Diurnal modulation signal from dissipative hidden sector dark matter

TL;DR

The work investigates a two-component hidden sector DM model with an unbroken $U(1)'$ gauge symmetry and kinetic mixing to the Standard Model that provides a heating mechanism for dissipative halos, requiring $\epsilon \sim 10^{-9}$. It shows that DM captured within the Earth can shield detectors from the halo wind, producing a diurnal modulation that is especially pronounced for Southern-hemisphere sites, and derives a shielding radius $R_s$ with a parameterized dependence on $m_{F_2}$, $\alpha'$, and $|Z'|$. The authors quantify the daily rate suppression ${\cal R}(t)$ and find a large modulation across broad parameter space, with $R_{\max}$ typically exceeding $10\%$ and described by analytic scalings for specific sites such as Stawell and Andes Lab. These results yield a concrete, testable signature of self-interacting dissipative DM in upcoming or existing Southern-hemisphere direct-detection experiments, linking astrophysical halo dynamics to terrestrial measurements.

Abstract

We consider a simple generic dissipative dark matter model: a hidden sector featuring two dark matter particles charged under an unbroken $U(1)'$ interaction. Previous work has shown that such a model has the potential to explain dark matter phenomena on both large and small scales. In this framework, the dark matter halo in spiral galaxies features nontrivial dynamics, with the halo energy loss due to dissipative interactions balanced by a heat source. Ordinary supernovae can potentially supply this heat provided kinetic mixing interaction exists with strength $ε\sim 10^{-9}$. This type of kinetically mixed dark matter can be probed in direct detection experiments. Importantly, this self-interacting dark matter can be captured within the Earth and shield a dark matter detector from the halo wind, giving rise to a diurnal modulation effect. We estimate the size of this effect for detectors located in the Southern hemisphere, and find that the modulation is large ($\gtrsim 10\%$) for a wide range of parameters.