Документ взят из кэша поисковой машины. Адрес оригинального документа : http://xmm.vilspa.esa.es/docs/documents/CAL-SRN-0290-1-0.pdf Дата изменения: Thu Sep 27 11:46:37 2012 Дата индексирования: Tue Feb 5 20:08:11 2013 Кодировка: Поисковые слова: m 2

XMM-Newton CCF Release Note
XMM-CCF-REL-290 Implementation of the Variable Boresight for RGS
R. Gonzґlez-Riestra a July 20, 2012

1

CCF comp onents
Name of CCF XMM BORESIGHT 0023 VALDATE 2000-01-01T00:00:00 EVALDATE ­ Blocks changed RGS1 ANGVAR RGS2 ANGVAR XSCS flag NO

2

Changes

Starting in SASv12.0, a variable b oresight is b eing used for the processing of data from the Optical Monitor and the EPIC cameras on-b oard XMM-Newton [1]. The variable b oresight has b een implemented in SAS adding new extensions to the CCF XMM BORESIGHT, one p er instrument, holding the values of the variation of the Euler angles as a function of time. XMM BORESIGHT 0022 is the first version of the CCF having these new extensions for OM and the three EPIC cameras. We present in this Release Note an extension of this approach to the RGS instruments. Any change in the astrometric solution implies a p otential change in the wavelength scale. The main goal of this document is the evaluation of this effect.

3

Scientific Impact of this Up date

Talavera et al. [1] have shown that the use of the variable b oresight in the processing of EPIC data results in a change in the average offset of the source p ositions along the spacecraft Z axis (i.e. the RGS disp ersion direction) from -0.6 to +0.1 arcsec. This is equivalent to a shift from -1.4 to 0.3 m° A in RGS first order sp ectra, half these values in second order sp ectra. The width of the distribution A of the offsets along this axis decreases from 1.5 to 1.2 arcsec (i.e from 3.5 to 2.7 m°). 1


XMM-Newton CCF Release

XMM-CCF-REL-290

Page:

2

Figure 1: Consistency check: Top: Comparison of the wavelength displacement b etween the sp ectra processed with fixed and variable b oresight to the corresp onding angle (converted to m°) comA puted from the CCF. The dashed lines corresp ond to the 1:1 relation. Bottom: Comparison of the displacement b etween sp ectra as a function of time (p oints) to the time dep endence of the EPIC Z correction given in [1] (solid line).


XMM-Newton CCF Release

XMM-CCF-REL-290

Page:

3

To evaluate the effect of the variable b oresight on the RGS wavelength scale, we have followed the same reduction and analysis procedures on the same data sample as in [2], but using the new CCF1 and SASv12.02 , and we have compared the wavelengths of the emission lines measured in the sp ectra processed with "Fixed" and "Variable" b oresight. We have first checked that the changes in the line p ositions are consistent with the correction applied to the b oresight. The top panel of Fig. 1 shows the comparison of the difference in the average wavelength shift p er sp ectrum (converted to arcsec) to the correction in the Euler angle for the date of the observation, computed from the values given in the CCF. Dashed lines corresp ond to the 1:1 relation, and show the overall good agreement. In the b ottom panel we represent again the average wavelength shift p er sp ectrum, but now as a function of time. Overplotted is the parametrisation of the variation of the EPIC shift along the spacecraft Z axis, as given in [1]. The agreement is also good and therefore, the correctness of the implementation of the RGS variable b oresight in SASv12.0 is confirmed.

4

Estimated Scientific Quality

The application of the variable b oresight leads to average wavelength shifts that are 1 m° smaller A ° than those obtained with the fixed b oresight. The scatter of the data is 1 mA smaller as well (see Fig. 2). The shift b etween instruments and orders remains the same, not surprisingly, as the correction applied is the same for b oth RGS and sp ectral orders. A similar result is found if, instead of comparing the average shifts p er sp ectrum, we compare the p ositions of individual lines, as shown if Fig. 3. Shifts are, on average, 1 m° smaller, as it is A also the scatter in the data.

5

Exp ected Up dates

The RGS extensions of this CCF should b e up dated to keep them aligned with the values used for the MOS cameras.
1 We have used an 'ad-hoc' CCF, similar to the public XMM BORESIGHT 0022.CCF, but with two new extensions (RGS1[2] ANGVAR) in which the values corresp onding to MOS1[2] have b een replicated. 2 Release track version xmmsas 20120517 1702-12.0.0-Alpha


XMM-Newton CCF Release

XMM-CCF-REL-290

Page:

4

Figure 2: Comparison of the average sp ectrum shifts with resp ect to lab oratory wavelengths measured in sp ectra processed with fixed and variable b oresight.

Figure 3: Comparison of individual line shifts with resp ect to lab oratory wavelengths measured in data processed with fixed and variable b oresight.


XMM-Newton CCF Release

XMM-CCF-REL-290

Page:

5

6

Test pro cedures

General checks:

· use fv (or another FITS viewer) for file insp ection. It should contain six binary extensions. · use the SAS task cifbuild to see if the CAL digests and creates correctly the calibration index file.

7

Summary of the test results

We have confirmed the applicability of the XMM-Newton variable b oresight to RGS data. The measured change in the wavelength scale is fully compatible with the applied correction. The use of the variable b oresight results in a marginal improvement in the accuracy of the RGS wavelength scale. Line shifts with resp ect to lab oratory wavelengths are now 1 m° smaller. Also, A ° as well. the scatter in the shifts is reduced by approximately 1 mA This improvement is fully consistent with the results obtained for EPIC. The use of the variable b oresight in the EPIC cameras makes the average p osition shift along the spacecraft Z axis to decrease by less than 1 arcsec (2.3 m°), and the width of the distribution by 0.5 arcsec (1.2 m°). A A These values are in good agreement with the change in line p ositions rep orted here.

References
[1] "Astrometry: time-dep endent b oresight", A. Talavera, P. Rodrґguez-Pascual and M. Guainazzi, i XMM-CCF-REL-286, May 2012 http://xmm2.esac.esa.int/docs/documents/CAL-SRN-0286-1-1.p df . [2] "The effect of the variable b oresight on the RGS Wavelength scale", R. Gonzґlez-Riestra, XMMa SOC-TN-0101, June 2012 http://xmm2.esac.esa.int/ xmmdoc/CoCo/CCB/DOC/Attachments/CAL-TN-0101-0-0.p df