Astrophysical and dark matter interpretations of extended gamma-ray emission from the Galactic Center

We construct empirical models of the diffuse gamma-ray background toward the Galactic Center. Including all known point sources and a template of emission associated with interactions of cosmic rays with molecular gas, we show that the extended emission observed previously in the Fermi Large Area Telescope data toward the Galactic Center is detected at high significance for all permutations of the diffuse model components. However, we find that the fluxes and spectra of the sources in our model change significantly depending on the background model. In particular, the spectrum of the central Sgr A* source is less steep than in previous works and the recovered spectrum of the extended emission has large systematic uncertainties, especially at lower energies. If the extended emission is interpreted to be due to dark matter annihilation, we find annihilation into pure $b$-quark and $\tau$-lepton channels to be statistically equivalent goodness of fits. In the case of the pure $b$-quark channel, we find a dark matter mass of $39.4(\genfrac{}{}{0ex}{}{+3.7}{-2.9}\text{stat})(\pm 7.9\mathrm{sys})\mathrm{GeV}$, while a pure ${\tau }^{+}{\tau }^{-}$-channel case has an estimated dark matter mass of $9.43(\genfrac{}{}{0ex}{}{+0.63}{-0.52}\text{stat})(\pm 1.2\mathrm{sys})\mathrm{GeV}$. Alternatively, if the extended emission is interpreted to be astrophysical in origin such as due to unresolved millisecond pulsars, we obtain strong bounds on dark matter annihilation, although systematic uncertainties due to the dependence on the background models are significant.

Figure 1

For the full model, $2\mathrm{FGL}+2\mathrm{PS}+\mathrm{I}+\mathrm{MG}+\mathrm{ND}+\mathrm{GCE}$ (see text and Table 1), we show here the multicomponent diffuse model (the combined $\mathrm{I}+\mathrm{MG}+\mathrm{ND}$) residuals, i.e., the counts subtracting all model components other than the $\mathrm{I}+\mathrm{MG}+\mathrm{ND}$ components (top row), the multicomponent diffuse model, $\mathrm{I}+\mathrm{MG}+\mathrm{ND}$, (middle row), and the GCE source residuals within our ROI (bottom row), at $|b|<3.5°$ (vertical axis) and $|\ell |<3.5°$ (horizontal axis). The maps are shown on the same color scale to show the components’ relative strength for the counts per pixel, Gaussian filtered spatially with $\sigma =0.3°$.

Figure 2

For the full model, $2\mathrm{FGL}+2\mathrm{PS}+\mathrm{I}+\mathrm{MG}+\mathrm{ND}+\mathrm{GCE}$ (see text and Table 1), we show the full model residuals after including all diffuse components, in units of $\sigma$. Here, $|b|<3.5°$ (vertical axis) and $|\ell |<3.5^{\circ }$ (horizontal axis). The residual count map was Gaussian filtered spatially with $\sigma =0.3°$. The 20 point sources modeled simultaneously with the diffuse and extended sources in the ROI are shown as circles.

Figure 3

Shown are two cases of our determination of the Sgr A* source spectrum. The $2\mathrm{FGL}+2\mathrm{PS}+\mathrm{I}$ binned spectrum is in pink circles, with best fit binned log-parabola spectrum in pink. The full model $2\mathrm{FGL}+2\mathrm{PS}+\mathrm{I}+\mathrm{MG}+\mathrm{ND}+\mathrm{GCE}$ spectrum is in blue squares, with best fit binned log-parabola spectrum in blue. The presence of GCE associated photons at 1 to 3 GeV in the Sgr A* spectrum is evident in the case of the $2\mathrm{FGL}+2\mathrm{PS}+\mathrm{I}$ modeling. The errors shown are solely the Poisson errors within the energy band and do not reflect covariances or systematic uncertainties.

Figure 4

Shown are two cases of our determination of the GCE source spectrum. The $2\mathrm{FGL}+2\mathrm{PS}+\mathrm{I}+\mathrm{GCE}$ binned spectrum is in pink circles, with best fit binned log-parabola spectrum in pink. The full model $2\mathrm{FGL}+2\mathrm{PS}+\mathrm{I}+\mathrm{MG}+\mathrm{ND}+\mathrm{GCE}$ spectrum is in blue squares, with best fit binned log-parabola spectrum also in blue. We also show the spectrum using FRONT converting only photons in green stars, with its best fit binned log-parabola spectrum in green. The errors shown are solely the Poisson errors within the energy band and do not reflect covariances or systematic uncertainties.

