🔭 astrophysics

RAMSES-MCR: A consistent multi-group treatment of cosmic rays physics in momentum-space with the RAMSES code

This paper introduces RAMSES-MCR, a consistent multi-group spectral method implemented in the RAMSES code that models cosmic ray protons and electrons in momentum space by evolving their energy and number densities alongside flux, thereby capturing complex physical processes like anisotropic diffusion and losses to accurately simulate their impact on astrophysical environments such as supernova remnants.

Original authors: Nimatou-Seydi Diallo, Yohan Dubois, Alexandre Marcowith, Joki Rosdahl, Benoît Commerçon

Published 2026-02-25

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Nimatou-Seydi Diallo, Yohan Dubois, Alexandre Marcowith, Joki Rosdahl, Benoît Commerçon

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the universe as a giant, chaotic kitchen. In this kitchen, there are invisible, super-fast particles called Cosmic Rays (CRs). They are like tiny, hyperactive ghosts that zip around at nearly the speed of light. They are everywhere, carrying a massive amount of energy, and they play a huge role in how galaxies cook up stars and gas.

For a long time, scientists trying to simulate this kitchen on computers had a problem. They treated these cosmic rays like a single, uniform soup. They assumed all the ghosts were the same speed and behaved the same way. But in reality, cosmic rays are a crowd of individuals: some are slow and heavy, others are light and zooming at top speed. Because they move at different speeds, they interact with the gas and magnetic fields in the universe very differently.

The Problem: The "One-Size-Fits-All" Mistake
Imagine trying to predict how a crowd of people moves through a hallway.

The Old Way (Grey Approximation): You assume everyone is walking at the same average speed. You tell the computer, "Everyone moves at 3 mph."
The Reality: Some people are sprinting (high-energy particles), some are jogging, and some are strolling (low-energy particles). The sprinters will get to the end of the hallway first, while the strollers get stuck in the middle. If you treat them all as one group, your prediction of where the crowd ends up will be wrong.

The Solution: RAMSES-MCR
The authors of this paper have built a new, super-smart tool called RAMSES-MCR. Think of this tool as a high-tech traffic controller for the cosmic ray crowd.

Instead of treating the crowd as one blob, RAMSES-MCR splits them into different groups (or "bins") based on how fast they are moving.

Group A: The slow walkers.
Group B: The joggers.
Group C: The sprinters.

The code tracks each group separately. It knows that the sprinters (high energy) can zip through magnetic fields easily, while the slow walkers (low energy) get stuck and lose their energy to the surrounding gas.

How It Works (The Magic Tricks)
The paper describes how this tool handles three main things:

The Traffic Jam (Diffusion): Cosmic rays don't just fly in straight lines; they bounce around magnetic fields like pinballs. The new tool calculates that the sprinters bounce further and faster than the joggers. This allows the simulation to show how high-energy particles escape a supernova explosion quickly, while low-energy ones linger around.
The Energy Transfer (Cooling): When cosmic rays crash into gas, they lose energy. The slow walkers lose energy very quickly (like a car hitting a wall), while the sprinters lose it slowly. RAMSES-MCR tracks exactly how much heat is dumped into the gas by each group. This is crucial because that heat can stop gas from collapsing into new stars or, conversely, push gas away to create galactic winds.
The Feedback Loop: The tool doesn't just watch the particles; it lets the particles push back on the gas. If the cosmic rays get too energetic, they can blow a bubble in the gas, changing the shape of the galaxy.

The Big Test: The Supernova Remnant
To prove their tool works, the authors simulated a Supernova Remnant (the expanding shell of gas after a star explodes).

What happened? They watched the explosion evolve over millions of years.
The Discovery: They found that the "sprinters" (high-energy particles) ran away from the explosion first, leaving a trail behind. The "joggers" stayed closer to the center, keeping the pressure up.
The Result: Because the tool could see these different groups, it showed that the explosion pushed the gas much harder and for a longer time than the old "one-size-fits-all" models predicted. It's like realizing that a group of runners can push a heavy cart further than a group of walkers, even if they start with the same total energy, because the runners keep pushing while the walkers stop.

