New search for monochromatic neutrinos from dark matter decay

From data recently reported from the IceCube telescope, we derive new bounds on the monochromatic neutrino signal produced from dark matter particle decays. In the few TeV to tens of TeV energy range, these bounds turn out to be better than previous limits by more than an order of magnitude. As a result, intensity constraints on neutrino lines at energies above a few TeV are now comparable to those on gamma-ray lines. From the same data sample, we also perform a detailed search for a neutrino line, showing that there is no significant hint for such a signal.

Figure 1

Deposited energy spectrum from the full sky (black curve) as measured by IceCube. The colored regions show the expected atmospheric muon (bottom green) and neutrino (middle red) fluxes as well as the best-fit astrophysical neutrino power-law background model contribution (top blue). Data taken from Ref. [32].

Figure 2

Deposited-energy ($E_{\mathrm{dep}}$) spectrum for DM decay into monochromatic neutrinos of energy $E_{\nu }^{\mathrm{line}}={10}^5\mathrm{GeV}$. The signal normalization is for a DM particle with mass $2\times {10}^5\mathrm{GeV}$ and lifetime ${\tau }_{\mathrm{DM}}={10}^{28}\mathrm{s}$, with democratic neutrino flavor mixture and equal parts of $\nu$ and $\overline{\nu }$ in the flux reaching the Earth’s surface. The diagram shows the total number of events (black curve) after integration over the full sky and including propagation effects through the Earth. The dominating Galactic DM contribution (blue) and the extragalactic DM contribution (red) are shown separately. The dashed curves enclose the effective area uncertainty range when the presented uncertainties of IceCube’s instrument response function in the different sky directions are propagated and added in quadrature. The energy binning is the same as in the published instrument response function [32].

Figure 3

95% C.L. limits on the lifetime of DM particle decay into monochromatic neutrinos. Solid (black, blue and red) curves are for DM decay into pure neutrino flavor states (${\nu }_e$, ${\nu }_{\mu }$ and ${\nu }_{\tau }$, respectively) at production, which after propagation to the Earth are no longer pure flavor states. The companion final state $X$ in the DM decay $\mathrm{DM}\to {\nu }_{\alpha }+X$ is assumed to be light compared to the DM particle mass $m_{\mathrm{DM}}$. The dashed red curve shows the limit for $\mathrm{DM}\to {\nu }_{\tau }+X$ if no atmospheric background subtraction is made. For comparison, we show the limits by Rott, Kohri and Park [23] (90% C.L., green dotted curve) as well as Esmaili, Kang and Serpico [34] (90% C.L., green dashed-dotted curve) who analyzed the 3-year high-energy data set from Ref. [35]. Equal parts of $\nu$ and $\overline{\nu }$ are assumed for the DM signals.

Figure 4

TS distribution from $2.02\times {10}^4$ monochromatic neutrino-line searches (at our 101 used DM masses) in Monte Carlo-generated pseudodata from our NULL background model assumption. The data follows well a $\frac{1}{2}\delta (\mathrm{TS})+\frac{1}{2}{\chi }^2(\mathrm{TS})$ distribution (shown by the dashed blue line after multiplication by $2.02\times {10}^4$). Dotted grey curves show the TS distribution from $m_{\mathrm{DM}}<560\mathrm{TeV}$ (dark grey) and $m_{\mathrm{DM}}>560\mathrm{TeV}$ (light grey) line searches separately.

Figure 5

Local fit significance of a monochromatic line from DM decays vs DM mass (solid red curve). For comparison we also show the TS evaluated for signals at our “robust” limits with the method of Sec. 2c (black dashed) and our used profile log-likelihood ratio limit at $\mathrm{TS}=2.71$ (blue dotted curve). Democratic flavor composition and equal parts of $\nu$ and $\overline{\nu }$ are assumed at the Earth’s surface.

Figure 6

95% C.L. lifetime limits (solid curve) on the DM particle decay lifetime into monochromatic neutrinos. The expected sensitivity reach (dashed curve) and its 68% (yellow) and 95% (green) containment bands are also shown. The thin red line shows the 99.9% C.L. limits. Democratic flavor composition and equal parts of $\nu$ and $\overline{\nu }$ are assumed for the incoming neutrino flux at the Earth’s surface.

Figure 7

95% C.L. lifetime limits on various neutrino-line-like signals from DM decay: monochromatic line, internal bremsstrahlung ($dN/d{E}_{\nu }\propto E_{\nu }$ and $\propto E_{\nu }^3$) and box-like spectrum ($dN/d{E}_{\nu }=\text{constant}$). Democratic flavor composition and equal parts of $\nu$ and $\overline{\nu }$ are assumed after propagation to the Earth.

Figure 8

Lifetime limits on DM particle decay into monochromatic lines $\nu +\gamma$ from Fermi-LAT [80] (blue, dashed) and H.E.S.S. [12, 28] (blue, dotted) using gamma-ray data compared to the neutrino-line bounds derived in this study (solid black) as well as previous IceCube limits [31] (thick green).

Figure 9

Observed deposit energy spectra compared to our best-fit model with a DM particle mass of 44.8 TeV (blue, and blue-dotted for the DM contribution), our second best-fit model with a DM particle mass of 2.52 PeV (red, and red-dotted for the DM contribution), our best-fit background-only model (orange) and the IceCube Collaboration’s [32] best-fit background model (grey). The best-fit DM decay signal contributions are shown for the 44.8 TeV and 2.52 PeV DM particles by the dashed red and dashed blue curves, respectively. DM models are for democratic flavor composition and with equal parts of $\nu$ and $\overline{\nu }$ flux at the Earth’s surface. The significance for both these DM signals is less than $1.5\sigma$.

Figure 10

Comparison of the 95% C.L. lower limit ${\tau }_{\text{limit}}^{95\%\mathrm{C}.\mathrm{L}.}$ to the true DM lifetime value ${\tau }_{\mathrm{DM}}$ for a set of 100 Monte Carlo realizations (for each of the 101 DM masses tested between 2 TeV and 200 PeV). Each realization is from a model with a DM particle lifetime ${\tau }_{\mathrm{DM}}$ randomly drawn between 0.1 and 10 times the lifetime limit derived from the IceCube data ${\tau }_{\text{limit}}^{\mathrm{IC}}$. The color of each point gives the found significance $\sqrt{\mathrm{TS}}$ (i.e. the number of standard deviations $\sigma$) for the injected DM model. Points above the solid horizontal line represent realizations where the lower lifetime limit covers the true injected value. Democratic flavor and equal parts of $\nu$ and $\overline{\nu }$ composition was assumed at the Earth’s surface for the DM signals.

Figure 11

95% C.L. limits on the lifetime of DM particle decay into monochromatic neutrinos for pure $e$ (black), $\mu$ (blue) or $\tau$ (red) neutrinos injected at the source and with equal parts of $\nu$ and $\overline{\nu }$.

Figure 12

Same as Fig. 11 with the pure ${\nu }_e$, ${\nu }_{\mu }$ and ${\nu }_{\tau }$ at the source, but also separated into neutrinos (solid) or antineutrinos (dashed).

Figure 13

Same as Fig. 12 but in the case of pure ${\nu }_e$, ${\nu }_{\mu }$ and ${\nu }_{\tau }$ fluxes at the Earth’s surface. A pure flavor flux is typically not realistic due to neutrino oscillations, but the plot clearly illustrates the differences in effective area depending on the incoming neutrino flavor.