Документ взят из кэша поисковой машины. Адрес оригинального документа : http://xmm.vilspa.esa.es/docs/documents/CAL-SRN-0281-1-0.pdf Дата изменения: Thu May 31 17:07:28 2012 Дата индексирования: Mon Oct 1 21:11:41 2012 Кодировка:

XMM-Newton CCF Release

Page:

1

XMM-Newton CCF Release Note XMM-CCF-REL-281 Supp ort to the X-ray Loading correction in EPIC-pn Fast Mo des
M. Guainazzi, M. Smith, M. Freyberg, R. Saxton 12 April 2012

1

CCF comp onents
Name of CCF EPN REJECT 0006.CCF VALDATE 2001-01-01 EVALDATE 2020-01-01 Blocks changed XRL2PHA XSCS flag NO

A systematic reanalysis of the offset maps taken prior to each EPIC-pn exposures in "Fast Mo des" (Burst and Timing) unveiled that X-ray Loading (XRL; Smith 2005) affects them almost ubiquitously (Guainazzi et al. 2011). In a nutshell, XRL o ccurs when a source "contaminates" the dark current level measured through the offset maps. When this happens, the energy scale may be significantly distorted. This discovery has potentially a ma jor impact on the calibration of both mo des. The spectra of a sizable fraction of the astrophysical sources used for the on-flight energy calibration of EPIC-pn Fast Mo des may have been affected by XRL, i.e. by an uncalibrated distortion in their energy scale proportional to their count rate. This phenomenon offers a natural physical explanations to rate-dependent energy calibration effects such as the "Rate-Dependent CTI" (Guainazzi et al. 2009). Hence it stems the importance of a proper XRL correction prior to any calibration of the energy scale. The CCF update described in this Release Note embeds the XRL correction. In general terms, one should expect that: (P H A) L, where (P H A) is the (in principle energyindependent) shift yielded by the XRL level L (expressed "ADU units"; for the EPIC-pn: 1 ADU = 5 eV). Existing calibration data based on a recent (September 2011) observation of the Crab Nebula are consistent with this assumption, at least for EPIC-pn Timing Mo de. The SAS assumes for the time being a linear correction.


XMM-Newton CCF Release

Page:

2

2

Changes

A new extension has been added to the EPN REJECT CCF constituent: XRL2PHA. It contains the co efficients of the linear relation: (P H A) = a + b в L. for the EPIC-pn Fast Mo des. This change does not affect EPIC-pn imaging mo des. In this version: a = 0 and b = 0 for both Burst and Timing Mo de. As of Version 12, these co efficients are used by the SAS task epreject to event-by-event basis the event PHA value for the XRL measured in the RAWX the event is detected. Readers are referred to the task description for a description of the XRL correction. The values in this CCF version ensure, no correction is applied even if the task is run on an event list. correct on an column where more detailed however, that

3

Scientific Impact of this Up date

This CCF aims at ensuring that SASv12 supports the algorithm for the XRL correction. However, this correction cannot be delivered to the public yet, because its should be accompanied by other calibrations, whose content depends on a proper XRL correction. Nullifying the XRL correction for any value of L is obtained if a = 0 and b = 0, i.e. if the event PHA values remain untouched independently of the level of XRL measured.

4

Estimated Scientific Quality

The release of this CCF is a preliminary step to achieve the full recalibration of EPICpn Fast Mo des. Readers are referred to Guainazzi et al. (2011) for a more detailed description of the basic scientific ob jectives of this exercise.

5

Test pro cedures and results

The content, formal integrity and compliance to FITS standards of the new CCF have been checked with fverify and fv. We have applied epreject with EPN REJECT 0006.CCF and the EPN TICLOSEDODI *.CCF (Smith et al., 2012) to the event list of one of the most recent Crab observations in EPIC-pn Timing Mo de (Obs.#0611181101, offset map in THICK filter). Spectra extracted from event lists to which the XRL correction was not applied were indistinguishable from spectra extracted under the same conditions from event lists on which epreject was run, mo dulo a correction for offset shifts due to Minimum Ionising Particles (MIPs) during


XMM-Newton CCF Release

Page:

3

offset calculation, as expected. For further details on the SASv12 epreject validation results, including non-trivial cases of XRL correction under different assumptions, readers are referred to the SASv12 Scientific Validation Report (Gabriel et al., in preparation).

6

References

Guainazzi et al., 2009, XMM-CCF-REL-245 (available at: http://xmm2.esac.esa.int/docs/documents/CAL-SRN-0245-1-0.ps.gz) Guainazzi et al., 2011, XMM-SOC-CAL-TN-0083 (available at: http://xmm2.esac.esa.int/docs/documents/CAL-TN-0083.pdf) Smith M., 2005, XMM-SOC-CAL-TN-0050 (available at: http://xmm2.esac.esa.int/docs/documents/CAL-TN-0050-1-0.ps.gz) Smith M. et al., 2012, XMM-CCF-REL-282 (available at: http://xmm2.esac.esa.int/docs/documents/CAL-SRN-0282-1-0.ps.gz)