XMM-Newton Trainee Project 2004

Jenny Carter

I am from Swindon in Great Britain. I attended the University of Bristol for four years and was awarded a first class honours degree in Physics with Astrophysics in July 2002. After travelling for four months in South America, I spent the following year working at the National Physical Laboratory in London within the Optical Radiation Measurement department, primarily concerned with the calibration and maintenance of primary standards for radiometry and photometry. I joined ESA as a YGT in May 2004 at ESAC, Spain.

This document describes one of the projects assigned to me as part of my Young Graduate Trainee placement at ESAC, Spain.

During my year as a YGT, I am working as a member of the Reflection Grating Spectrometry (RGS) team, which monitors and continously calibrates the two Reflection Grating Spectrometers onboard XMM-Newton. One project as part of placement involves the development of a new software task, which will form part of the general XMM-Newton scientific analysis package, SAS.

The project

The motivation behind the project is to be able to look in unprecendented detail at the spectrum of a particular object using an accumulation of data taken for this object of interest. The prescence or absence of an ion species will become more apparent. As examples, and due to the wealth of data already collected by XMM-Newton, I have concentrated the development and testing of the software on three specific objects often used for calibration purposes: HR1099, Capella and AB Doradus.



To be able to look in great detail at a star's spectrum, the addition of all the individual spectra for this one star collected by the telescope could be simply added together, to give an increasingly clear picture of the nature of its spectrum. The problem, however, lies in the way the data are binned in the original SAS process. The RGS works on the principal of diffraction, which is dependent on the angle of incident , α for the incoming photon, and this photon's wavelength is determined by the angle of diffraction, β detected by the RGS instruments.

Diffraction equation: mλ = d(cosβ - cosα)
Where m is the spectral order, λ the wavelength, and d the diffraction grating spacing

Each observation taken by XMM-Newton will have a different incidence angle as the spacecraft positioning will be slightly different each time. This will give rise to different β, and the SAS tasks uses binning in terms of bins in β space to produce the resultant spectrum files. As part of the SAS processing it is possible to ensure that the β binning for different observations are actually aligned and so two spectra can be compared. However, it is essential for further and quantitative analysis, for example in XSpec that the response of the instruments are taken into account. The response matrices are large data sets that use variable length arrays to save precious space. These have to be manipulated to ensure the addition of the correct energy channels to which each response values applies, and, like the spectrum files, the length of the observation has to be taken into account to establish the correct weighting of each observation's influence to the addition as a whole.

The software has first been developed in IDL, for designed and testing, and is currently being applied to C++, within the SAS development environment.

As part of this project, and the ongoing obtainment of skills and development as a YGT, I have attended beneficial courses for using the SAS, and programming techniques in IDL