HEAD 2002 Pileup Talk  



In this section, a few examples of the usefulness of the pileup model are presented. These include:

For each of these observations, the spectral shape was assumed to be given by an absorbed powerlaw, and isis was used for the spectral analysis. For each example, isis was fed the following script to set it up for the fitting process:
      load_dataset ("pha.fits", "rmf.fits", "arf.fits");
      fit_fun ("phabs(1)*powerlaw(1)");
      set_par ("phabs(1)", 0.1, 0, 0, 1);
      set_par ("powerlaw(1)", [0.006, 1.7], [0,0], [0,0], [0.1,3.0]);
      set_kernel (1, "pileup");
      set_par ("pileup<1>", [1,1,0.5,0.95], [1,1,0,0]);
      group_data (1, 8);
      xnotice_en (1, 0.5, 12);
Note the group_data statement. This was used to group every 8 PHA channels together. With the ability to dynamically group data on the fly, there is no need for auxiliary programs such as dmgroup or grppha. Finally note that only data ranges from 0.5 to 12 keV were used for the analysis.

The actual fits were carried out using the isis fit_counts function. The "residual" plots were created using rplot_counts, and contour plots were created using the conf_map_counts and conf_plot functions.

For sherpa, the (almost) equivalent script is:

      DATA 1 pha.fits
      INSTRUMENT 1 = RSP[my1]("rmf.fits", "arf.fits")
      source 1 = w*p
      w.nH = 0.1
      w.nH.min = 0
      w.nH.max = 1
      thaw w.nH
      p.norm = 0.006
      thaw p.norm p.PhoIndx
      PILEUP 1 = jdp1
      jdp1.ftime = 3.2
      jdp1.n = 1
      jdp1.alpha = 0.5
      jdp1.f = 0.95
      thaw jdp1.f jdp1.alpha
Please note: sherpa does not have the ability to dynamically group pha data so you will have to use the dmgroup CIAO tool before starting sherpa for this task. Also sherpa does not permit data ranges to be excluded when using the pileup model. Follow the following links for more information about how the model was applied to Q0836+7104, NGC 4579, and M81.

Previous: Spectral Fitting Software Next: Q0836+7104

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