Discovery of Localized Regions of Excess 10-TeV Cosmic Rays

The 7 year data set of the Milagro TeV observatory contains $2.2\times {10}^{11}$ events of which most are due to hadronic cosmic rays. These data are searched for evidence of intermediate scale structure. Excess emission on angular scales of $\sim 10°$ has been found in two localized regions of unknown origin with greater than $12\sigma$ significance. Both regions are inconsistent with pure gamma-ray emission with high confidence. One of the regions has a different energy spectrum than the isotropic cosmic-ray flux at a level of $4.6\sigma$, and it is consistent with hard spectrum protons with an exponential cutoff, with the most significant excess at $\sim 10\mathrm{TeV}$. Potential causes of these excesses are explored, but no compelling explanations are found.

Figure 1

Map of significances for the Milagro data set without any cuts to remove the hadronic cosmic-ray background. A 10° bin was used to smooth the data, and the color scale gives the significance. The solid line marks the Galactic plane, and every 10° in Galactic latitude are shown by the dashed lines. The black dot marks the direction of the heliotail, which is the direction opposite the motion of the solar system with respect to the local interstellar matter. The fractional excess of region $A$ is $\sim 6\times {10}^{-4}$, while for region $B$ it is $\sim 4\times {10}^{-4}$. The deep deficits bordering the regions of excess appear because the background calculation has been raised by the excess.

Figure 2

Signal and background events vs RA for $10°<\delta <20°$. The plot was made using independent 10° $\delta$ by 1° RA bins (i.e., no smoothing). A subset of the data was used in which there are only full days of data in order to give an approximately uniform exposure in RA. Region $A$ corresponds to the excess at $\mathrm{RA}\approx 70°$. This plot shows that the region $A$ excess is inherent in the raw signal data and is not due to an underestimation of the background.

Figure 3

(a) Differential plot of the fractional excess versus ${log}_e(f_{\mathrm{out}})$ for regions $A$ and $B$, where $f_{\mathrm{out}}$ is the fraction of the outriggers hit. The spectrum of region $A$ is significantly different than the background ($2\times {10}^{-6}$), which is represented by the horizontal line. (b) Profile plot of the simulated energy of protons for the as a function of ${log}_e(f_{\mathrm{out}})$. The ranges are asymmetric and contain the inner 68% of simulated events.

Figure 4

Results of a ${\chi }^2$ fit to the excesses in regions $A$ and $B$, assuming a pure-proton spectrum of the form in Eq. (1). The top panel shows the results for region $A$ and the bottom panel shows the results for region $B$. The $1\sigma$, $2\sigma$ and $3\sigma$ allowed regions of the spectral index $\gamma$ and the cutoff energy $E_c$ are indicated by the shaded regions.