Application of the Eddington inversion method to constrain the dark matter halo of galaxies using only observed surface brightness profiles

Abstract

Context. The halos of low-mass galaxies may allow us to constrain the nature of dark matter (DM), but the kinematic measurements needed to diagnose the required properties are technically extremely challenging. However, the photometry of these systems is doable. Aims. Using only stellar photometry, we wanted to constrain key properties of the DM halos in low-mass galaxies. Methods. Unphysical pairs of DM gravitational potentials and starlight distributions can be identified if the pair requires a distribution function, f , that is negative somewhere in the phase space. We used the classical Eddington inversion method (EIM) to compute f for a battery of DM gravitational potentials and ∼100 observed low-mass galaxies with M? between 106 and 108 M . The battery includes Navarro, Frenk, and White (NFW) potentials (expected from cold DM) and potentials stemming from cored mass distributions (expected in many alternatives to cold DM). The method assumes spherical symmetry and an isotropic velocity distribution and requires fitting the observed profiles with analytic functions, for which we used polytropes (with zero inner slope, also known as core) and profiles with variable inner and outer slopes. The validity of all these assumptions is analyzed. Results. In general, the polytropes fit the observed starlight profiles well. If they are the correct fits (which could be the case), then all galaxies are inconsistent with NFW-like potentials. Alternatively, when the inner slope is allowed to vary for fitting, between 40% and 70% of the galaxies are consistent with cores in the stellar mass distribution and thus inconsistent with NFW-like potentials. Conclusions. Even though the stellar mass of the observed galaxies is still not low enough to constrain the nature of DM, this work shows the practical feasibility of using the EIM technique to infer DM properties only from photometry.