⚛️ general relativity

Beyond dust in local cosmography: covariant kinetic theory

This paper develops a covariant kinetic framework for collisionless cold dark matter that extends beyond the dust approximation by incorporating velocity dispersion to model stream crossing, thereby generating intrinsic 4-acceleration and vorticity that modify local cosmographic observables like the Hubble dipole and deceleration multipoles.

Original authors: Chris Clarkson, Roy Maartens

Published 2026-10-06

📖 7 min read🧠 Deep dive

CC BY 4.0

Original authors: Chris Clarkson, Roy Maartens

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

In the vast, expanding universe, astronomers have long relied on a simple mental model to describe how matter moves: they imagine the invisible scaffolding of the cosmos as a smooth, flowing river of dust. In this picture, every tiny patch of space has a single, clear speed and direction, moving along the curves of gravity without ever bumping into itself or changing course. This "dust" model has been the workhorse of cosmology for decades, allowing scientists to map the expansion of the universe and measure how fast galaxies are moving away from us. However, this simplicity is an approximation. In reality, the universe is clumpy. Gravity pulls matter together into dense clusters and halos, causing the paths of individual particles to cross and crisscross like a chaotic swarm of bees. When these paths cross, the idea of a single, smooth flow breaks down. If the universe is not a smooth river but a turbulent, multi-streaming cloud, then our maps of its expansion might be slightly off, particularly in the local neighborhood of our own galaxy.

This is the precise problem that Chris Clarkson and Roy Maartens set out to solve. They have developed a new, more realistic way to describe how matter moves in the local universe, one that accounts for the fact that particles can have different speeds in the same small patch of space. Their work moves beyond the old "dust" model to treat the universe as a collisionless gas, where particles interact only through gravity but can stream past one another. By doing so, they have uncovered a hidden force that arises naturally from this chaos: a subtle acceleration that pushes the average flow of matter off its straight path. This discovery provides a new tool for testing whether the strange discrepancies seen in recent measurements of the local universe are due to a flaw in our maps or simply a failure of the old, overly simple model.

For years, cosmographers have tried to measure the expansion of the universe without assuming a specific shape or background structure, a method known as "covariant cosmography." This approach is powerful because it lets the data speak for itself, rather than forcing observations to fit a pre-existing theory. However, when researchers applied this method to the local universe, specifically looking at distances and redshifts for galaxies within about 40 million light-years, the model began to struggle. The data did not fit the predictions of the smooth dust model. Some scientists suspected this was a sign that the dust model itself was breaking down on these small scales, but without a better theory of how matter actually behaves in these crowded regions, they could not be sure. They needed a way to describe matter that was more complex than dust but still simple enough to calculate.

Clarkson and Maartens answered this need by building a mathematical framework based on kinetic theory, which describes the behavior of a gas of particles. Instead of assuming every particle moves at the same speed, they allowed for a spread of speeds, known as velocity dispersion. Imagine a crowd of people walking down a hallway; in the old dust model, everyone walks at the exact same speed in the same direction. In the new model, some walk faster, some slower, and some drift slightly to the side, even though they are all moving generally forward. When you average this crowd to find the "center of mass" motion, the differences in speed create a kind of internal pressure. In a gas, this pressure is usually thermal, caused by heat. But in the cold, dark matter of the universe, this pressure comes purely from the random motions of the particles as they cross paths.

The researchers found that this internal pressure does something surprising: it creates a force that accelerates the average flow of matter. In the old dust model, matter follows the curves of gravity perfectly, like a marble rolling down a hill. In their new model, the internal jostling of the particles pushes the average flow slightly off that perfect path. This acceleration is not caused by an external push but is an intrinsic property of the matter itself, arising from the way the different speeds of the particles interact. This is a crucial distinction. It means that the "Hubble dipole"—a specific pattern in how fast the universe appears to expand in different directions—is not just a result of our own motion through space, but also a signature of the internal stress of the matter itself.

Furthermore, this acceleration has a second, even more subtle effect. Because the force is generated by the uneven distribution of speeds, it can twist the flow of matter, creating a swirling motion known as vorticity. In the smooth dust model, such swirling is impossible to generate from a calm start. But in this new kinetic framework, the very act of particles crossing paths and dispersing their speeds can spin up the cosmic flow. The authors derived exact equations showing how this happens, proving that the universe's local expansion is not just stretching and speeding up, but also potentially twisting in ways that were previously invisible to the standard model.

