The population of Galactic supernova remnants in the TeV range

Abstract

Context. Supernova remnants (SNRs) are likely to be significant sources of cosmic rays up to the knee of the local cosmic-ray (CR) spectrum. They produce gamma rays in the very-high-energy (VHE) (E > 0.1 TeV) range mainly via two mechanisms: hadronic interactions of accelerated protons with the interstellar medium and leptonic interactions of accelerated electrons with soft photons. Observations with current instruments have lead to the detection of about a dozen SNRs emitting VHE gamma rays and future instruments should significantly increase this number. Yet, the details of particle acceleration at SNRs and of the mechanisms producing VHE gamma-rays at SNRs remain poorly understood. Aims. We aim to study the population of SNRs detected in the TeV range and its properties and confront it to simulated samples in order to address fundamental questions concerning particle acceleration at SNR shocks. Such questions concern the spectrum of accelerated particles, the efficiency of particle acceleration, and the gamma-ray emission being dominated by hadronic or leptonic interactions. Methods. By means of Monte Carlo methods, we simulated the population of SNRs in the gamma-ray domain and confronted our simulations to the catalogue of sources from the High Energy Stereoscopic System (H.E.S.S.) Galactic Plane Survey (HGPS). Results. We systematically explored the parameter space defined in our model, including for example, the slope of accelerated particles α, the electron-to-proton ratio Kep, and the efficiency of particle acceleration ξ. In particular, we found possible sets of parameters for which ≳90% of Monte Carlo realisations are found to be in agreement with the HGPS. These parameters are typically found at 4.2 ≳ α ≳ 4.1, 10−5 ≲ Kep ≲ 10−4.5, and 0.03 ≲ ξ ≲ 0.1. We are able to strongly argue against some regions of the parameter space describing the population of Galactic SNRs in the TeV range, such as α ≲ 4.05, α ≳ 4.35, or Kep ≳ 10−3. Conclusions. Our model is so far able to explain the SNR population of the HGPS. Our approach, when confronted with the results of future systematic surveys, such as the Cherenkov Telescope Array Observatory, will help remove degeneracy from the solutions and to better understand particle acceleration at SNR shocks in the Galaxy.