High-energy neutrinos from multibody decaying dark matter

Since the report of the PeV–TeV neutrinos by the IceCube Collaboration, various particle physics models have been proposed to explain the neutrino spectrum by dark matter particles decaying into neutrinos and other standard model particles. In such scenarios, simultaneous $\gamma$-ray emission is commonly expected. Therefore, multimessenger connections are generally important for the indirect searches of dark matters. The recent development of $\gamma$-ray astronomy puts stringent constraints on the properties of dark matter, especially by observations with the Fermi $\gamma$-ray satellite in the last several years. Motivated by the lack of $\gamma$-ray as well as the shape of the neutrino spectrum observed by IceCube, we discuss a scenario in which the DM is a PeV scale particle which couples strongly to other invisible particles and its decay products do not contain a charged particle. As an example to realize such possibilities, we consider a model of fermionic dark matter that decays into a neutrino and many invisible fermions. The dark matter decay is secluded in the sense that the emitted products are mostly neutrinos and dark fermions. One remarkable feature of this model is the resulting broadband neutrino spectra around the energy scale of the dark matter. We apply this model to multi-PeV dark matter, and discuss possible observable consequences in light of the IceCube data. In particular, this model could account for the large flux at medium energies of $\sim 10–100\mathrm{TeV}$, possibly as well as the second peak at PeV, without violating the stringent $\gamma$-ray constraints from Fermi and air-shower experiments such as CASA-MIA.

Figure 1

Neutrino deposited energy histogram predicted by a decaying DM without assuming astrophysical contributions (Model 1). The dots with error bars represent the observed data points. The solid line is the theoretical prediction in the model. We assume that the DM with its mass $m_{\mathrm{DM}}=4\mathrm{PeV}$ decays into $N\sim 30$ particles. The branching ratio into the two-body decay is taken to be ${\mathrm{BR}}_{\mathrm{line}}=0.034$. In this model, the lifetime is fitted to be $1.46\times {10}^{27}\sec$. We also show the atmospheric contribution with short-dashed lines with its uncertainty (gray shaded regions).

Figure 2

Neutrino source spectrum in Model 1. Black points with error bars are data. Dashed line represents the model prediction. Model parameters are same in those used in Fig. 1. The IceCube data for the sum of all flavors come from the results of the combined likelihood analysis.

Figure 3

Deposited energy histogram of the neutrino spectrum combined the astrophysical component with those of the two- and multi-body decaying DM contributions (Model 2a). The total (astrophysical) contribution is represented in the solid (long-dashed) line. The short-dashed and shaded region corresponds to the atmospheric contributions and its uncertainty, which is same as those in Fig. 1. In this case, the DM with its mass $m_{\mathrm{DM}}=4\mathrm{PeV}$ also decays into $N\sim 30$ particles. The branching ratio into the line spectrum and the lifetime is assumed to be ${\mathrm{BR}}_{\mathrm{line}}=0.080$, and $\tau =3.41\times {10}^{27}\sec$, respectively.

Figure 4

Neutrino source spectrum in Model 2a corresponds to the Fig. 3. The dashed (dotted) line corresponds to the DM (astrophysical) contributions. The solid line represents the sum of those components.

Figure 5

Deposited energy histogram of the neutrino by both the decaying DM and the astrophysical components (Model 2b). Here we assume that the DM with its mass of $m_{\mathrm{DM}}=600\mathrm{TeV}$ decays into the $N\sim 30$ particles. Contribution from the two-body decay mode is negligible. In this case, the lifetime of the DM particle is fitted to be $\tau =5.56\times {10}^{26}\mathrm{s}$. The normalization of the astrophysical component is same as those of [14] in order.

Figure 6

Source spectrum of the neutrino in Model 2b derived with the same parameters assumed in those of Fig. 5. Lines are the same as those in Fig. 4.