Detection of ultra-high-energy cosmic rays in the southern hemisphere with FAST: data acquisition and preliminary results

TL;DR

FAST tackles the sparse-fluorescence detection problem for ultra-high-energy cosmic rays by developing autonomous triggers robust to floating baselines. The authors introduce two in-house triggering approaches and compare them with reference methods using Monte Carlo EAS simulations and real southern-hemisphere FAST data, finding the in-house methods superior at low signal levels. They provide a detailed threshold calibration and baseline characterization, and report preliminary southern UHECR detections (≈$268$ events) with indicative energy-distance reach near $60\,\mathrm{EeV}$ up to $\sim 20\,\mathrm{km}$. The work supports deploying a FAST mini-array with stereo coincidences to enable full geometry/reconstruction and contribute to resolving energy-scale differences between major UHECR observatories, with implications for future GCOS-scale experiments.

Abstract

Ultra-high-energy cosmic rays (UHECRs) remain one of the greatest mysteries in astroparticle physics. The Fluorescence detector Array of Single-pixel Telescopes (FAST) is a next-generation cosmic ray experiment which utilizes ground-based fluorescence telescopes designed to detect these extremely rare particles at energies exceeding 30 EeV. FAST offers a cost-effective and low-maintenance solution to cover the huge detection areas required for UHECR observation. FAST telescopes are currently installed and remotely operated in both hemispheres, at the Pierre Auger Observatory and the Telescope Array experiment. To enable fully autonomous operation, a sophisticated trigger for data acquisition is essential. In this paper, we present two novel triggering algorithms inspired by those used at the largest observatories, but improved to meet the specific requirements imposed by the FAST design. Their performance is validated using Monte Carlo simulations of extensive air showers and UHECR events detected by the FAST telescope in the southern hemisphere. Finally, we present the sensitivity analysis estimate for FAST.

Figures (22)

• Figure 1: Schematic of the first-generation FAST telescope Fujii2016.
• Figure 2: Examples of acquired floating pedestals and their extracted baselines, recorded with the telescope shutter closed. The $x-$axis represents time, and the $y-$axis shows the estimated number of photoelectrons per $20 \, \unit{ns}$. The baseline can either overestimate (left panel) or underestimate (right panel) a potential signal.
• Figure 3: Left panel: The baseline combined with a simulated NB, i.e., it does not contain a signal. Right panel: The baseline combined with a simulated NB and a signal. The filtering process based on the Auger first-level trigger method is applied for both panels. Due to the floating baseline, the filtered data without a signal in the left panel reaches a higher maximum value ($\approx 4.30\,N_{\mathrm{p.e.}}/20\,\unit{ns}$) than in the case when a signal is present in the right panel ($\approx 4.14\,N_{\mathrm{p.e.}}/20\,\unit{ns}$).
• Figure 4: A general scheme for SNR calculation. $P_{\mathrm{test}}$ denotes the current tested position, and $P_{\mathrm{ma}}$ is the position used to estimate the baseline. The standard deviation of the filtered trace is computed over the interval between $P_{\mathrm{std}}$ and $P_{\mathrm{ma}}$. Values between $P_{\mathrm{nan}}$ and $P_{\mathrm{test}}$ are excluded from both the baseline and standard deviation calculations. In this example, the parameters are $w_{\mathrm{std}} = 2048$, $w_{\mathrm{ma}} = 513$, and $w_{\mathrm{nan}} = 256$ bins.
• Figure 5: Two-dimensional histogram of all baselines extracted from pedestal data acquired with the telescope shutter closed. The values, which are colored according to the colour bar on the right, are normalized to represent a PDF for each bin. Note that values above two have the same colour to show details for lower values.