New limits on the dark matter lifetime from dwarf spheroidal galaxies using Fermi-LAT

Dwarf spheroidal galaxies (dSphs) are promising targets for the indirect detection of dark matter through gamma-ray emission due to their proximity, lack of astrophysical backgrounds and high dark matter density. They are often used to place restrictive bounds on the dark matter annihilation cross section. In this paper, we analyze six years of Fermi-LAT gamma-ray data from 19 dSphs that are satellites of the Milky Way, and derive from a stacked analysis of 15 dSphs, robust 95% confidence level lower limits on the dark matter lifetime for several decay channels and dark matter masses between $\sim 1\mathrm{GeV}$ and 10 TeV. Our findings are based on a bin-by-bin maximum likelihood analysis treating the $J$-factor as a nuisance parameter using the Pass 8 event class. Our constraints from this ensemble are among the most stringent and solid in the literature, and competitive with existing ones coming from the extragalactic gamma-ray background, galaxy clusters, AMS-02 cosmic ray data, Super-K and ICECUBE neutrino data, while rather insensitive to systematic uncertainties. In particular, among gamma-ray searches, we improve existing limits for dark matter decaying into $\overline{b}b$ (${\mu }^{+}{\mu }^{-}$) for dark matter masses below $\sim 30(200)\mathrm{GeV}$, demonstrating that dSphs are compelling targets for constraining dark matter decay lifetimes.

Figure 1

Individual lower limits, at 95% C.L., on the dark matter lifetime from the most constraining dSphs for the $bb$ (left panel) and $\tau \tau$ (right panel) final state. We explicitly derived individual limits from all 19 dSphs and decided to plot the ones which yield the most restrictive bounds for clarity. It is clear that Draco, Ursa Minor and Ursa Major II provide the best limits and are in the ballpark of $10^{26–27}\mathrm{s}$.

Figure 2

Stacked analysis, 95% C.L., bound on DM lifetime for several decay channels, encompassing both fermionic and bosonic DM. Refer to text for the list of dSphs used in the stacked analysis. The decay into $qq$ takes into account all light quarks. For heavy DM, decays into $\overline{b}b$, $hh$ and $qq$ provide the strongest limits, whereas for relatively light dark matter $\overline{b}b$, $qq$ and $\tau \tau$ are dominant. As we shall see further in Fig. 3 these are the most stringent limits in the literature for DM from gamma-ray searches masses below $\sim 30(200)\mathrm{GeV}$ for decays into $bb$ (${\mu }^{+}{\mu }^{-}$).

Figure 3

We compare our results with existing gamma-ray observations, namely bounds from extragalactic gamma-ray background from [4] (dashed) and bounds from galaxy clusters from [9] (dotted), using an optimized ROI strategy [17] (solid gray) employing Fermi-LAT data. See text for detail. Left: limits for dark matter decay into $\overline{b}b$. Right: limits for dark matter decay into ${\mu }^{+}{\mu }^{-}$. We conclude that our limits are the strongest for dark matter masses below $\sim 30$ and $\sim 200\mathrm{GeV}$ for the $bb$ and ${\mu }^{+}{\mu }^{-}$ decay channels respectively.

Figure 4

95% C.L bound on DM lifetime for $b\overline{b}$ and ${\tau }^{+}{\tau }^{-}$ channels using Reticulum II data only. Notice the anomalous behavior for $b\overline{b}$, which provides a limit still weaker than our combined one. As for the other dSphs, the shapes of the limit curves result from a combination of both the energy spectrum of the final state and the upper flux limit reported in [43].