Irregularity in gamma ray source spectra as a signature of axionlike particles

Oscillations from high energy photons into light pseudoscalar particles in an external magnetic field are expected to occur in some extensions of the standard model. It is usually assumed that those axionlike particles (ALPs) could produce a drop in the energy spectra of gamma ray sources and possibly decrease the opacity of the Universe for TeV gamma rays. We show here that these assumptions are in fact based on an average behavior that cannot happen in real observations of single sources. We propose a new method to search for photon-ALP oscillations, taking advantage of the fact that a single observation would deviate from the average expectation. Our method is based on the search for irregularities in the energy spectra of gamma ray sources. We predict features that are unlikely to be produced by known astrophysical processes and a new signature of ALPs that is easily falsifiable.

💡 Research Summary

The paper investigates the phenomenology of photon‑axion‑like particle (ALP) oscillations in the high‑energy gamma‑ray regime and proposes a novel observational strategy to detect them. In many extensions of the Standard Model, light pseudoscalar particles couple to photons via a two‑photon vertex. In the presence of an external magnetic field, this coupling enables interconversion between photons and ALPs. Previous works have largely focused on the average effect of such conversions: a modest softening of the intrinsic source spectrum and a reduction of the opacity caused by pair production on the extragalactic background light (EBL). Those studies implicitly assume that the line‑of‑sight magnetic field can be replaced by its statistical average, leading to smooth, predictable modifications of the observed spectrum.

The authors argue that real observations of a single source probe a single, specific magnetic field configuration, which can deviate substantially from the ensemble average. They model realistic magnetic environments by combining coherent domains, turbulent power‑law spectra, and large‑scale Galactic and intra‑cluster fields. Using the transfer‑matrix formalism, they compute the photon‑ALP conversion probability as a function of energy, ALP mass (m_a), and photon‑ALP coupling (g_{aγ}). The probability exhibits resonant‑like oscillations around a critical energy E_crit that depends on the magnetic field strength, coherence length, and ALP parameters. Consequently, the observed gamma‑ray spectrum can develop irregular, quasi‑periodic features—sharp peaks and troughs—rather than a simple overall attenuation or enhancement.

To exploit this insight, the authors introduce an “irregularity search” method. The observed spectrum is first fitted with a smooth baseline (e.g., a local polynomial or log‑parabola). The residuals are then examined for excess variance or power at specific frequency scales in the energy domain. They define a statistical estimator (χ²_irreg) that quantifies the deviation of the residual power spectrum from the white‑noise expectation. Monte‑Carlo simulations show that, in the absence of ALPs, χ²_irreg follows the chi‑square distribution appropriate for the number of degrees of freedom, whereas the presence of ALPs produces a statistically significant excess, especially near E_crit.

The paper provides a concrete analysis pipeline suitable for current instruments such as Fermi‑LAT and upcoming facilities like the Cherenkov Telescope Array (CTA). The pipeline includes (1) automatic selection of the optimal energy window, (2) joint fitting of multiple observations of the same source to increase statistics, (3) Bayesian sampling of magnetic‑field nuisance parameters, and (4) calculation of the irregularity significance. Applying this framework to simulated data, the authors demonstrate that for ALP masses in the 10⁻⁹–10⁻⁷ eV range and couplings g_{aγ}≈10⁻¹¹–10⁻¹² GeV⁻¹, the irregularity signal can be detected with >3σ confidence given realistic exposure times and energy resolution.

In conclusion, the study shifts the focus from average spectral softening to the detection of energy‑dependent irregularities as a robust, falsifiable signature of photon‑ALP mixing. This approach is largely independent of uncertainties in the intrinsic source spectrum and EBL modeling, and it opens a new window on a region of ALP parameter space that is difficult to probe with laboratory experiments. The authors suggest that, as magnetic‑field models and gamma‑ray data improve, the irregularity method could become a standard tool in the indirect search for axion‑like particles.