Speaker
Description
Measurement of the intracluster medium (ICM) in X-ray is being carried out with ever increasing influx of data from telescopes like Chandra, eROSITA, XRISM, XMM-Newton. The data we obtain from these telescopes is generally incomplete, uncertain (noisy) and convolved with complex instrument responses. Reconstructing multi-domain (spatial, spectral and temperature for example) images of the multiphase ICM from the data we obtain becomes a complex high dimensional, non-linear inverse problem. In this study, we use information field theory (IFT), which provides a robust mathematical framework to build a Bayesian signal inference algorithm that identifies lines on top of continuum emission. Numerical Information Field Theory (NIFTy), a python package is used to implement the signal inference algorithm.
Our developed method is used for reconstructing the multiphase ICM emission from the central regions of the Perseus cluster. The main algorithmic challenge, to split the total emission into continuum and line emission components, is dealt with a phenomenological modelling approach. For solving the degeneracy between the two components, we develop a method which first fits a purely continuum flux model, and then detects, guesses and initialises line emission components therein and automatically adds them to the total flux model. The proposed method identifies and adds 7 phenomenological lines which correspond to emission from Fe-K, Fe-L/Ne, Si, S, Mg and Ar.
