Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://xmm.vilspa.esa.es/docs/documents/CAL-SRN-0297-1-0.pdf Äàòà èçìåíåíèÿ: Wed May 8 18:05:08 2013 Äàòà èíäåêñèðîâàíèÿ: Mon Mar 3 13:09:24 2014 Êîäèðîâêà:

XMM-Newton CCF Release Note
XMM-CCF-REL-297 Sun Angle correction to the RGS Wavelength scale
R. Gonz´lez-Riestra a April 17, 2013

1

CCF comp onents
Name of CCF RGS1 SAACORR 0001.CCF RGS2 SAACORR 0001.CCF VALDATE 2000-01-01T00:00:00 2000-01-01T00:00:00 EVALDATE ­ ­ Blocks changed SAACORR SAACORR XSCS flag NO NO

2

Intro duction

Studies of RGS sp ectra of emission line sources have shown that line p ositions are systematically shifted with resp ect to lab oratory wavelengths, and that the wavelength scales of b oth RGS are displaced by a few m°. These works have also established that wavelengths measured in RGS A sp ectra are accurate to 7 m° in first order and to 5 m° in second order (Lorente et al. 2003). A A Along the last years, several studies have b een carried out to clarify the origin of these systematic effects (see e.g. Coia and Pollock 2007). Gonz´lez-Riestra (2008) found a strong correlation b etween the angular distance b etween the a spacecraft p ointing direction and the Sun, hereafter "Solar Angle"1 . This correlation was confirmed indep endently by Kaastra et al. (2011) in their analysis of the RGS sp ectra of Mrk 509. This dep endence was ment and sp ectral order. of the RGS instruments resulting the values show parametrised as a linear relation, with different coefficients for each instruOnce the Variable Boresight was implemented in the reduction of the data the coefficients of this relation were re-derived (Gonz´lez-Riestra, 2012), a n in Table 1.

The Sun Angle correction to the RGS wavelength scale has b een implemented for the first time in SAS13.0, as a non-default processing option. In this first implementation, the same linear relation
1

This angle is equivalent to the Fine Sun Sensor Pitch Angle + 90 degrees. Op erational range is 70-110 degrees.

1


XMM-Newton CCF Release

XMM-CCF-REL-297

Page:

2

Table 1: RGS Wavelength shifts: Fits to a b RGS1 o1 1.2±0.2 -0.52±0.02 RGS2 o1 6.1±0.2 -0.50±0.02 RGS1 o2 1.8±0.2 -0.29±0.02 RGS2 o2 2.7±0.2 -0.33±0.02 ° correction (in mA) = a + b â (SA - 90) Res: residuals of the fit in m°, errors are standard deviations. A

Solar Angle Res 0±5 1±5 0±3 0±2

Table 2: RGS Wavelength shifts Average shifts p er sp ectrum (in m°) A SAS12 SAS13 RGS1 o1 2±6 1±5 RGS2 o1 8±6 1±5 RGS1 o2 2±3 1±3 RGS2 o2 3±4 1±3 order 1 - order 2 RGS 1 1±4 0±4 RGS 2 4±4 -1±3 RGS 1 - RGS 2 order 1 -5±2 1±3 order 2 -2±2 1±2 Shifts of individual lines (in m°) A SAS12 SAS13 RGS1 o1 2±7 1±6 RGS2 o1 7±7 1±6 RGS1 o2 1±5 1±5 RGS2 o2 3±4 0±6

is applied to first and second order data (but different for each instrument). The error introduced by this approximation has b een estimated to b e less than 2 m° for second order data taken at extreme A values of Solar Angle.

3

Scientific Impact of this Up date
is correction, the scatter in the line shifts is reduced, and the wavelength scale and orders is aligned. This is shown in Table 2 and Fig. 1, that compare the sample of 60 sp ectra of the four wavelength calibrators (Cap ella, AB Dor, ) processed with SAS12 and with SAS13+Sun Angle correction.

After application of th of b oth sp ectrographs wavelength shifts of a HR 1099 and Procyon


XMM-Newton CCF Release

XMM-CCF-REL-297

Page:

3

Figure 1: Comparison of the wavelength shifts in sp ectra processed with SAS12 and with SAS13 applying the Sun Angle correction


XMM-Newton CCF Release

XMM-CCF-REL-297

Page:

4

Figure 2: Sp ectra of Cap ella taken at Sun Angles 73 and 108, processed with SAS12 and with SAS13 applying the Sun Angle correction.

4

Estimated Scientific Quality

We have compared two observations of Cap ella taken at extreme Solar Angle values (108 and 73) processed with SAS12 and SAS13+Sun Angle correction. Sp ectra of b oth instruments and orders are shown in Fig. 2. Shift b etween b oth sp ectra (in m°) A SAS12 SAS13 RGS1 o1 22 2 RGS2 o1 23 5 RGS1 o2 12 3 RGS2 o2 15 6 As exp ected, the application of the Sun Angle correction improves substantially the agreement ° b etween the wavelengths measured in b oth datasets: from approx. 22 mA to approx. 3 m° for first A ° to 4 m° for second order. A small, intrinsic, shift due to orbital motion A order, and from 13 mA cannot b e excluded.


XMM-Newton CCF Release

XMM-CCF-REL-297

Page:

5

5

Exp ected Up dates

The relation b etwwen line shifts and Solar Angle will b e revised regularly, as new data of the wavelength calibrators b ecome available. The coefficients will b e up dated as needed.

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.

References
Coia, D. and Pollock, A., "A survey of lines in coronal sources for the RGS wavelength scale", Decemb er 2007 SOC-CAL-TN-0079-1-0 Gonz´lez-Riestra,R., "Systematic Effects in the RGS Wavelength scale", Septemb er 2008 a SOC-CAL-TN-0082-1-0 Gonz´lez-Riestra, R., "The effect of the variable b oresight on the RGS Wavelength scale", June a 2012 SOC-CAL-TN-0101-0-0 Kaastra, J. et al.," Multiwavelength campaign on Mrk 509. I I. Analysis of high-quality Reflection Grating Sp ectrometer sp ectra", 2011 A&A 534, A37 Lorente, R. et al.,"The RGS Wavelength Scale", July 2003 SOC-CAL-TN-0041-1-0