About hydromass

hydromass is a Python package for the analysis of galaxy cluster mass profiles from X-ray and/or Sunyaev-Zel’dovich observations. The code builds upon several decades of development and tens of scientific papers. The code was originally developed as an IDL script by Stefano Ettori (see Ettori et al. 2010 ) and translated into Python by Vittorio Ghirardini during his PhD. The new Python code was completely revised by Dominique Eckert in the framework of the X-COP gravitational field project and turned into a general framework for the reconstruction of galaxy cluster mass and thermodynamic profiles. The code is released together with a series of two papers describing it in detail and applying it to the X-COP galaxy cluster sample.

Motivation

The X-COP collaboration is committed to delivering high-level data products and advanced analysis tools to allow for an easy replication of our results and extension of our work to a wider range of applications than can be pursued only with our limited manpower. The hydromass package is an important part of this philosophy, as it will allow the user to easily load the public X-COP data products and apply our reconstruction tools directly within a Jupyter notebook. The framework will later be extended to the use of new data such as the CHEX-MATE Heritage program on XMM-Newton and eventually to the data of upcoming missions (e.g. eROSITA, ATHENA).

Features

  • Joint modeling of X-ray surface brightness, X-ray spectroscopic temperature, and SZ pressure

  • A global framework for mass modeling, deprojection and PSF deconvolution of thermodynamic gas profiles

  • Efficient Bayesian optimization based on Hamiltonian Monte Carlo using PyMC3

  • Parametric mass model reconstruction including Navarro-Frenk-White, Einasto and several other popular mass models, with automatic or custom priors

  • Decomposition of the hydrostatic mass profile into baryonic and dark matter components

  • Non-parametric temperature deprojection and hydrostatic mass reconstruction using a log-normal mixture model

  • Parametric forward model fitting and effective polytropic reconstruction

  • Non-thermal pressure modeling and marginalization

  • Diagnostic tools to investigate goodness-of-fit through posterior predictive checks and WAIC

  • Easy visualization of the output mass and thermodynamic profiles

  • Saving/reloading options

The current implementation has been developed in Python 3 and tested on Python 3.6+ under Linux and Mac OS.

Citation

If you find the software useful, please cite Eckert et al. 2022 . The associated paper also provides extensive validation tests of the provided software.