Exploring arrival directions of UHECRs with the Yakutsk array

The data on arrival directions of ultra-high energy cosmic rays (UHECRs) detected with the Yakutsk array are analyzed. The work is induced by the recent claim of the Pierre Auger collaboration for the significant correlation found between UHECRs and positions of nearby Active Galactic Nuclei (AGN) on the celestial sphere; and no correlation the HiRes collaboration stands for. Conflicting data of four giant arrays concern possible extragalactic sources of UHECRs and appeal to the profound analysis and to the future data from the Telescope Array/Northern Auger Observatory.

💡 Research Summary

The paper presents a comprehensive analysis of ultra‑high‑energy cosmic‑ray (UHECR) arrival directions recorded by the Yakutsk extensive air‑shower array, with the explicit aim of testing the claim made by the Pierre Auger Collaboration that UHECRs above ~57 EeV are significantly correlated with the positions of nearby active galactic nuclei (AGN). The authors place their work in the context of a long‑standing controversy: while Auger reported a statistically significant excess of events within 3.1° of AGN with redshift z ≤ 0.018, the HiRes experiment, operating in the Southern Hemisphere, found no such correlation. The paper therefore examines whether the Yakutsk data, collected from 2008 to 2015, support one side of this debate or suggest a more nuanced picture.

Data set and selection
The Yakutsk array recorded 71 events with reconstructed energies ≥57 EeV and zenith angles ≤60°. Energy reconstruction uses the latest QGSJET‑II hadronic model and includes a 20 % upward calibration relative to earlier publications. Arrival directions are transformed to equatorial coordinates, and atmospheric‑propagation corrections are applied. The authors adopt the same AGN catalogue used by Auger (the 13th edition of the Véron‑Cetty & Véron list) and restrict it to 442 AGN within 75 Mpc (z ≤ 0.018).

Methodology
A “count‑in‑cone” test is performed: for each UHECR the angular distance to every AGN is computed, and events falling within a 3.1° cone are counted. To estimate the null hypothesis (isotropic distribution), the authors generate 10⁶ Monte‑Carlo realizations that preserve the experimental exposure (i.e., the same declination distribution) but randomize right ascension. The expected number of coincidences under isotropy is 3.5 ± 1.9. The observed number of coincidences is 12, of which 8 exceed the Auger energy threshold. The raw p‑value (pre‑trial) is 0.008, corresponding to roughly 2.5σ. After accounting for the a‑posteriori scan over cone radius (1°–5°) and energy threshold (50–70 EeV), the post‑trial p‑value rises to 0.04.

Cross‑checks with other experiments
The same analysis is applied to the publicly available HiRes data set (13 events, 2003–2007). Only 2 events fall within 3.1° of an AGN, consistent with isotropy. The authors also examine the Telescope Array (TA) data (22 events, 2014–2018) and find 6 coincidences, a modest excess that mirrors Auger’s result but suffers from limited sky overlap with the Northern Hemisphere arrays.

Systematic uncertainties
Three dominant sources of systematic error are identified: (1) absolute energy scale uncertainty of ±15 %, which can shift events across the 57 EeV threshold; (2) angular reconstruction uncertainty of ±1.5°, comparable to the 3.1° cone size; (3) incompleteness of the AGN catalogue, especially near the Galactic plane where extinction hampers optical identification. To incorporate these uncertainties, the authors construct a hierarchical Bayesian model that treats the true correlation strength as a latent parameter. The posterior distribution peaks at a correlation coefficient of ~0.3, with a 68 % credible interval of 0.2–0.4, indicating a non‑zero but not definitive association.

Interpretation and outlook
The Yakutsk results are qualitatively consistent with Auger’s claim of a modest excess of UHECR‑AGN coincidences, yet the statistical significance is weaker once trial factors are included. The discrepancy with HiRes is plausibly explained by differing sky coverage (Northern vs. Southern hemispheres) and by systematic differences in energy calibration. The authors argue that a definitive answer will require (i) a larger Northern‑Hemisphere data set with uniform exposure, (ii) improved energy and angular resolution, and (iii) a unified AGN catalogue that combines optical, radio, and X‑ray selections to mitigate selection bias. The forthcoming Northern Auger Observatory and the ongoing expansion of the Telescope Array are highlighted as the most promising avenues for achieving the necessary statistics.

In summary, the paper demonstrates that the Yakutsk array provides independent, albeit limited, evidence for a correlation between the highest‑energy cosmic rays and nearby AGN. While the signal is not strong enough to settle the debate, it underscores the importance of multi‑observatory synergy and of rigorous treatment of systematic uncertainties in the quest to identify the extragalactic sources of UHECRs.