skip to content
XMM-Newton Guest Observer Facility


    Frequently Asked (and Answered) Questions

    This page contains some of the common questions about the observation enhancement process.
    You may want to look at the enhancement page for the general information.

    We thank all the people who helped us generating this page in particular Matteo Guainazzi, Rudy Much, Dave Lumb, and Matthias Ehle.

    Question: I've just received an e-mail from the XMM folks and they want to "enhance" my proposal -- They suggest using a filter because of some potential problems in the EPIC due to nearby bright optical source. Should I follow their suggestions??

    Answer: Any bright sources within 2 arcmin of the observation target can contribute some optical light which causes a shift in the X-ray energy scale because the offset map is non-uniform. (Have a look at the EPIC optical loading paper for more information. In order to avoid this, a catalogue (the Optical Monitor Operation Catalogue or OMOC) is searched and any bright sources are flagged. The standard recommendation is then to increase the thickness of the filter (changing from thin to medium for example). The OMOC used in this task is a compilation of several catalogues. For more informations on OMOC see Specification for the OMOC and Bright Source Check with OMOC.
    There are many caveats to note:

    1) It appears that the software does not distinguish between the target and other sources. In other words, the bright optical source flagged by the code may be your target.

    2) If your source is heavily absorbed (more than several e21 cm^2), a change from thin to medium filter will probably have NO effect. We still recommend to CHECK the resulting count rates using quicksim or any programs you used to estimate the count rates in your proposal.

    3) Currently OMOC does not give entirely reliable brightness products, assuming all sources are point-like. In other words some extended objects which will NEVER be a problem with normal filters get flagged.

    4) The calculations are for the brightest pixel of a point source optical PSF. If that is at the edge of the nominal 2 arcmin box then the likely effect on one's on-axis source are smaller but still flagged by the software.

    As a result, we offer a cautionary note concerning the enhancement process because it cannot be pre-judged for each science program what is critical or what is not.

    5) In the EPIC-PN the offset map (the bias frame in optical astronomer parlance) is made on a pixel by pixel basis and should be able to cope with local brightness fluctuations. However any optically-bright leakage would certainly affect the CTI and energy scale of X-ray source signal in those columns.

    6) The EPIC-MOS uses either a fixed offset or calculates row and column averages (depending on the mode) and therefore does NOT accurately subtract a local offset. One can check the effect locally to a bright source by looking at the raw data files, checking histograms of E3 and E4 values around the bright star and comparing with global histograms - a difference can translate to a misplaced energy scale. In principle the pipeline software (emchain) can make its own average background map to make a subtraction. However this hasn't been properly tested and in any case the additional noise, pattern migration, and relative threshold selection differences will also change detection efficiency and response distributions in a way yet uncharacterized.

    Question: I've asked for GRISM observation but the USER DEFINED mode is not implemented. What's the deal?

    Answer: Many observers, following the recommendations in the AO1 documentation, have indeed requested grisms observation in OM SCI USER DEFINED mode. In this mode windows can be defined in pixel or in sky coordinates. Most observers have chosen the latter option (it is easier and not boresight dependent). However, the conversion from sky coordinates to OM pixels in the ground segment does not work properly. This bug is being fixed now, but in the meantime an observer in this situation has only one of the following choices:

    1) Put the enhancement of the whole proposal on-hold, waiting for this problem to be fixed. This, of course, can imply a long and unknown delay in the scheduling of all the observations in her/his proposal

    2) Convert sky coordinates in pixels manually. Most of the observers (basically everybody except some members of the OM team itself) are currently unable to do this.

    3) Convert the SCI USER DEFINED mode to standard IMAGE mode. This implies a potential loss of efficiency by a factor 5. One of the current suggestion is to keep the GRISM filter and ask for the RUDI5 exposure type. Note that there is currently no distortion map for the grism and this requires some further calibration analysis at MSSL (not yet completed).

    4) Change the filter. This choice is the ultimate responsibility of the observer because it concerns the scientific goals of the proposal and it depends on how critical OM grism data is to reach such goals.

    If your question is not answered here, you can send a mail to the XMM-Newton US GOF from our Feedback form.

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