SAETASS: Solver for Astroparticle Equation of Transport Analysis in Spherical Symmetry

In order to model astrophysical environments characterized by radial stratification, such as stellar winds, supernova remnants or expanding superbubbles; correctly understanding the transport of non-thermal particles in astrophysical plasmas is essential. SAETASS presents a novel, open-source numerical tool designed to solve the time-dependent transport equation for astroparticles in one-dimensional spherical symmetry.

Mathematical Formulation and Operator Splitting

The evolution of the nearly isotropic distribution function, f(t, r, p), is governed by the transport equation: ∂f/∂t + ∇ · (uwf) + ∂/∂p (˙pf) = ∇ · (D∇f) + Q. This equation is decomposed into four distinct physical operators:

1. Advection Operator (Ladv[f]): Defined as Ladv[f] = −1/r2 ∂/∂r (r2uwf), this term represents the bulk transport of particles with an advection velocity of uw, and it is a hyperbolic term prone to producing artificial oscillations or excessive numerical diffusion.

2. Diffusion Operator (Ldiff[f]): Defined as Ldiff[f] = 1/r2 ∂/∂r (r2D∇f), this operator accounts for stochastic scattering processes and is a parabolic term requiring implicit time-stepping.

3. Loss Operator (Lloss[f]): Defined as Lloss[f] = −∂/∂p(˙pf), this describes the variation of particle energy in momentum space, acting as a hyperbolic term analogous to advection but representing flow toward lower momenta.

4. Source Operator (Lsrc[f]): Defined simply as Lsrc[f] = Q, this is an algebraic term representing the injection of new particles into the system.

Numerical Discretization and Spatial/Momentum Treatment

The solver employs a conservative finite-volume framework to ensure exact particle conservation. The computational domain is defined in the (r, q) plane, where q ≡ log10 p, and utilizes cell-centered values fi,j(t) defined by the integral form of the transport equation. For spatial discretization in spherical symmetry, a two-dimensional domain is used: radial cells centered at ri and momentum cells centered at qj. The finite-volume formulation advances the cell-average fi,j(t) by computing fluxes through intercell faces. For hyperbolic operators like radial advection and momentum losses, the MUSCL-Hancock scheme is employed to achieve second-order accuracy in space by using a piecewise linear function reconstruction governed by limited slopes σk to prevent non-physical extrema.

Temporal Integration Strategy

The solver utilizes an operator-splitting architecture to manage the complexity of the coupled operators. The temporal evolution is advanced via a sequence of substeps, where each operator is treated individually over a time interval ∆tn. The paper implements two primary strategies:

1. Lie splitting [31] (sometimes called Lie-Trotter splitting), which is exactly the algorithm introduced, providing a light and easy-to-implement approach that advances operators sequentially.

2. Strang splitting [33], a second-order alternative that symmetrizes the composition and reduces the splitting error by applying substeps of different operators in a specific sequence.

Validation and Application

The reliability of SAETASS is rigorously validated through several tests, including:

1. Advection validation: Testing pure advection with constant radial velocity confirms the scheme's ability to handle geometrical dilution and recovers the expected asymptotic behavior, such as f(r) ∝ r−2.

2. Diffusion validation: Testing a simple constant-diffusion scenario verifies the correct handling of the Laplacian operator in spherical coordinates, showing convergence with increasing spatial resolution.

3. Loss validation: Testing a steady-state solution for losses (where α > β > 1) confirms the scheme correctly reproduces the balance between source injection and losses in momentum space.

4. Time dependence validation: The method of manufactured solutions (MMS) is used to test non-stationary conditions, such as an advection problem with a time-varying velocity field, demonstrating the solver's ability to capture pre-equilibrium temporal dynamics.

Software Architecture

The SAETASS package is designed with modularity in mind. The architecture reflects the operator splitting approach, separating the core engine (src/saetass) from physical utilities and solvers. Key components include:

/src/saetass/

/grid.py and state.py manage physical data structures (spatial and momentum coordinates). 2.

/splitting.py orchestrates the temporal substepping logic, defining the sequence of operators to be applied. 3.

**/solver.py acts as the main orchestrator, managing global advancement by delegating operations to specialized SubSolver instances (advection, diffusion, loss). 4.

As a fastidious researcher, my analysis focuses on how the mathematical rigor of SAETASS (Solver for Astroparticle Equation of Transport Analysis in Spherical Symmetry) translates into tangible, high-value improvements for AI systems, specifically those dealing with complex physical simulations and modeling.

Here are the specific improvements and capabilities an AI system could gain by integrating SAETASS methodologies:

),

1. Implementation of a robust, modular operator-splitting architecture (Advection/Loss/Diffusion).

2. Utilization of a conservative finite-volume framework ensuring exact particle conservation and stability against steep gradients.

3. Application of high-order MUSCLHancock schemes for shock-capturing hyperbolic advection and momentum losses.

4. Integration of an implicit, batched CrankNicolson algorithm for the diffusive operator to handle global coupling without restrictive explicit timestep constraints.

5. Adoption of a generalized logarithmic momentum variable transformation to efficiently manage high-dimensional phase space and resolve stiffness across orders of magnitude in momentum (e.g., from low-energy protons to PeV protons).

6. Implementation of a hierarchical, multi-level time stepping control strategy that dynamically adjusts timesteps based on splitting requirements, user substepping, and local CFL stability conditions.

The improved AI system could perform the following specific tasks:

1. Predictive Modeling of High-Energy Astrophysical Phenomena:

2. Accurate Simulation of Non-Steady State Environments:

3. Modeling Complex Particle Transport in Stratified Media:

4. Quantifying Pre-Equilibrium Dynamics in Dynamic Systems:

In order to model astrophysical environments characterized by radial stratification, such as supernova remnants or expanding superbubbles; correctly understanding the transport of non-thermal particles in astrophysical plasmas is essential. While large-scale Galactic propagation codes exist, they are often optimized for Cartesian or cylindrical geometries and lack the efficiency of one-dimensional spherically symmetric problems. In this work, we present SAETASS (Solver for Astroparticle Equation of Transport Analysis in Spherical Symmetry), a novel, open-source numerical tool designed to solve the time-dependent transport equation for astroparticles. The solver is built upon a conservative finite-volume framework that ensures exact particle conservation and numerical stability. To manage the interplay between diverse physical processes, SAETASS employs a modular operator-splitting architecture. Radial advection and continuous momentum losses are treated using a second-order, shock-capturing MUSCL-Hancock scheme, while the diffusive operator is integrated via an implicit, batched Crank-Nicolson algorithm. This approach allows for the robust handling of steep gradients, spatial discontinuities and regularity conditions at the origin. We rigorously validate the code through a suite of tests for pure advection, diffusion and losses. Finally, we demonstrate the solver's capabilities by modelling cosmic-ray proton transport in a real astrophysical scenario. Our results successfully recover established steady-state limits while revealing relevant pre-equilibrium temporal dynamics across Kolmogorov, Kraichnan and Bohm diffusion regimes. SAETASS provides the community with a lightweight, flexible tool for investigating particle acceleration and propagation in complex, radially dependent astrophysical environments.