Python package to reconstruct masses of galaxy clusters by making use of the self-similar behavior of cluster outskirts (e.g., Zhang et al., 2007; Ghirardini et al., 2018a; Käfer et al., 2019), that is the universality of the shape of the X-ray emission.
Illustrative flow chart to reconstruct the galaxy cluster mass using an MCMC posterior sampling technique with a Poisson likelihood to compare the measured aperture photon counts to the β-model predictions:
Requires:
- Numpy
- Scipy
- Astropy
- emcee
- PyXspec (optional)