Galactic Center Gamma-Ray Excess from a Generic Triaxial Halo

Abstract

Recent studies of Galactic surveys, such as Gaia, have revealed that the Milky Way's gravitational potential comes from a matter distribution that is triaxial and rotated with respect to the Galactic center-Sun axis. This, in turn, could mean that the dark matter halo also shares these properties. In this work, by fitting to the Fermi-LAT gamma-ray observations, we test the compatibility of the morphology of the Galactic Center Excess (GCE) from dark matter annihilation with a triaxial dark matter halo. In particular, we consider both untilted triaxial halos and halos whose principal axes are tilted with respect to the Galactic disk. In our fits of the Fermi-LAT data, by testing over a large library of galactic diffuse emission models, we quantify how the halo triaxiality and tilt affect the line-of-sight-integrated annihilation signal and, consequently, the preferred GCE spatial templates. We find that the GCE spectrum and inner cuspiness are robust against variations in the triaxiality and tilt of the dark matter halo. However, in terms of its overall morphology, the GCE in the gamma-ray data can discriminate between choices for the dark matter halo's triaxiality and tilt. Finally, we find that the GCE is more compatible with originating from a triaxial and tilted halo of dark matter than originating from a triaxial and tilted halo of stars, a result important for understanding the GCE's origin.

Figures (11)

• Figure 1: 3D view of the tilted triaxial halo of BM I (red) and BM II (cyan). $a_i$, $b_i$, $c_i$ ($i=\text{I}, \text{II}$) represent the directions of major, intermediate, and minor axes of the halos, respectively.
• Figure 2: GCE count maps for a spherical dark matter halo (left), tilted halo BM I (middle left), tilted halo BM II (middle right), and the flipped BM I (right). For all models, we assume a cuspiness of $\gamma = 1.2$, take the energy range of $1.02-1.32$ GeV, and mask the disk region with $|b| < 2^\circ$. The white dashed lines show contours of equal count to guide the eye.
• Figure 3: The cuspiness $0.9 \leq \gamma \leq 1.5$, of the GCE morphology for a generalized NFW triaxial dark matter annihilation profile tilted to our line of sight, for the assumptions of Ref. 2022AJ....164..249H, i.e., BM I of Fig. \ref{['fig:example']}. The best-fit GDE models give a preference for $\gamma\simeq 1.2-1.3$. Our $y$-axis gives the difference in the fit between choices for the GCE morphology, defined as $-2 \ln \mathcal{L}$ for the alternative model minus $-2 \ln \mathcal{L}$ of the best fit GCE from the entire sample. Colored lines showcase the results for some of the GDE models that give the best fit, while the gray lines show the GCE under every single GDE model.
• Figure 4: The morphology of the GCE for all 80 GDE models. We compare for an inner slope $\gamma=1.2$ of the dark matter halo profile, the spherical GCE template to those of BM I, BM II and the flipped BM I of Fig. \ref{['fig:example']}. We find that there is a systematic preference for the flipped benchmark I assumptions.
• Figure 5: We assume that the GCE follows the stellar halo profile with cuspiness of $1.3 \leq \gamma_{\textrm{s.h.}} \leq 2.0$. Like in Figure \ref{['fig:GCE_cuspiness']}, we assume that the profile is triaxial and tilted to our line of sight, for the assumptions of Ref. 2022AJ....164..249H. The best-fit results go beyond what is considered the expected range of cuspiness for a stellar halo. Our $y$-axis gives the difference in the fit $-2 \Delta \ln(\mathcal{L})$ between choices for the GCE morphology. Colored lines show the results for some of the best-fit GDE models, while the gray lines show the GCE under every single GDE model.