Assessing Astrophysical Uncertainties in Direct Detection with Galaxy Simulations

TL;DR

This study uses high-resolution galaxy simulations including baryons to analyze local dark matter velocity distributions, revealing significant differences from dark matter-only models and implications for direct detection experiments.

Contribution

It provides the first detailed comparison of dark matter velocity distributions in baryonic versus dark matter-only galaxy simulations and assesses their impact on direct detection experiments.

Findings

Baryons increase high-velocity dark matter particles.

Maxwellian fits overpredict high-velocity tail, affecting detection bounds.

Velocity structures are more complex with baryons, especially in the Earth frame.

Abstract

We study the local dark matter velocity distribution in simulated Milky Way-mass galaxies, generated at high resolution with both dark matter and baryons. We find that the dark matter in the Solar neighborhood is influenced appreciably by the inclusion of baryons, increasing the speed of dark matter particles compared to dark matter-only simulations. The gravitational potential due to the presence of a baryonic disk increases the amount of high velocity dark matter, resulting in velocity distributions which are more similar to the Maxwellian Standard Halo Model than predicted from dark matter-only simulations. Further, the velocity structures present in baryonic simulations possess a greater diversity than expected from dark matter-only simulation. We show the impact on the direct detection experiments LUX, DAMA/Libra, and CoGeNT using our simulated velocity distributions, and explore how resolution and halo mass within the Milky Way's estimated mass range impact the results. A Maxwellian fit to the velocity distribution tends to overpredict the amount of dark matter in the high velocity tail, even with baryons, and thus leads to overly optimistic direct detection bounds on models which are dependent on this region of phase space for an experimental signal. Our work further demonstrates that it is critical to transform simulated velocity distributions to the lab frame of reference, due to the fact that velocity structure in the Solar neighborhood appears when baryons are included. There is more velocity structure present when baryons are included than in dark matter-only simulations. Even when baryons are included, the importance of the velocity structure is not as apparent in the Galactic frame of reference as in the Earth frame.