Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://xmm.vilspa.esa.es/docs/documents/CAL-SRN-0209-1-0.ps.gz Äàòà èçìåíåíèÿ: Wed Jan 24 14:23:09 2007 Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 22:45:30 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: south pole

XMM­Newton CCF Release Note
XMM­CCF­REL­209
OM Grisms Astrometry
A. Talavera, A. Breeveld
June 13, 2006
1 CCF components
Name of CCF VALDATE EVALDATE List of Blocks
changed
XSCS flag
OM ASTROMET 0012 2000­01­01T00:00:00 --- FILTER­GRISM10 NO
--- FILTER­GRISM20 NO
--- POLYNOM MAP NO
--- POLYNOM MAP2 NO
2 Changes
When using the OM grisms, the position of the sources in the field of view is translated to the grism
image in the position of their corresponding zero orders. This translation can be mapped so as to
recover from any zero order position the corresponding position of the source in the ``direct'' image.
Then the normal SAS OM astrometric tools can be used to compute the astronomical coordinates
of the source producing a spectrum through the OM grisms.
The transformation between direct image and grism image can be considered as a form of ge­
ometric distortion that can be mapped if we have both images. We have used observations of Sco
X­1 field with the V filter and both OM grisms, obtained in revs. 402 and 688.
All images were processed with SAS so as to obtain modulo 8 and geometric distortion corrected
images corresponding to the V filter and grisms images. (It should be noted that the grisms images
were not rotated). Then a similar procedure to the one used to derive the geometric distortion was
used.
A 4th order polynomial provides a good fit. Its coe#cients and the corresponding regularly
sampled distortion map, for both grisms, have been incorporated into the CCF.
1

XMM­Newton CCF Release XMM­CCF­REL­209 Page: 2
This will be used by the SAS in the following way. First, the position of the zero order of each
spectrum determined by omdetect and omgrism tasks in the omgprep rotated image has to be de­
rotated. Then the new grism distortion map can be applied to compute the position of the source in
the direct image, and from there the algorithms in omatt can be used to derive the Right Ascension
and Declination of the source producing the spectrum.
3 Scientific Impact of this Update
Astrometry in the OM grisms will allow the user to know the coordinates of the objects producing
a spectrum through the grisms. This is particularly interesting in the case of multi­object spec­
troscopy (remember that OM grisms can be used in User Defined window configuration to register
the spectrum of the target sitting at the Boresight, or in full frame to have the spectra of all objects
in the field of view).
4 Estimated Scientific Quality
The position of the sources will not have the same accuracy that can be achieved with the OM
lenticular filters. The internal consistency of the transformation is 1.1 and 1.3 pixels RMS error for
the UV and V grisms respectively.
After all transformations are applied, we expect to obtain coordinates within better than 10
arcsec for the objects whose spectrum is observed.
5 Expected Updates
This is the first version of the transformation map. The corresponding SAS tasks to use it are
already implemented and an extensive testing phase is starting now. Modification of the maps after
more data are analyzed will be implemented if necessary.
6 Test procedures
Grism observations will be processed using the newly developed code and the new contents of this
CCF. Cross­correlation with USNO catalog provided by SAS task omatt will give us the errors in
the overall transformation process.

XMM­Newton CCF Release XMM­CCF­REL­209 Page: 3
7 Summary of the test results
As pointed out before we expect errors of less than 10 arcsec. Preliminary results confirm this
expectation. Detailed results will be included in this RN at a later stage, when more extensive
testing has been performed.
References