TL;DR
The Pierre Auger Observatory has conducted a comprehensive search for EeV (10^18 eV) photons from point sources across the entire observable sky. Despite the sensitivity of the experiment, no statistically significant excess of photon events was found, leading to the establishment of stringent upper limits on the photon flux from any direction. These results help constrain astrophysical models for the origin of the highest-energy cosmic rays.
What Problem Does This Paper Address?
The origin of the highest-energy cosmic rays (EeV energies) remains one of the most significant mysteries in astrophysics. While protons are thought to be the dominant component, their exact sources are unknown. Photons, produced by processes like pion photoproduction or inelastic nuclear collisions near sources, can act as messengers, pointing to the origin of these energetic particles. Detecting EeV photons from discrete sources would provide direct evidence for such accelerators. However, EeV photons are attenuated by intergalactic photons, limiting the observable volume. This paper addresses the challenge of identifying such sources by searching for EeV photon fluxes from specific directions in the sky using data from the Pierre Auger Observatory.
Methods
The study employs a hybrid technique, combining data from the Pierre Auger Observatory's fluorescence detectors (FD) and surface detectors (SD). This allows for precise determination of shower geometry and energy. A multivariate analysis (MVA) using a Boosted Decision Tree (BDT) classifier is central to the method, aiming to distinguish between photon-induced showers and the much more abundant hadronic cosmic ray background.
Key observables used in the MVA include:
- FD observables:
- $X_{max}$: The depth of shower maximum, which is generally deeper for photon-induced showers due to their electromagnetic nature.
- $X_{arr}^2/\text{dof}$: A measure of the goodness of fit for the longitudinal profile using the Greisen function, expected to be better for pure electromagnetic showers.
- $E_{Gr}/E_{GH}$: The ratio of energies reconstructed by the Greisen and Gaisser-Hillas functions, also sensitive to shower development.
- SD observables:
- $S_b$: An estimator of the lateral distribution of the signal, which is steeper for photon-induced showers.
- ShapeP: A parameter related to the timing of signal arrival, expected to be larger for deeper developing showers (photons).
A "blind search" strategy was implemented, meaning the analysis was performed without looking at the data for specific candidate sources beforehand. This minimizes the risk of bias. The search targeted directions with a declination band from -85° to +20° and an energy range from 10¹⁷.³ eV to 10¹⁸.⁵ eV.
Datasets
The analysis utilizes data collected by the Pierre Auger Observatory between January 2005 and September 2011. For training and testing the MVA, simulated air showers were generated using the CORSIKA code with approximately 30,000 photon and 60,000 proton primaries. These simulations were reweighted to account for different spectral indices.
Key Insights and Ideas
- The "blind search" approach is crucial: By not pre-selecting candidate sources, the analysis avoids confirmation bias and provides a more robust assessment of the entire sky.
- Hybrid technique is powerful: Combining FD and SD data provides complementary information for precise shower characterization, essential for distinguishing between photon and hadron showers.
- No significant excesses found: The most striking finding is the absence of any statistically significant excess of photon events from any direction, despite the observatory's sensitivity.
- Constraining astrophysical models: The derived upper limits are significant because they rule out certain scenarios for the production of EeV cosmic rays, particularly those involving transient or steady sources within the Galaxy that would produce a detectable photon flux.
Key Results
- No point source detected: The blind search found no statistically significant excess of photon events from any direction.
- Upper limits on photon flux: Directional upper limits at the 95% confidence level were derived. The mean upper limit is 0.035 photons km⁻² yr⁻¹, with a maximum of 0.14 photons km⁻² yr⁻¹.
- Energy flux limits: Assuming an E⁻² spectrum, these flux limits correspond to an energy flux of 0.06 eV cm⁻² s⁻¹ on average, and a maximum of 0.25 eV cm⁻² s⁻¹.
- Exclusion of Galactic sources: These limits are stringent enough to exclude certain models of EeV cosmic ray production within the Galaxy, particularly those that would produce a detectable photon flux. For instance, a flux of 1.44 eV cm⁻² s⁻¹ would be detectable with 5σ significance.
Limitations and Future Work
- Systematic uncertainties: The study acknowledges systematic uncertainties, primarily related to the Auger energy scale, which impact the flux upper limits. These uncertainties range from ±30% at 10¹⁷.³ eV to ±10% above 10¹⁸ eV.
- Photon spectral index assumption: The analysis assumes a photon spectral index of -2. Variations in this assumption can significantly alter the derived upper limits.
- Future searches: The authors suggest that future work could focus on searching for photon fluxes from specific candidate sources or "stacks" of candidates with common astrophysical characteristics. A modest excess might become statistically significant with increased exposure. The study also notes that while this search is sensitive to sources within the Galaxy, extragalactic sources are also considered, with the caveat of intergalactic attenuation.
Summary updated