VERITAS Dark Matter search in dwarf spheroidal galaxies: an extended analysis

TL;DR

This paper presents an extended analysis of VERITAS observations of dwarf spheroidal galaxies to improve dark matter detection sensitivity by considering their spatial extent, compared to traditional point-source methods.

Contribution

It introduces an unbinned maximum likelihood method that incorporates the angular profiles of dSphs, enhancing dark matter search sensitivity with VERITAS data.

Findings

Extended-source analysis improves sensitivity over point-source methods.

The likelihood approach effectively utilizes spatial information of dSphs.

Results set new constraints on dark matter annihilation cross-sections.

Abstract

Dark matter (DM) is largely believed to be the dominant component of the matter content of the Universe. Astronomical measurements can be utilized to search for Standard Model (SM) annihilation or decay products of DM, complementing direct and collider-based searches. Among DM particle candidates, Weakly Interacting Massive Particles (WIMPs) are an attractive one. Their decay or annihilation could produce secondary particles including very-high-energy (VHE: $E>100$ GeV) gamma rays, which could be detected by imaging atmospheric Cherenkov telescopes (IACTs). One of the most favourable target classes for DM searches are dwarf spheroidal galaxies (dSphs), dark matter-dominated objects with a negligible predicted gamma-ray emission due to apparent absence of gas and on-going star formation. IACTs, whose point spread functions (PSFs, defined as 68\% containment radius) are typically $0.1^{\circ}$ at 1 TeV, have the necessary angular resolution to detect extended emission from some dSphs. Thus, an extended-source analysis may give an improvement to DM sensitivity, compared to a point-source analysis. In this work, we used observations made since 2007 to 2013 by VERITAS, an array of four imaging atmospheric Cherenkov telescopes sensitive to VHE gamma rays in the 100 GeV - 30 TeV energy range. We performed an unbinned maximum likelihood estimation incorporating the dSph angular profiles of four dSphs and tested its effectiveness against the traditional spectral analysis.