Why Should You Care?
This isn't just about math; it's about understanding our universe.

Star Formation: Cosmic rays can heat up gas clouds, preventing them from collapsing into stars. If we get the physics wrong, we can't predict how many stars a galaxy will make.
Galactic Winds: Cosmic rays can act like a wind, blowing gas out of galaxies. This regulates the size of galaxies.
Observations: When we look at the sky with telescopes (like radio or gamma-ray telescopes), we see the light emitted by these particles. To understand what we see, we need to know exactly how these particles are moving and losing energy.

In a Nutshell
The authors built a new, more accurate "traffic controller" for the universe's most energetic particles. By sorting them into groups based on their speed, they can now simulate how galaxies evolve with much greater precision. It's a bit like upgrading from a blurry, black-and-white photo of a crowd to a high-definition, slow-motion video where you can see exactly what every single person is doing. This helps us understand the cosmic kitchen much better than ever before.

1. Problem Statement

Cosmic rays (CRs) are a dynamically significant component of the interstellar medium (ISM), influencing shock dynamics, galactic winds, and the thermochemistry of gas. However, standard numerical simulations often rely on a "grey" approximation, treating CRs as a single-fluid component with a single effective energy and transport coefficient. This approach fails to capture critical physical realities:

Energy-Dependent Transport: CR diffusion and streaming are intrinsically dependent on particle momentum (energy). High-energy particles diffuse faster than low-energy ones.
Energy-Dependent Losses: Radiative losses (Coulomb, hadronic, synchrotron) scale differently with energy, causing the CR spectrum to evolve spatially and temporally.
Feedback Loops: The coupling between CRs and gas dynamics (e.g., pressure support, heating via streaming instability) depends on the spectral shape of the CR distribution.

The lack of spectrally resolved models in adaptive mesh refinement (AMR) hydrodynamical codes limits the ability to accurately simulate CR feedback in complex astrophysical environments like supernova remnants (SNRs) and galactic winds.

2. Methodology: RAMSES-MCR

The authors introduce RAMSES-MCR, a new module for the AMR MHD code RAMSES. It implements a consistent multi-group spectral method based on the two-moment formalism (evolving both CR energy density and number density, along with their fluxes).

Key Numerical Features:

Two-Moment Formalism: The code evolves the CR energy density ( $e_{c,i}$ ) and number density ( $n_{c,i}$ ) for each momentum bin $i$ , coupled with their respective fluxes ( $F^e_{c,i}, F^n_{c,i}$ ). This results in a hyperbolic system with source terms, allowing for larger timesteps compared to explicit diffusion schemes.
Momentum-Space Discretization: The CR distribution function $f(p)$ is discretized into logarithmic momentum bins. Within each bin, the distribution is modeled as a broken power-law. The code tracks both the amplitude and the local spectral slope ( $q_i$ ) of the distribution.
Fluid-Frame Formulation: Unlike previous implementations (e.g., in RAMSES as described in Rosdahl et al. 2025), RAMSES-MCR solves the transport equations in the comoving fluid frame. This is crucial for:
- Ensuring consistency under adiabatic transformations.
- Correctly handling the coupling between CRs and gas momentum/energy.
- Improving numerical stability when using a reduced speed of light ( $\tilde{c}$ ).
Physical Processes Modeled:
- Advection: Coupling with gas velocity.
- Diffusion: Anisotropic diffusion along magnetic field lines, with momentum-dependent coefficients ( $\kappa \propto p^\delta$ ). The code includes spectral correction factors to account for the differential diffusion of number vs. energy density within a bin.
- Streaming: CR streaming instability driving Alfvén waves, which damp and heat the gas.
- Losses: Coulomb, ionization, and hadronic losses (for protons); synchrotron and inverse Compton (for electrons, extensible).
- Injection: Power-law injection at shocks.
Operator Splitting: The solver uses an operator-splitting approach:
1. Advection and hyperbolic transport (fluxes) are solved explicitly/implicitly.
2. Source terms (losses/gains) and momentum-space transport (spectral evolution) are solved separately using a spectral timestep.

