⚛️ phenomenology

Correlation of galactic very-high-energy gamma-ray sources detected by LHAASO with high-energy neutrinos

This study finds no compelling evidence for a spatial correlation between LHAASO-detected very-high-energy gamma-ray sources and IceCube high-energy neutrino alerts, suggesting that current LHAASO sources are unlikely to be the primary counterparts of these neutrinos and highlighting the need for more sensitive future instruments like IceCube-Gen2.

Original authors: Long-Yu Zhang, Kai Wang

Published 2026-10-15

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Long-Yu Zhang, Kai Wang

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Deep within the frozen ice of the Antarctic continent lies a massive detector designed to catch the universe's most elusive messengers: high-energy neutrinos. These are ghostly particles, nearly massless and electrically neutral, that can travel across the cosmos without being stopped by stars, gas, or dust. For decades, scientists have hunted for their origins, hoping they would reveal the locations of the most violent and energetic engines in our galaxy. At the same time, a different kind of observatory in China, perched on a high mountain plateau, scans the sky for very-high-energy gamma rays. These are the most energetic forms of light, often produced when cosmic rays—fast-moving atomic nuclei—are accelerated to incredible speeds. A long-standing mystery in astrophysics asks whether these gamma rays are created by electrons or by protons. If they come from protons, the same collisions should also produce neutrinos. Finding a gamma-ray source and a neutrino coming from the exact same spot in the sky would be the smoking gun, proving that the universe is accelerating protons to energies far beyond what any human-made machine can achieve.

A team of researchers recently set out to test this connection using data from two major observatories: the Large High Altitude Air Shower Observatory, known as LHAASO, and the IceCube Neutrino Observatory. LHAASO has mapped the sky and identified ninety distinct sources of very-high-energy gamma rays, some of which reach energies a million times higher than what medical X-rays use. IceCube, meanwhile, has recorded hundreds of high-energy neutrino events that have been flagged as potential signals from deep space. The scientists asked a simple but profound question: do these gamma-ray sources and neutrino events overlap in the sky? If the gamma rays are indeed the byproduct of proton collisions, the neutrinos should be coming from the same places.

To answer this, the researchers compared the positions of the ninety gamma-ray sources found by LHAASO against the locations of 348 neutrino events recorded by IceCube. They focused specifically on the most energetic neutrinos, those with energies above 100 TeV, because these are the ones most likely to be linked to the powerful accelerators LHAASO has spotted. The team looked for any instance where the uncertain position of a neutrino event touched the area of a gamma-ray source. They found several such overlaps. In nine cases, a neutrino event appeared to share a location with a gamma-ray source when using a strict definition of position uncertainty, and in ten cases when the definition was slightly relaxed.

However, finding two things in the same place does not always mean they are related. The sky is crowded with sources, and neutrinos can arrive from anywhere. To determine if these overlaps were meaningful or just a random accident, the team ran a massive computer simulation. They shuffled the neutrino positions around the sky thousands of times, keeping the gamma-ray sources fixed, to see how often a random neutrino would land near a source by pure chance. The results of this simulation showed that the overlaps they found were not rare. In fact, the statistical probability of seeing this many overlaps by chance was quite high. This means the data does not support the idea that the gamma-ray sources and the neutrinos are physically connected. The spatial coincidence appears to be accidental rather than a sign of a shared origin.

The researchers took the investigation a step further by asking a theoretical question: even if the positions matched, could these sources actually produce enough neutrinos to explain what IceCube sees? They assumed the most favorable scenario possible: that every single gamma ray detected by LHAASO was created by protons colliding with gas, a process that inevitably produces neutrinos. Under this assumption, they calculated the maximum possible number of neutrinos these sources could be sending our way. When they compared this theoretical maximum to the actual number of neutrinos detected by IceCube, the results were disappointing. For the vast majority of the sources, the theoretical neutrino output was far too low to account for the observed neutrino events. Even in the most optimistic case, where every gamma ray is a sign of proton activity, the sources simply do not have the power to produce the neutrinos IceCube is catching.

This study does not rule out the possibility that some of these sources are indeed cosmic particle accelerators, but it suggests that they are not the ones responsible for the specific neutrino events detected so far. The lack of a strong link implies that either the gamma rays are being produced by electrons rather than protons, or that the neutrinos are coming from different, yet-to-be-discovered sources. The findings highlight a gap in our current understanding and point toward the need for more sensitive instruments. The researchers conclude that to truly solve the mystery of galactic cosmic rays, we will need detectors with much greater sensitivity, such as the upcoming IceCube-Gen2, which will be able to see fainter signals and pinpoint sources with much higher precision. Until then, the connection between these high-energy gamma-ray beacons and the ghostly neutrinos remains an open question, waiting for clearer evidence from the next generation of cosmic observatories.

