General Data Analysis Facilities

Other STSDAS Packages

Some STSDAS packages that are still in development contain useful tasks that may be robust enough for some purposes. Because these packages continue to be developed, the descriptions contained in this manual are only brief synopses of their functions. Refer to the online help for more current summaries of the package capabilities. These packages include the testdata and playpen packages. You may also want to investigate the contrib package, which contains programs developed by users outside the STSDAS group.

There are also two other packages that may be of interest to some users, but that we do not necessarily recommend using: sobsolete and vdisplay. The sobsolete is where tasks that are about to be retired from STSDAS are stored. By putting these tasks in their own package, users become aware that a task's functionality has either been replaced by another task or is no longer deemed useful (type help sobsolete opt=sys for more information). If you need and use a task in the sobsolete package, you should notify the HotSeat staff to find out why the task is being retired.

The vdisplay package is provided to support Gould DeAnza image display devices on VAX systems. These devices are not often used any longer, and this package will eventually be retired.

Test Data

The STSDAS testdata package will generate simple test spectra. Task compspec creates a one-dimensional image using a user-specified algebraic expression to compute the pixel values. The task overscan modifies an existing spectrum to simulate the effect of overscanning with the Faint Object Spectrograph.


Tasks or packages that are being developed are often placed in the playpen package so that users can exploit the latest software even before it is fully integrated into STSDAS. While playpen may rightly be considered a testbed for new utilities, the tasks themselves have undergone the same testing and validation as other STSDAS tasks. Users should simply be aware that the functionality, the user interface, the parameters, and even the task names themselves may change with little or no warning. As tasks or packages are fully developed, they may be moved elsewhere within STSDAS--for example, to the analysis package--but this will usually occur during major releases, not minor patch releases.

The playpen package currently contains a prototype of the registration package, which allows images to be aligned based on common features or on FITS-style World Coordinate information stored in the image headers. This package currently accommodates sets of images with different pixel scales, central positions, and rotations. Accommodation of more complicated transformations is planned.

The newredshift task is a handy utility for modifying the redshift of one-dimensional spectra. The task operates by modifying the dispersion relation, which is stored in various keywords in the image header, and can handle different input and output redshifts, as well as logarithmic wavelength scales.

The hsubtract and fill tasks implement the Baade-Lucy algorithms for dealing with sky background in direct images. This method is designed to take care of the spatial high-frequency component in the background. It fills pits (pixels below the average background) by local flux reshuffling. The hsubtract task subtracts a reference background, either from a reference image or a computed background. These tasks, along with the edge task for modifying image edges by windowing or overlapped edge extension, are useful for conditioning images before using the image restoration tasks.

The lgrlist task is an enhancement of the stplot.grlist task. It expands a list of image names, including a range of group numbers, into an explicit list of individual image names with an explicit section and group number. This input list of images can include wildcards, @-lists, etc.

Two HST-specific utilities are also included in the playpen package. The jimage task uses a list of U2-U3 pointing coordinates and creates an image that reflect the jitter information. The hstpos task plots the position of HST at specified times on a Mollweide projection of the world. The task uses information contained in the headers of uncalibrated, GEIS-format HST data, and is most useful for determining whether a given observation was likely to be affected by terminator crossings, or by the South Atlantic Anomaly. The geo2mag task for converting geocentric coordinates to and from geomagnetic coordinates, while not HST specific, is also useful for gaining insight about the quality of HST data.

User-Contributed Software: contrib

The contrib package contains tasks or small packages that are contributed from the HST user community. They are included in the STSDAS distribution as a service to our users, but the responsibility for the code, the accuracy of the output, and advice for using this software rest with the authors.

The acoadd task optimally coadds images with differing resolution using an algorithm developed by Leon Lucy and Richard Hook ( at the ST-ECF in Garching.

The gaussfit task is a general utility for solving least-squares and robust estimation problems. It includes a programming language and an algebraic manipulator for analytically calculating the required partial derivatives. While this version of the program is outdated, it still provides some functionality beyond that found in the stsdas.fitting package. A more recent version (which was developed for the PC and Macintosh environments) can be obtained from the University of Texas at Austin.

The redshift package contains two tasks for determining redshifts and dispersion relations, using either the Fourier Quotient technique (Sargent et al., 1977, ApJ, 212, 326) or the cross-correlation technique (Tonry and Davis, 1979, AJ, 84, 1511). This package was developed by Gerard Kriss at Johns Hopkins University (

The specfit task is very useful for fitting multi-component models to one-dimensional spectra. The components include polynomial, blackbody, and powerlaw continua, and Gaussian, Lorentzian, and user-defined line profiles. Users can select either the downhill simplex or the Levenberg-Marquardt method of solution. The task was written by Gerard Kriss at Johns Hopkins University (

Finally, the vla package contains three IRAF scripts to reduce radio or optical imaging spectroscopy data cubes into a set of two-dimensional images representing various moments of the data cube along the velocity axis. These moments include the total intensity, the mean and peak velocity, and the profile dispersion (i.e., the width). These scripts were written by Diane Gilmore ( and Ron Allen at STScI.

Test Data
User-Contributed Software: contrib

Generated with WebMaker