The EPIC-MOS particle-induced background spectra

Abstract

Context. To analyze diffuse emission that fills the field of view, one must accurately characterize the instrumental backgrounds. For the XMM-Newton EPIC-MOS instrument these backgrounds include a temporally variable "quiescent" component, as well as the strongly variable soft proton contamination. Aims. We have characterized the spectral and spatial response of the EPIC-MOS detectors to these background components and have developed tools to remove these backgrounds from observations. Methods. The "quiescent" component was characterized using a combination of the filter-wheel-closed data and a database of unexposed-region data. The soft proton contamination was characterized by differencing images and spectra taken during flared and flare-free intervals. Results. After application of our modeled backgrounds, the differences between independent observations of the same region of blank sky are consistent with the statistical uncertainties except when there is clear spectral evidence of solar wind charge exchange emission. Using a large sample of blank sky data, we show that strong magnetospheric SWCX emission requires elevated solar wind fluxes; observations through the densest part of the magnetosheath are not necessarily strongly contaminated with SWCX emission.Key words. methods: data analysis -instrumentation: detectors -X-rays: general MotivationAlthough XMM-Newton has a large field of view (FOV) compared to other current X-ray observatories, there are often observations in which the object of interest (such as diffuse Galactic emission, nearby galaxies, or a large cluster of galaxies) fills the entire FOV. Determining the background in such a situation can be problematic. The non-cosmic background is due primarily to energetic particles interacting directly with the detector, or interacting with material around the detector and producing fluorescent X-rays that then strike the detector. This "particleinduced background" has multiple components and each component is temporally variable, although on different scales. Since the particle background is temporally variable, using the particle background derived from another observation is likely to be unsatisfactory."Blank sky" observations are often used to subtract a spectrum combining the particle background and the Galactic emission to allow the measurement of the spectrum of clusters of galaxies. Given that the Galactic background varies strongly with Galactic coordinate, use of blank sky data to remove the particle background and Galactic foreground emission in order to study some other object requires evaluation of the suitability of each blank sky field for its appropriateness based on two criteria: the Galactic spectrum and the particle background spectrum. This method works well at E > 2 keV where neither the bulk of the Galaxy (away from the Galactic center and plane) nor the particle background varies greatly, but is much more Based on observations obtained with XMM-Newton, an ESA science mission with instruments and contributions directly fun...

A systematic analysis of the XMM-Newton background: IV

“…For the chips MOS1-4, MOS1-5, MOS2-2 and MOS2-5 anomalous states have been identified, characterized by a significant increase of the count rates below 1 keV. These anomalous states can be recognized by performing an analysis of the corner data in a hardness-ratio-count rate diagnostic diagram [6].…”

“…The knowledge of these components has been growing thanks to the many efforts involved in collecting suitable blank sky fields to be used as template background by the XMM-Newton users [4,5], the analysis of the XMM-Newton Guest Observer Facility leading to the XMMNewton Extended Source Analysis Software [6,7], the efforts of the XMMNewton SOC 1 and the contributions of various research teams, among them our in Milan has been particularly active on this topic [8,9,10]. A summary table of the EPIC instrumental background components is available at this link 2 .…”

“…The former is obtained by fitting the histogram of the counts with a Gaussian and excluding time periods above 3σ from the mean, in a similar fashion as filtering soft proton flares in the light curves of EPIC observations. The latter has been obtained by using the SEP events list found on the ESA Solar Energetic Particle Environment Modelling (SEPEM) application server 6 . The time duration of the SEP event in the list is usually conservative, even though sometime this is not true and leads to low residual level of outliers (see an example in Figure 5).…”

Modeling the magnetospheric X‐ray emission from solar wind charge exchange with verification from XMM‐Newton observations

“…The results of each comparison show very minor differences in shape and magnitude, certainly not enough for the Hodges model to be the cause of the variations between the modeled and observed light curves. It should also be noted that when comparing SWCX through different parts of the magnetosheath, Kuntz and Snowden [2008] demonstrate that the solar wind flux is a more important factor than the magnetosheath density along the line of sight. This indicates that small differences in the Hodges number density are unlikely to make any significant differences.…”

“…To make the conversion from one instrument to the other requires the assumption of a spectral model. Formally, the diffuse X-ray background spectrum is well represented by a two-component thermal Astrophysical Plasma Emission Code model for the Galactic emission and a power law for the unresolved power law [Kuntz and Snowden, 2008]. We have used the spectral parameters from a deep analysis of case 01-3 previously studied by Carter et al [2010] modified by the appropriate absorption in the light of sight which is provided by the HEASARC toolkit and derived from the Leiden/Argentine/Bohn neutral hydrogen survey [Kalberla et al, 2005].…”