Jeans that fit: weighing the mass of the Milky Way analogues in the $\Lambda{\rm CDM}$ universe

TL;DR

This study evaluates the effectiveness of the spherical Jeans equation in estimating the mass of Milky Way-like galaxies within the $ m ext{Lambda CDM}$ framework, revealing inherent biases and limitations due to halo properties.

Contribution

It demonstrates that the spherical Jeans equation can recover galaxy mass profiles with about 12% bias and 14% dispersion, highlighting fundamental limits imposed by halo asphericity and non-equilibrium states.

Findings

Mass recovery bias of ~12% for stellar halos

Bias of ~-2.4% for dark matter halos

Dispersion of ~14% limits accuracy regardless of data quality

Abstract

The spherical Jeans equation is a widely used tool for dynamical study of gravitating systems in astronomy. Here we test its efficacy in robustly weighing the mass of Milky Way analogues, given they need not be in equilibrium or even spherical. Utilizing Milky Way stellar halos simulated in accordance with $\Lambda{\rm CDM}$ cosmology by Bullock and Johnston (2005) and analysing them under the Jeans formalism, we recover the underlying mass distribution of the parent galaxy, within distance $r/{\rm kpc}\in[10,100]$, with a bias of $\sim12\%$ and a dispersion of $\sim14\%$. Additionally, the mass profiles of triaxial dark matter halos taken from the SURFS simulation, within scaled radius $0.2<r/r_{\rm max}<3$, are measured with a bias of $\sim-2.4\%$ and a dispersion of $\sim10\%$. The obtained dispersion is not because of Poisson noise due to small particle numbers as it is twice the later. We interpret the dispersion to be due to the inherent nature of the $\Lambda{\rm CDM}$ halos, for example being aspherical and out-of-equilibrium. Hence the dispersion obtained for stellar halos sets a limit of about $12\%$ (after adjusting for random uncertainty) on the accuracy with which the mass profiles of the Milky Way-like galaxies can be reconstructed using the spherical Jeans equation. This limit is independent of the quantity and quality of the observational data. The reason for a non zero bias is not clear, hence its interpretation is not obvious at this stage.