3. Key Contributions

First Consistent Multi-Group Solver in RAMSES: Extends the grey two-moment method to a full spectral treatment, allowing for the simultaneous evolution of CR energy and number densities across multiple momentum bins.
Fluid-Frame Implementation: Derives and implements the two-moment equations in the fluid frame, ensuring rigorous conservation properties during adiabatic compression/expansion and improving stability for reduced light speeds.
Spectral Correction for Diffusion: Introduces analytically derived correction factors for diffusion coefficients within bins, ensuring that the spectral slope evolves correctly due to momentum-dependent diffusion, rather than preserving the initial slope artificially.
Extensibility: The framework is designed to handle multiple CR species (protons and electrons) and can be extended to other astrophysical contexts.

4. Results and Validation

The authors validated RAMSES-MCR through a suite of idealized tests and a complex astrophysical application:

Radiative Cooling: The code accurately reproduces the analytical evolution of CR spectra under Coulomb and hadronic losses, capturing the "break" in the spectrum at low energies and the preservation of slopes at high energies.
Steady-State Injection: In a system with continuous injection and cooling, the code converges to the analytical steady-state solution, with errors scaling linearly with the number of momentum bins ( $N_c$ ).
Adiabatic Compression: Tests confirmed that the semi-implicit time integration scheme minimizes errors in preserving the spectral slope during gas compression/expansion.
Diffusion Tests:
- 1D Diffusion: Demonstrated that using spectrally averaged diffusion coefficients correctly captures the deformation of the CR distribution function (changing inner slopes), whereas a simple bin-centered approximation fails to capture this spectral evolution.
- 2D Anisotropic Diffusion: Successfully simulated CR diffusion in a circular magnetic field, showing that higher-momentum CRs diffuse faster and further along field lines.
Shock Tube with Streaming: Validated the interaction between CRs and gas in a shock tube. The multi-group approach reproduced the expected density depletion and velocity jumps caused by streaming instability, consistent with grey models but with the added capability of tracking spectral changes.
Supernova Remnant (SNR) Application:
- Simulated a 3D SNR expansion with anisotropic diffusion and energy losses.
- Momentum-Dependent Escape: High-energy CRs escaped the remnant rapidly, while low-energy CRs remained confined, sustaining internal pressure.
- Momentum Boost: The inclusion of CRs significantly enhanced the total radial momentum of the SNR compared to a pure hydrodynamic run. The enhancement was strongest for lower diffusion coefficients (where CRs are more confined), reaching a 135% increase in momentum at late times ( $t \approx 0.9$ Myr).
- Snowplough Phase: CRs delayed the transition to the momentum-conserving snowplough phase, maintaining pressure support longer than thermal gas alone.

5. Significance

RAMSES-MCR represents a major advancement in computational astrophysics by providing a robust, spectrally resolved tool for studying CR feedback. Its significance lies in:

Physical Realism: It moves beyond the "grey" approximation, allowing researchers to study how spectral evolution (e.g., hardening/softening of spectra) affects gas dynamics and observational signatures (e.g., $\gamma$ -ray emission).
Galaxy Evolution: By accurately modeling CR transport and losses, the code enables more reliable predictions of galactic wind mass-loading factors and the regulation of star formation, which are sensitive to the energy-dependent nature of CR pressure.
Observational Tracers: The ability to track specific momentum bins allows for direct comparison with multi-wavelength observations (radio synchrotron, $\gamma$ -rays) that are sensitive to specific energy ranges of CR electrons and protons.
Numerical Stability: The fluid-frame formulation and semi-implicit schemes ensure the method is stable and efficient for large-scale, high-resolution cosmological simulations.

In conclusion, RAMSES-MCR provides a consistent framework to investigate the complex, energy-dependent interplay between cosmic rays and the interstellar medium, bridging the gap between idealized spectral models and full 3D magnetohydrodynamic simulations.