Figure 5

Here we show the SED of the Sgr A* source for the full model, $2\mathrm{FGL}+2\mathrm{PS}+\mathrm{I}+\mathrm{MG}+\mathrm{ND}+\mathrm{GCE}$ (blue squares), as well as its best fit log-parabola spectrum (solid line). For comparison, we show the Sgr A* spectrum determined by Chernyakova et al. [17] (gray circles) and the 3 pc diffusion emission model from Linden et al. [18] (dashed line). The errors represent the SED-normalization statistical uncertainty within an energy band.

Figure 6

Here we show the spectrum for the MG and ND components for the $2\mathrm{FGL}+2\mathrm{PS}+\mathrm{I}+\mathrm{MG}+\mathrm{ND}+\mathrm{GCE}$ model. The MG spectrum is in pink circles, with the best fit log-parabola spectrum in pink. The ND spectrum is in orange triangles, with the best fit log-parabola spectrum in orange. For reference, we show the best fit GCE spectrum for the same full model, which shows how the GCE is detected at above $\sim 2\mathrm{GeV}$. The errors shown are solely the Poisson errors within the energy band and do not reflect covariances or systematic uncertainties.

Figure 7

Here we show the residual flux for the GCE for different spatial regions within the ROI for the $2\mathrm{FGL}+2\mathrm{PS}+\mathrm{MG}+\mathrm{GCE}$ model as well as the flux from the model counts for $\gamma =1.1$. It is clear that all the different regions are being well fit by the NFW-like density profile. The errors shown are solely the Poisson errors within the energy band and do not reflect covariances or systematic uncertainties.

Figure 8

Here we compare the flux spectra of the best fit GCE source with the flux spectra from eight globular clusters detected by Fermi LAT (47 Tuc, $\omega$ Cen, M62, NGC 6388, Terzan 5, NGC 6440, M28, NGC 6652). The three best fit GCE spectra shown are from the full model with a power-law exponential cutoff spectrum (solid), from the full model with a log-parabola spectrum (dashed) and from the 0.7–7 GeV analysis with a log-parabola spectrum (dotted). All the spectra are normalized by their fluxes for energies greater than 2 GeV.

Figure 9

Shown are the $2\mathrm{\Delta }\ln (\mathcal{L})$ for the best fit dark matter particle masses for (a) pure $b\overline{b}$ and (b) pure ${\tau }^{+}{\tau }^{-}$ annihilation channels, for several astrophysical model cases studied when varying all sources on the GC ROI. In both panels, the cases for $2\mathrm{FGL}+2\mathrm{PS}+\mathrm{GCE}$ show the exact particle mass runs in orange circles, $2\mathrm{FGL}+2\mathrm{PS}+\mathrm{I}+\mathrm{MG}+\mathrm{GCE}$ case in green triangles, and the full best fit model $2\mathrm{FGL}+2\mathrm{PS}+\mathrm{I}+\mathrm{MG}+\mathrm{ND}+\mathrm{GCE}$ case in blue squares. Fourth order spline interpolations are shown as lines for each case, which are used to find the minima and limits. For the full $2\mathrm{FGL}+2\mathrm{PS}+\mathrm{I}+\mathrm{MG}+\mathrm{ND}+\mathrm{GCE}$ model, the $b\overline{b}$ and ${\tau }^{+}{\tau }^{-}$ are equivalent in their goodness of fit, and there is no evidence for a mixed channel. The horizontal lines are for 2,3, and $5\sigma$ limits.

Figure 10

Shown are the systematic and statistical uncertainties in determining the GCE source spectrum. The errors represent the SED-normalization statistical uncertainty within an energy band, while the several cases represent the inherent systematic uncertainty present in the adoption of the GCE source’s spectral form.

Figure 11

Shown are limits on several channels when assuming that the new extended source is associated with MSP or other astrophysical emission in the models we study, for (a) the $b\overline{b}$, (b) ${\tau }^{+}{\tau }^{-}$, and (c) $W^{+}{W}^{-}$, in comparison with combined dwarf galaxy limits [14] and limits from HESS observations toward the Milky Way GC [25]. In (d) we show the strong model dependence of the limits, with the adopted full model limits being $2\mathrm{FGL}+2\mathrm{PS}+\mathrm{I}+\mathrm{MG}+\mathrm{MSP}+\mathrm{ND}$ solid (blue). The shaded box is for the case of $2\mathrm{FGL}+2\mathrm{PS}+\mathrm{MG}$, where there is the detection.