skip to content
XMM-Newton Guest Observer Facility


The XMM-Newton Extended Source Analysis Software, XMM-ESAS, is available in the SAS package. XMM-ESAS allows the user to model the quiescent particle background both spectrally and spatially for the EPIC detectors. It will produce background spectra for user-defined regions of the detectors and background images. The output files are FITS standard and can be used in spectral fitting packages (e.g., Xspec) or with FITS image display software (e.g., fv or ds9).

XMM-ESAS was developed at the NASA/GSFC XMM-Newton GOF in cooperation with the XMM-Newton SOC at ESAC and the Background Working Group. The Background Working Group consists of representatives from a number of institutions including the SOC, MPE, Leicester, Saclay, Milan, and GSFC. The focus is on extracting the greatest possible science from EPIC observations of extended sources and the diffuse background. Our goal has been to understand to the various background components experienced by the EPIC detectors, derive methods for modeling and/or removing the backgrounds, produce software to do so, and make the resulting products available to the general community. To aid astronomers, the SOC has created web pages detailing the current understanding of various background components.

A couple of papers demonstrate the utility of the XMM-ESAS analysis procedures. Snowden et al. 2004, ApJ,610,1182 Presented a serendipitous observation of solar wind charge exchange (SWCX) X-ray emission while Snowden et al. 2008, A&A, 478, 615 presented a catalog of galaxy clusters with their flux, temperature, and abundance radial profiles.

The XMM-ESAS users guide can be downloaded from the GOF at: xmm-esas.pdf. The document is in the process of being revised for the SAS V19 release but is not quite finished, but this partially upadted version is better than the old one as things have changed significantly.

The additional CalDB files required be XMM-ESAS can be downloaded from the GOF at: esas-caldb.tar.

If problems arise or if the results look odd, please contact the GSFC GOF Helpdesk at: Suggestions for improving the software or documentation are also welcome.

Warnings and Watchouts

General Notes

  • NOTE 1: Do not run multiple tasks in the same directory at the same time as this can cause conflicts leading to significant errors.
  • NOTE 2: Occasionally the backscale calculation will fail leaving the backscal keyword as 0 (zero). This will cause the calculation of the background spectrum to fail. This probably has something to do with the complication of the selection expression and we are looking into a fix. However, until a fix is found the affected CCD should be excluded.
  • NOTE 3: If a CCD is identified by mos-back as operating in an anomalous mode processing will stop. mos-spectra and mos-back should then be rerun excluding the CCD.
  • NOTE 4: The standard light-curve filtering by espfilt removes only those time intervals where the soft proton background count rate significantly fluctuates. It will not remove a relatively constant low level background. Filtering will not remove the SP background and in most cases the user will need to fit a residual component, and will occasionally need to exercise discretion on whether the observation is salvageable.
  • NOTE 5: Short exposures are problematic. Observations with clean exposures greater than ~7 ks will probably be OK. Shorter exposures should be approached with caution. Currently pn_back and mos_back do not die gracefully with a warning. The *-spec.qdp plots should be examined for reasonableness.

Future Developments

Since the last release of ESAS:

  • We have discovered the source of the temporal variation of the shape of the quiescent particle background spectrum. The spacecraft samples different particle environments over its orbit. Most of the time the spacecraft is in a rather benign and temporally stable environment. We can now identify those time intervals in which the spectrum will be different, and we can quantify the significance of the difference. This work was inspired by the AREMBES group.
  • We have identified a new algorithm for determining which pn observations have sufficient soft proton flare removal to be used for determining the quiescent particle background spectrum. Although this algorithm removes the bulk of available observations, the remaining data produce a far more consistent background spectrum.
  • We are developing a new tool for determining whether MOS chips are in an anomalous state. This will be a significant improvement over the emtaglenoise routine.
  • We have begun to produce average quiescent background spectra that are useful for the bulk of XMM observations. We will be producing a tool that will allow users to determine whether these average spectra are appropriate for any individual observation.
  • To see the work behind these changes, download our presentations from the XMM-Newton CalOps Meetings for:

If you have any questions concerning XMM-Newton send e-mail to