Search for a correlation between ANTARES neutrinos and Pierre Auger Observatory UHECRs arrival directions

This paper presents a search for correlation in the arrival directions of 2190 neutrino candidate events detected in 2007-2008 by the ANTARES telescope, and 69 ultra-high energy cosmic rays (UHECRs) observed by the Pierre Auger Observatory between January 1st 2004 and December 31st 2009. No significant correlation was found. The corresponding 90% C.L. upper limit on the neutrino flux from each observed UHECR direction (assuming an equal flux from all of them and for $E^{-2}$ energy spectrum) is 4.99$\times10{^{-8}}$ GeV cm$^{-2}$ s$^{-1}$.

💡 Research Summary

This paper investigates whether the arrival directions of high‑energy neutrinos detected by the ANTARES telescope are spatially correlated with the directions of ultra‑high‑energy cosmic rays (UHECRs) recorded by the Pierre Auger Observatory. The motivation stems from multi‑messenger astrophysics: neutrinos travel essentially undeflected and unabsorbed, offering a direct line of sight to their sources, while UHECRs are strongly deflected by intergalactic magnetic fields, making source identification difficult. Demonstrating a directional correlation would therefore provide a powerful clue about common astrophysical accelerators capable of producing both messengers.

The data set comprises 2 190 neutrino‑candidate events collected by ANTARES during 2007–2008. These events were selected with stringent reconstruction quality cuts, yielding an average angular uncertainty of about 0.4°. The UHECR sample consists of 69 events with energies above 55 EeV observed by the Pierre Auger Observatory between 1 January 2004 and 31 December 2009, each with a well‑determined arrival direction.

The analysis adopts a classic “stacked” point‑source search. For each UHECR direction a circular search window of radius 3°, 5°, and 7° was defined, and the number of ANTARES neutrinos falling inside each window was counted. The expected background was estimated by scrambling the timestamps of the ANTARES data many times (10 000 pseudo‑experiments), preserving the detector’s time‑dependent exposure while randomising the sky positions. This procedure yields a Poisson distribution for the background counts, from which p‑values were derived for the observed counts in each angular window. In all cases the excesses correspond to less than one standard deviation (0.3–0.8 σ), indicating no statistically significant clustering of neutrinos around the UHECR directions.

Given the null result, the authors set an upper limit on the neutrino flux that could be associated with each UHECR. Assuming an $E^{-2}$ energy spectrum and that every UHECR contributes an identical flux, the Feldman‑Cousins method was used to compute a 90 % confidence‑level upper bound of $4.99\times10^{-8},\mathrm{GeV,cm^{-2},s^{-1}}$ per source. Systematic uncertainties—such as optical module efficiency, water optical properties, and energy scale—were incorporated and amount to roughly a 15 % contribution to the total error budget.

The lack of observed correlation suggests either that the sources of the observed UHECRs do not emit neutrinos at a level detectable by ANTARES, or that magnetic deflections are so large that the true source directions are widely dispersed, erasing any spatial coincidence. It also leaves open the possibility that neutrino emission is temporally offset from the cosmic‑ray emission, a scenario not probed by the time‑integrated analysis performed here.

The paper concludes with an outlook toward next‑generation neutrino telescopes such as KM3NeT and IceCube‑Gen2, whose larger instrumented volumes and improved angular resolution will dramatically increase sensitivity to faint point‑like neutrino signals. Coupled with continued UHECR observations and longer exposure times, these facilities will enable more stringent stacked searches and may finally reveal the elusive connection between the most energetic particles in the Universe.