A new Monte Carlo Generator for Ultra-High Energy Cosmic Rays from the Local and Distant Universe

TL;DR

This paper introduces a novel Monte Carlo simulation method for ultra-high energy cosmic rays that combines distant source parameterizations with local universe tracking through realistic matter and magnetic field models, improving analysis and source modeling.

Contribution

The paper presents a new Monte Carlo generator that integrates large-scale structure simulations and magnetic field models for more accurate UHECR source and propagation analysis.

Findings

Generated detailed probability maps for UHECR observation directions.

Enhanced the ability to compare simulated data with actual measurements.

Provided a tool for optimizing UHECR analysis techniques.

Abstract

For the understanding of the origin and propagation of ultra-high energy cosmic rays (UHECR) we developed a new approach to simulating UHECRs from an arbitrary number of sources based on Monte Carlo technique. The method consists of a combination of three steps. For distant sources we apply commonly accepted parameterizations to calculate the contribution to the observed cosmic ray flux. For sources of the local universe we use forward tracking through realistic matter distributions and magnetic fields resulting from explicit simulations of large-scale structure formation. From the calculations and the forward tracking we generate maps of the probability to observe a particle with a given energy from a discrete direction. To account for deflections in the galactic field, these probability maps are transformed by matrices calculated from backtracking of antiparticles through field parameterizations. Based on the combined probability maps, Monte Carlo production of individual UHECR data is performed which are then used in comparisons with UHECR measurements. The simulated UHECR data serves for optimizing the analysis techniques used in UHECR measurements as well as for constraining the parameter space of the underlying source and magnetic field models.