Fit a power law in SPEX


The spectrum in the files powerl.spo and powerl.res was recorded from a source at 6 kpc distance.

  1. Load the spectrum into PYSPEX and plot it.

  2. The calibration of the instrument is not very accurate for energies below 0.3 keV and above 10 keV. Ignore those parts of the spectrum.

  3. How many data bins do you have in the spectrum? Now try to apply optimal binning to the spectrum. How many bins are left?

  4. Set up an absorbed powerlaw model. Do not forget to set the distance to the source.

  5. Next step: fit the spectrum. Is it a good fit?

  6. Calculate the errors on all free parameters and save your results in a python variable.

Learning goals

After having done this spectrum, you should know:

  • How to read a spectrum in SPEX (using

  • How to use the basic plot functionalities (using s.plot_data()).

  • How to ignore parts of the spectrum (using s.ignore()).

  • How to optimally bin a spectrum (using s.obin()).

  • How to set-up a simple spectral model (using and s.com_rel()).

  • How to set the distance (using s.dist()).

  • How to do spectral fitting (using

  • How to determine errors on parameters (using s.error()).