A distribution-function estimate of the Milky Way mass profile from outer halo stars in DESI DR2

Abstract

We present measurements of the Milky Way mass using blue horizontal-branch (BHB) and RR Lyrae (RRL) stars from the second data release of DESI (DR2), covering its first three years of operations. We employ 279 BHBs and 213 RRLs at $\sim\!50$--$80$\,kpc from the Galactic center, selected from catalogs roughly twice as large as those of DESI DR1, used in our previous mass measurement. We apply the Galactic Mass Estimator (GME), a hierarchical Bayesian framework that assumes power-law forms for the potential and the tracer density, and accounts for distance and velocity uncertainties. Validating the method on 24 AuriDESI mock catalogs (six Auriga halos at four viewing angles), we find that it overestimates the enclosed mass by $\sim\!15\%$ at 80\,kpc, and correct our results accordingly. We obtain $M(<80\,{\rm kpc}) = 0.64^{+0.06}_{-0.06}$ and $0.53^{+0.05}_{-0.05} \times 10^{12}\,M_\odot$ for BHBs and RRLs, respectively, while including Sagittarius stream stars lowers these masses by $\sim\!10$\%. Extrapolating to 100\,kpc, we obtain $M(<100\,{\rm kpc}) = 0.74^{+0.07}_{-0.07}$ and $0.62^{+0.06}_{-0.07} \times 10^{12}\,M_\odot$, and anchoring an NFW halo to our GME constraints at 65\,kpc yields virial masses of $M_{200} = 1.12^{+0.14}_{-0.12}$ and $0.89^{+0.11}_{-0.10} \times 10^{12}\,M_\odot$. BHBs yield systematically higher masses than RRLs across the full radial range, reflecting the impact of tracer choice even within a single method. This work is part of a wider DESI DR2 Key Project effort to measure the Milky Way mass and constrain systematics arising from different samples and methodologies.