Looking for the γ-Ray Cascades of the KM3-230213A Neutrino Source

TL;DR

This study tests whether the KM3NeT KM3-230213A ultrahigh-energy neutrino could be accompanied by GeV–TeV gamma-ray cascades in the extragalactic background light. Using the gamma-Cascade framework, it models cascades from a putative $\mathcal{O}(100\text{ PeV})$ hadronic transient and performs a comprehensive, time-resolved Fermi-LAT search over $17$ years, scanning a region around the neutrino localization. No significant gamma-ray counterpart is detected, enabling constraints on the source redshift $z$ and intergalactic magnetic field $B$, with mappings showing that $B_0 \gtrsim 10^{-14}$ G is disfavored for $z<1$ under the small-angle approximation; extending observations to ~$7$ years approaches the limits of this approximation. The analysis also identifies a nearby Fermi-LAT source with a hard 2010--2011 period but disfavors a cascade origin for that emission, and discusses alternative explanations including hidden accelerators producing neutrinos without detectable GeV cascades. The results highlight the complementary power of neutrino and gamma-ray data for probing UHE astrophysical engines and cosmological magnetic fields, and point to future MeV instruments and extended LAT monitoring as critical to further constrain cascade scenarios and IGMF.

Abstract

The extreme energy of the KM3-230213A event could transform our understanding of the most energetic sources in the Universe. However, it also reveals an inconsistency between the KM3NeT detection and strong IceCube constraints on the ultra-high energy neutrino flux. The most congruous explanation for the KM3NeT and IceCube data requires KM3-230213A to be produced by a (potentially transient) source fortuitously located in a region where the KM3NeT acceptance is maximized. In hadronic models of ultra-high-energy neutrino production, such a source would also produce a bright γ-ray signal, which would cascade to GeV--TeV energies due to interactions with extragalactic background light. We utilize the γ-Cascade package to model the spectrum, spatial extension, and time-delay of such a source, and scan a region surrounding the KM3NeT event to search for a consistent γ-ray signal. We find no convincing evidence for a comparable \textit{Fermi}-LAT source and place constraints on a combination of the source redshift and the intergalactic magnetic field strength between the source and Earth.

Figures (7)

• Figure 1: Comparison of visible cascades for $T_{\text{obs}} = 1$ year, $z=0.1$, and varying IGMF strengths, computed assuming IGMF evolution with $\alpha = 4$. Our results, computed using the semi-analytic $\gamma$-CascadeV4 code, are shown in darker shades of teal-green curves, and are compared to the Monte Carlo-based ELMAG package in Ref. Fang:2025nzg shown in lighter shades. The orange dashed line represents the approximation given in Eq. \ref{['eq:approx']}, applied to the universal spectrum of a fully developed electromagnetic cascade, $dN/dE_\gamma \propto E_\gamma^{-1.9}$Berezinsky:2016feh, from which we predict an energy spectrum as a power-law with spectral index of $\Gamma \sim$ 0.9. The expected KM3NeT neutrino spectrum under the assumption of photo-pion production is shown in pink.
• Figure 2: Spectral energy distribution (SED) and delta-log-likelihood ($\Delta \mathcal{L}$) profile for the region centered on KM3-230213A, covering 100 MeV to 500 GeV, during the 2023--24 time frame following the reported neutrino event. The black arrows represent 95% CL flux upper limits, obtained under the assumption of a power-law spectrum with index $\Gamma=0.9$, appropriate for a one-year cascade timescale. The color scale indicates $\Delta \mathcal{L}$ as a function of flux and energy.
• Figure 3: Time-resolved $\gamma$-ray residual maps from the Fermi-LAT, showing intervals of one week (left), one month (center), and the total (right) Fermi-LAT exposure relative to the KM3-230213A neutrino event time ($T_0$). The color scale represents $\sqrt{\mathrm{TS}}$, indicating the detection significance. The white star denotes the best-fit position of KM3-230213A. The solid white circle represents $1.5^\circ$ containment radius (68% KM3NeT containment) while the dashed circle corresponds to $3^\circ$ (99% KM3NeT containment). No significant transient or steady-state $\gamma$-ray emission is observed at the location of KM3-230213A. We note a new bright, unassociated excess observed in the upper-right region of the "Fermi-LAT [start] $\rightarrow$$T_0$" residual map (green circle), approximately $2^\circ$ northwest from the reported neutrino position. The known 4FGL J0616.2-0653 source is also shown in the same residual plot, indicated by the yellow circle.
• Figure 4: Analogous to Fig. \ref{['fig:panel1']}, but showing $\gamma$-ray residual maps binned in one-year intervals defined relative to the KM3-230213A neutrino event time $T_0$. The final panel shows the residual map for the entire period from 2008 to 2025. No significant transient or steady-state $\gamma$-ray emission is observed at the KM3-230213A neutrino position. The new, unassociated excess (green circle) is still observed in the upper-right region, particularly in the 2010--11 and 2008--25 residual maps, approximately $2^\circ$ northwest from the best-fit neutrino position. The known 4FGL J0616.2-0653 source is also shown in the 2008--25 residual plot as a marginal detection, indicated by the yellow circle.
• Figure 5: Constraints on the IGMF strength as a function of the redshift to the putative KM3NeT source, derived assuming IGMF evolution with $\alpha = 4$ and no detection of the cascaded $\gamma$-ray flux within periods of 1 year (black), 2 years (cyan), or 7 years (green-dashed) after the neutrino event. Shaded regions indicate parameter space excluded by our analysis. Constraints are shown at the best-fit location of the KM3NeT event, but change only negligibly over the angular uncertainty region of the event. The 1- and 2-year constraints rely on the small-angle approximation ($\delta\ll\pi$), assuming minimal $\gamma$-ray deflection. We demonstrate that extending the observation window to 7 years following the KM3-230213A (i.e., 5 more years of Fermi-LAT observations) reaches the limit of this approximation ($\sim10^{-13}$ G).