more than one numeric Independent Variable, each Primary Variable can also be contoured as a function of any two of the Independent Variables, if there are enough data points to construct the plot.
The plot operation uses a sequence of HTML forms to collect the information needed to construct the plot. In the process, the range of each variable is determined and displayed in square brackets, [ ], at the end of each variable description. This can be a handy way to tell whether or not a particular variable is worth plotting.
Many of the archived data are not on a uniform grid, so in the interest of producing a plot, some data may be interpolated when there is more than one Independent Variable. The procedure for contouring irregularly gridded data uses a triangulation scheme which can give the appearance there are data in regions where there are none, so re-drawing the plot with a judicious modification of the range of the variables may be necessary to determine which regions of the plot actually contain observations.
The plot routines use the variable names from the data files to construct the titles for the plot and the axes. Titles which are too long to fit on the plot are truncated.
Some operations, like reading large data files or contouring very detailed fields, may take longer than the web server allows for send/receive operations. When that occurs your browser will post an error message indicating no data was received because of a server timeout. To help reduce this kind of wasted resources, the plot routines will refuse to read large data files, and inform you of the maximum file size they think they can read within the allotted time. However, the speed at which the system can read files and contour the data depends on the system's load, so the plot routines may overestimate the system's capabilities and you may still get a server timeout message.
Curator: Steven Gaines - NASA Official: Mike Craig Updated: 2004-10-07