Overcoming Gamma Ray Constraints with Annihilating Dark Matter in Milky Way Subhalos

TL;DR

This paper explores how dark matter annihilations in nearby subhalos can explain observed cosmic ray excesses while avoiding gamma ray constraints, proposing a particle physics model with Sommerfeld enhancement.

Contribution

It demonstrates that subhalo annihilations can reconcile leptonic signals with gamma ray constraints, introducing a simple leptophilic U(1) model with Sommerfeld enhancement.

Findings

Subhalo annihilations can produce observed lepton excesses without violating gamma ray constraints.

Gamma ray constraints on main halo annihilations are very restrictive, requiring subdominant dark matter components.

A simple particle physics model with a hidden leptophilic U(1) vector boson explains the observations.

Abstract

We reconsider Sommerfeld-enhanced annihilation of dark matter (DM) into leptons to explain PAMELA and Fermi electron and positron observations, in light of possible new effects from substructure. There is strong tension between getting a large enough lepton signal while respecting constraints on the fluxes of associated gamma rays. We first show that these constraints become significantly more stringent than in previous studies when the contributions from background e^+ e^- are taken into account, so much so that even cored DM density profiles are ruled out. We then show how DM annihilations within subhalos can get around these constraints. Specifically, if most of the observed lepton excess comes from annihilations in a nearby (within 1 kpc) subhalo along a line of sight toward the galactic center, it is possible to match both the lepton and gamma ray observations. We demonstrate that this can be achieved in a simple class of particle physics models in which the DM annihilates via a hidden leptophilic U(1) vector boson, with explicitly computed Sommerfeld enhancement factors. Gamma ray constraints on the main halo annihilations (and CMB constraints from the era of decoupling) require the annihilating component of the DM to be subdominant, of order 10^-2 of the total DM density.