Technical Summary: Correlation of Galactic Very-High-Energy Gamma-Ray Sources Detected by LHAASO with High-Energy Neutrinos

Problem Statement
The Large High Altitude Air Shower Observatory (LHAASO) has identified 90 very-high-energy (VHE, $E > 0.1$ TeV) gamma-ray sources, many of which are potential PeVatrons capable of accelerating particles to PeV scales. A critical unresolved question in high-energy astrophysics is whether these VHE gamma-rays originate from hadronic processes (proton interactions) or leptonic processes (electron interactions). Hadronic interactions ($pp$ or $p\gamma$ ) produce both VHE gamma-rays and high-energy neutrinos, whereas leptonic processes produce gamma-rays without accompanying neutrinos. Consequently, high-energy neutrinos serve as an "exclusive probe" for identifying hadronic activity. Despite the detection of a galactic diffuse neutrino signal by the IceCube Neutrino Observatory, no credible individual galactic neutrino source has been identified. This study investigates the spatial and physical correlation between the 90 LHAASO sources and high-energy track-like neutrino events from the IceCube Event Catalog of Alert Tracks (IceCat-1) to determine if LHAASO sources are the counterparts to these neutrino alerts.

Methodology
The analysis utilizes two primary datasets:

LHAASO Sample: The First LHAASO Catalog containing 90 VHE gamma-ray sources. The study considers the full catalog, as well as subsets detected solely by the Kilometer Squared Array (KM2A) and the Water Cherenkov Detector Array (WCDA).
IceCube Sample: The IceCat-1 catalog, comprising 348 track-like neutrino alert events. To focus on potential PeVatron candidates, the authors apply strict selection criteria: neutrino energy $E_\nu \geq 100$ TeV and angular uncertainty $\Omega_{90} \leq 10, 15,$ and $20$ deg $^2$ . This yields 202, 234, and 249 qualified events, respectively.

The methodology proceeds in three stages:

Spatial Coincidence Search: The authors define a spatial coincidence as an intersection between the effective extension radius of a LHAASO source (calculated as $\sqrt{\sigma_{p,95}^2 + r_{39}^2}$ ) and the 90% positional uncertainty region of a neutrino alert.
Statistical Significance (Monte Carlo): To distinguish physical associations from random spatial overlaps, the authors perform $10,000$ Monte Carlo (MC) simulations. In these simulations, the Right Ascension (RA) of neutrino events is randomized while keeping declination and error regions fixed (to preserve IceCube's zenith-dependent sensitivity). The $p$ -value is calculated as the probability of obtaining a coincidence count equal to or greater than the observed count ( $\nu_{real}$ ) under the null hypothesis of random distribution.
Flux Comparison: Assuming the gamma-rays are produced exclusively via hadronic $pp$ interactions, the authors estimate the "most optimistic" theoretical neutrino flux for the overlapping sources. They model the parent proton spectrum as an exponential cut-off power law and calculate the expected neutrino flux. This theoretical flux is then compared against the observed flux constraints derived from IceCat-1 alert events within the energy range $[30 \text{ TeV}, 300 \text{ TeV}]$ .

Key Results

Spatial Coincidences: The search identified spatial overlaps between LHAASO sources and neutrino alerts. For the baseline selection ( $\Omega_{90} \leq 10$ deg $^2$ ), 9 coincidences were found involving 7 neutrino events and 8 LHAASO sources.
Statistical Significance: The MC simulations yielded $p$ -values of 0.237 (all sources), 0.312 (KM2A only), and 0.115 (WCDA only) for the baseline selection. These values are well above the standard significance threshold (0.05), indicating that the observed spatial overlaps are consistent with random chance. The results remained consistent across different $\Omega_{90}$ thresholds ($15$ and $20$ deg $^2$ ).
Flux Constraints: For the majority of the overlapping sources, the theoretical neutrino flux (assuming 100% hadronic origin) is lower than the flux required to explain the observed IceCube events. Specifically, for most pairs, the predicted flux falls below the 95% confidence level lower limit of the observed flux. Only three specific pairs (e.g., LHAASO J0634+1741u with IC111209A) showed theoretical fluxes that could potentially match the observations, but these remain statistically insignificant in the broader context.
Diffuse Emission Interference: The authors note that many LHAASO sources lie near the Galactic plane ( $|b| < 12^\circ$ ), where significant diffuse neutrino emission exists. This mixing of discrete sources and diffuse background further weakens the reliability of associating specific neutrino alerts with specific LHAASO sources.

Significance and Conclusions
The paper concludes that while LHAASO sources are potential galactic neutrino emitters, current data does not provide compelling evidence for a direct physical connection between the 90 cataloged sources and the IceCat-1 high-energy neutrino alerts. The spatial coincidences appear to be accidental rather than physical. Furthermore, under the assumption of purely hadronic gamma-ray production, the maximum theoretical neutrino flux from these sources is generally insufficient to account for the detected neutrino events.

The authors emphasize that these findings underscore the necessity for instruments with elevated sensitivity, such as IceCube-Gen2, to definitively identify galactic PeVatrons. The study highlights limitations in the current analysis, including the use of a simple overlap-counting statistic (which is conservative), the restriction to high-threshold alert events rather than the full IceCube dataset, and the reliance on a single power-law spectral index. The paper asserts that future analyses utilizing the complete IceCube event sample and next-generation detectors are required to resolve the connection between galactic VHE sources and high-energy neutrinos.

More like this