1 The Datacube Package

DATACUBE is a collection of C-shell scripts layered on top of various pieces of the Starlink Software Collection (SSC) for the analysis of spectral datacubes, such as from integral-field spectrometers. The scripts permit graphical selection of spatial and spectral regions and features using the cursor.

Datacube provides facilities:

The backbone of the Datacube Package is the IFU Data Product Cookbook (SC/16). This includes illustrated examples of the Datacube scripts, script programming tips, and lists of potentially useful applications to manipulate and analyse cubes.

The C-Shell approach was a deliberate design decision which was made for two main reasons: first, due to the still developing nature of the field it is difficult to pin down exactly visualisation tasks will be necessary, and the choice of csh means that the existing scripts can be easily modified by the end user to do exactly the job required; second, after examining the Starlink Software Collection (SSC) it was realised that a great deal of the functionality required to analyse IFU data already existed and it would be wasteful to re-invent the wheel by re-coding large chunks of software.

Some tasks once mature can, and should, be re-coded in FORTRAN or C, mainly for the speed and memory advantages that this will bring. For instance the velmap script as implemented is rather slow and memory intensive, it seems likely that this will be the first candidate to be re-coded in a lower-level language.

Hopefully, while I haven’t managed to think of everything, there is enough ground work here so that the approach to solving your data visualisation or data cube manipulation problem is obvious, even if the package doesn’t have a script or application to do exactly what you require.

While this document contains the detailed descriptions of the shell scripts included with the package, the “how to” information detailing their use can be found in SC/16.

I would welcome comments, criticisms, contributions and corrections to the package once people start to figure out what they want to do with this sort of data. Comments can be sent either to the authors at aa@astro.ex.ac.uk or mjc@star.rl.ac.uk or to the Starlink software librarian starlink@jiscmail.ac.uk.