HEAD 2002 Pileup Talk  

Spectral Analysis Programs


The pileup model has been implemented in one form or another in isis, xspec, and sherpa. The following sections describe some important points to keep in mind when using these programs for the analysis of piled spectra. The reader is also encouraged to read Mike Nowak's page describing his experience with these programs in the context of the pileup model.


Isis is where the pileup model was first implemented and is arguably the most well-tested platform for using the model. It also does not have the limitations regarding the ignoring of PHA channels that sherpa has. Hence, it is program of choice for modeling pileup. All the examples given here were created using isis. Isis is an open source program developed principally by John Houck at the MIT Center for Space Research to support the analysis of high resolution grating spectroscopy. It was written with an open architecture in anticipation of future spectral analysis techniques, and because of this, the pileup model could be easily integrated. It is highly customizable via the S-Lang programming language and this extensibility has influenced the CIAO 3 development effort. Isis is heavily used by the MIT CXC and HETGS teams for grating spectroscopy as well as for imaging spectroscopy and spatial/spectral research programs on extended sources.

See the isis features page for more information.


Currently the xspec implementation of the pileup model is incomplete, and for this reason it cannot be recommended at this time. However, a more complete implementation is being tested. Keep an eye on the xspec web page for new xspec releases.


sherpa has a full implemention of the pileup model. Nevertheless there are a few sherpa limitations that affect the ability to use the model in a convenient or practical way. Chief among them is that one cannot ignore any PHA channels when using the pileup model. This is unfortunate because one almost always needs to ignore channels below about 0.5 keV and above about 12 keV where calibration information is uncertain or lacking. Also keep in mind that sherpa uses the value of the EXPOSURE keyword from the PHA file where it may be incorrect. The sherpa developers are aware of these limitations and they may be fixed in later releases.

Previous: Data Preparation Next: Examples

This page was last updated May 31, 2002 by John E. Davis. To comment on it or the material presented here, send email to davis@space.mit.edu.
Valid HTML 4.01! Made with JED Viewable With Any Browser