[Paper Review] Legacy Analysis of Dark Matter Annihilation from the Milky Way Dwarf Spheroidal Galaxies with 14 Years of Fermi-LAT Data

Alex McDaniel, M. Ajello+4 more|arXiv (Cornell University)|Nov 8, 2023

Dark Matter and Cosmic Phenomena4 citations

TL;DR

This study performs a legacy analysis of dark matter annihilation using 14 years of Fermi-LAT data from Milky Way dwarf spheroidal galaxies (dSphs), employing updated J-factor estimates and an expanded dSph sample. Although no significant individual dSph detection is found, a marginal local excess of $2-3\sigma$ is observed at $M_\chi \approx 150-230$ GeV ($b\bar{b}$) and $30-50$ GeV ($\tau^+\tau^-$), with global significance near $0.5\sigma$. Updated constraints on the annihilation cross-section remain consistent with prior results, and projected sensitivity suggests a $4\sigma$ signal could be achieved with ~10 more years of observation if the excess is real.

ABSTRACT

The Milky Way (MW) dwarf spheroidal satellite galaxies (dSphs) are particularly intriguing targets to search for gamma rays from Weakly Interacting Massive Particle (WIMP) dark matter (DM) annihilation or decay. They are nearby, DM-dominated, and lack significant emission from standard astrophysical processes. Previous studies using the Fermi-Large Area Telescope (LAT) of DM emission from dSphs have provided the most robust and stringent constraints on the DM annihilation cross section and mass. We report an analysis of the MW dSphs using over 14 years of LAT data and an updated census of dSphs and $J$-factors. While no individual dSphs are significantly detected, we find slight excesses with respect to background at the $\gtrsim 2\,σ$ local significance level in both tested annihilation channels ($b\bar{b}$, $τ^+τ^-$) for 7 dSphs. We do not find a significant DM signal from a combined likelihood analysis of the dSphs ($s_{global}\sim 0.5σ$), yet a marginal local excess relative to background at a $2-3\,σ$ level is observed at a DM mass of $M_χ=150-230$ GeV ($M_χ=30-50$ GeV) for annihilation into $b\bar{b}$ ($τ^+τ^-$). Given the lack of a significant detection, we place updated constraints on the $b\bar{b}$ and $τ^+τ^-$ annihilation channels that are generally consistent with previous recent results. As in past studies, tension is found with some WIMP DM interpretations of the Galactic Center Excess (GCE), though the limits are consistent with other interpretations given the uncertainties of the Galactic DM density profile and GCE systematics. Based on conservative assumptions of improved sensitivity with increased LAT exposure and moderate increases in the sample of dSphs, we project the local $\sim 2\,σ$ signal, if real, could approach the $\sim 4\,σ$ local confidence level with additional $\sim 10$ years of observation.

Motivation & Objective

• To improve constraints on weakly interacting massive particle (WIMP) dark matter annihilation using the most up-to-date Fermi-LAT data and dSph sample.
• To assess whether marginal excesses observed in previous analyses persist with 14 years of data and refined J-factor estimates.
• To evaluate the impact of diffuse background systematics on low-mass signal significance using a weighted likelihood method.
• To project future sensitivity gains for detecting a potential dark matter signal based on extended observation time.
• To test the compatibility of dSph constraints with interpretations of the Galactic Center Excess as arising from dark matter annihilation.

Proposed method

• Analysis of 14 years of Fermi-LAT data from Milky Way dSphs using a likelihood framework to search for gamma-ray signals from dark matter annihilation.
• Incorporation of updated J-factor estimates based on improved line-of-sight integration of dark matter density squared ($\rho_\chi^2$) along the line of sight.
• Application of a weighted log-likelihood method to mitigate systematic uncertainties from the diffuse background model, particularly at low energies and low DM masses.
• Comparison of standard and weighted likelihood approaches to assess the robustness of signal significance and background fluctuations.
• Global and local significance evaluation across multiple dSph samples and annihilation channels ($b\bar{b}$, $\tau^+\tau^-$) using test statistics (TS) and p-values.
• Projection of future sensitivity based on conservative assumptions of exposure time and sample size increases.

Figure 1: $J$ -factors and uncertainties for all 50 dSphs considered in this work. Marker shapes and colors indicate whether the $J$ -factor is determined by measurement (blue circles), the kinematic scaling relation of Equation 3 (green squares), or the photometric scaling relation of Equation 4 (o

Experimental results

Research questions

• RQ1 Does the marginal $2-3\sigma$ excess observed in previous dSph analyses persist with 14 years of Fermi-LAT data and updated J-factors?
• RQ2 To what extent are low-mass signal peaks driven by systematic uncertainties in the diffuse background model?
• RQ3 How do the updated constraints on the dark matter annihilation cross-section compare with previous results and with the Galactic Center Excess interpretation?
• RQ4 What exposure time is required to elevate a marginal local excess to a $4\sigma$ significance level, assuming it is a real signal?
• RQ5 How do the results from the weighted likelihood method affect the significance and upper limits compared to the standard likelihood approach?

Key findings

• A marginal local excess of $2-3\sigma$ is observed at $M_\chi \approx 150-230$ GeV for $b\bar{b}$ annihilation and $30-50$ GeV for $\tau^+\tau^-$ channels, though the global significance remains low at $\sim 0.5\sigma$.
• The weighted likelihood method reduces low-mass signal peaks, indicating that some of the TS enhancement at low masses is attributable to diffuse background systematics.
• Updated upper limits on the annihilation cross-section for $b\bar{b}$ and $\tau^+\tau^-$ channels are consistent with previous recent results, showing no significant improvement.
• The global significance for the benchmark sample improves slightly from $0.5\sigma$ to $0.7\sigma$ under the weighted likelihood method, but this does not alter the upper limits.
• The analysis finds tension with WIMP interpretations of the Galactic Center Excess, though this is within uncertainties of the Galactic dark matter density profile and GCE systematics.
• With an additional ~10 years of observation, the current $\sim 2\sigma$ local excess could reach $\sim 4\sigma$ significance under conservative assumptions of improved sensitivity and sample size.

Figure 2: TS distribution for combined blank fields corresponding to each of our sub samples. For reference, we show the distribution of $\chi^{2}/2$ with two degrees of freedom, as well as vertical lines comparing the TS value corresponding to $3\,\sigma$ for the Chernoff’s theorem approach (red-da