Abstract: On 2023 February 13, the KM3NeT/ARCA telescope observed a track-like event compatible with a ultra-high-energy muon with an estimated energy of 120 PeV, produced by a neutrino with an even higher energy, making it the most energetic neutrino event ever detected. A diffuse cosmogenic component is expected to originate from the interactions of ultra-high-energy cosmic rays with ambient photon and matter fields. The flux level required by the KM3NeT/ARCA event is, however, in tension with the standard cosmogenic neutrino predictions based on the observations collected by the Pierre Auger Observatory and Telescope Array over the last decade of the ultra-high-energy cosmic rays above the ankle (hence from the local Universe, z less than or similar to 1). We show here that both observations can be reconciled by extending the integration of the equivalent cosmogenic neutrino flux up to a redshift of zmax=6 and considering either source evolution effects or the presence of a subdominant independent proton component in the ultra-high-energy cosmic-ray flux, thus placing constraints on known cosmic accelerators.

Abstract: The KM3NeT Collaboration is building and operating two deep sea neutrino telescopes at the bottom of the Mediterranean Sea. The telescopes consist of latices of photomultiplier tubes housed in pressure-resistant glass spheres, called digital optical modules and arranged in vertical detection units. The two main scientific goals are the determination of the neutrino mass ordering and the discovery and observation of high-energy neutrino sources in the Universe. Neutrinos are detected via the Cherenkov light, which is induced by charged particles originated in neutrino interactions. The photomultiplier tubes convert the Cherenkov light into electrical signals that are acquired and timestamped by the acquisition electronics. Each optical module houses the acquisition electronics for collecting and timestamping the photomultiplier signals with one nanosecond accuracy. Once finished, the two telescopes will have installed more than six thousand optical acquisition nodes, completing one of the more complex networks in the world in terms of operation and synchronization. The embedded software running in the acquisition nodes has been designed to provide a framework that will operate with different hardware versions and functionalities. The hardware will not be accessible once in operation, which complicates the embedded software architecture. The embedded software provides a set of tools to facilitate remote manageability of the deployed hardware, including safe reconfiguration of the firmware. This paper presents the architecture and the techniques, methods and implementation of the embedded software running in the acquisition nodes of the KM3NeT neutrino telescopes. Program summary Program title: Embedded software for the KM3NeT CLB CPC Library link to program files: https://doi.org/10.17632/s847hpsns4.1 Licensing provisions: GNU General Public License 3 Programming language: C Nature of problem: The challenge for the embedded software in the KM3NeT neutrino telescope lies in orchestrating the Digital Optical Modules (DOMs) to achieve the synchronized data acquisition of the incoming optical signals. The DOMs are the crucial component responsible for capturing neutrino interactions deep underwater. The embedded software must configure and precisely time the operation of each DOM. Any deviation or timing mismatch could compromise data integrity, undermining the scientific value of the experiment. Therefore, the embedded software plays a critical role in coordinating, synchronizing, and operating these modules, ensuring they work in unison to capture and process neutrino signals accurately, ultimately advancing our understanding of fundamental particles in the Universe. Solution method: The embedded software on the DOMs provides a solution based on a C-based bare-metal application, operating without a real-time embedded OS. It is loaded into the RAM during FPGA configuration, consuming less than 256 kB of RAM. The software architecture comprises two layers: system software and application. The former offers OS-like features, including a multitasking scheduler, firmware updates, peripheral drivers, a UDP-based network stack, and error handling utilities. The application layer contains a state machine ensuring consistent program states. It is navigated via slow control events, including external inputs and autonomous responses. Subsystems within the application code control specific acquisition electronics components via the associated driver abstractions. Additional comments including restrictions and unusual features: Due to the operation conditions of the neutrino telescope, where access is restricted, the embedded software implements a fail-safe procedure to reconfigure the firmware where the embedded software runs.

Abstract: High-energy muon and electron neutrinos yield a non-negligible flux of tau neutrinos as they propagate through Earth. In this Letter, we address the impact of this additional component in the PeV and EeV energy regimes for the first time. Above 300 TeV, this contribution is predicted to be significantly larger than the atmospheric background, and it alters current and future neutrino telescopes' capabilities to discover a cosmic tau-neutrino flux. Further, we demonstrate that Earth-skimming neutrino experiments, designed to observe tau neutrinos, will be sensitive to cosmogenic neutrinos even in extreme scenarios without a primary tau-neutrino component.

Abstract: The Forward Physics Facility (FPF) plans to use neutrinos produced at the Large Hadron Collider to make a variety of measurements at previously unexplored TeV energies. Its primary goals include precision measurements of the neutrino cross section and using the measured neutrino flux both to uncover information about far-forward hadron production and to search for various beyond standard model scenarios. However, these goals have the potential to conflict: Extracting information about the flux or cross section relies upon an assumption about the other. In this paper, we demonstrate that the FPF can use the low-nu method-a technique for constraining the flux shape by isolating neutrino interactions with low energy transfer to the nucleus-to break this degeneracy. We show that the low-nu method is effective for extracting the nu μflux shape, in a model-independent way. We discuss its application for extracting the nu over bar μflux shape but find that this is significantly more model dependent. Finally, we explore the precision to which the nu μflux shape could be constrained at the FPF for a variety of proposed detector options. We find that the precision would be sufficient to discriminate between various realistic flux models.

Abstract: The measurement of the flux of muons produced in cosmic ray air showers is essential for the study of primary cosmic rays. Such measurements are important in extensive air shower detectors to assess the energy spectrum and the chemical composition of the cosmic ray flux, complementary to the information provided by fluorescence detectors. Detailed simulations of the cosmic ray air showers are carried out, using codes such as CORSIKA, to estimate the muon flux at sea level. These simulations are based on the choice of hadronic interaction models, for which improvements have been implemented in the post-LHC era. In this work, a deficit in simulations that use state-of-the-art QCD models with respect to the measurement deep underwater with the KM3NeT neutrino detectors is reported. The KM3NeT/ARCA and KM3NeT/ORCA neutrino telescopes are sensitive to TeV muons originating mostly from primary cosmic rays with energies around 10 TeV. The predictions of state-of-the-art QCD models show that the deficit with respect to the data is constant in zenith angle; no dependency on the water overburden is observed. The observed deficit at a depth of several kilometres is compatible with the deficit seen in the comparison of the simulations and measurements at sea level.