The significance of this work lies in its ability to connect the microscopic behavior of dark matter to the macroscopic maps we use to measure the cosmos. The researchers constructed a self-consistent system where the pressure and the acceleration are not just guessed at, but calculated directly from the distribution of particle speeds. They showed that if you stop at a certain level of complexity—considering the average speed and the spread of speeds, but ignoring the most extreme outliers—you get a complete and accurate picture of the matter's behavior. This "truncation" allows them to close the equations and make predictions without needing to assume a specific shape for the universe or rely on approximations that only work for small disturbances.

This new framework offers a direct way to test the breakdown of the dust model. By looking for the specific signatures of this intrinsic acceleration and the resulting vorticity in local distance-redshift data, astronomers can determine if the strange signals they see are indeed caused by the crossing of particle streams. The authors note that their companion paper applies these ideas to a specific, perturbed version of the universe to predict the size of these signals, suggesting that the effects are large enough to be detected with current and upcoming surveys. If the data confirms these predictions, it would mean that the "Hubble tension"—the disagreement between different measurements of the universe's expansion rate—might be partially explained by the fact that our local neighborhood is not a smooth river of dust, but a complex, multi-streaming cloud of particles.

The work does not claim to have solved the mystery of the universe's expansion, nor does it suggest that the standard model of cosmology is wrong. Instead, it provides a more precise tool for understanding the local universe. It clarifies that the "dust" model is a useful approximation for large scales but fails when the universe becomes crowded enough for particle paths to cross. By replacing that approximation with a kinetic description, the authors have identified exactly what quantities need to be measured to test this idea. They have shown that the acceleration of the matter flow is a real, physical effect generated by the velocity dispersion of dark matter, and that this effect leaves a distinct fingerprint on the way we observe the universe's expansion.

Ultimately, this paper represents a shift from assuming the universe is simple to describing it as it is: a complex system where the collective behavior of countless particles creates forces that shape the cosmos. The researchers have built a bridge between the chaotic, multi-streaming reality of dark matter and the smooth, large-scale maps we use to navigate the universe. Their findings suggest that the next generation of cosmological surveys will not just be measuring how fast the universe is expanding, but also probing the internal dynamics of the invisible matter that drives that expansion. If the data aligns with their predictions, it will confirm that the local universe is far more dynamic and interconnected than the simple dust model ever allowed us to see.

Technical Summary: Beyond Dust in Local Cosmography: Covariant Kinetic Theory

Problem Statement
Standard cosmographic analyses of the local Universe typically model cold dark matter (CDM) as pressure-free dust. This assumption implies that the matter congruence is geodesic, thereby removing any intrinsic dipole from the local Hubble flow. However, recent covariant cosmographic tests [1–4] revealed that the dust model fails to consistently fit Cosmicflows-4 data at redshifts $z \lesssim 0.01$ (radial scales $\lesssim 40$ Mpc). This breakdown is interpreted as a signal that the single-stream dust approximation is invalid on small scales where gravitational collapse leads to "stream crossing" (or shell crossing), resulting in multiple particle velocities within a single volume element. While the dust model assigns a unique velocity to each spacetime event, a collisionless gas accommodates multi-streaming via velocity dispersion in phase space, generating isotropic and anisotropic stresses. The challenge is to develop a model-independent, nonperturbative framework that incorporates these kinetic effects without assuming a Friedmann-Lemaître-Robertson-Walker (FLRW) background or specific geometrical symmetries.

Methodology
The authors construct a fully covariant kinetic framework for a massive, collisionless gas interacting only via gravity. The approach proceeds as follows:

Covariant Kinetic Theory: Starting from the exact massive Vlasov equation, the distribution function $f(x^a, p^b)$ is expanded in tensor multipoles $F_{A_\ell}$ in phase space.
Energy-Integrated Hierarchies: The authors derive an infinite hierarchy of energy-integrated moments in spacetime. Unlike massless cases, massive particles require a two-index hierarchy $(\ell, n)$ to account for independent towers of velocity weights, where $\ell$ denotes the angular multipole order and $n$ the velocity weight.
Velocity Ordering and Truncation: The hierarchy is organized by velocity weight. For cold dark matter, higher-order moments are suppressed by powers of particle speed $v$ . The authors impose a self-consistent truncation at the quadrupole (second velocity order, $O(v^2\rho)$ ). This closure retains the energy density $\rho$ , isotropic pressure $P$ , and anisotropic pressure $\pi_{ab}$ , while neglecting higher-order moments (e.g., the octupole $\Pi_{abc}$ ) that would otherwise source the lower moments.
Matter Frame Definition: A specific "matter frame" 4-velocity $u^a$ is defined such that there is no energy flux ( $q_a = 0$ ). In this frame, the momentum conservation equation transforms from an evolution equation for the heat flux into a constraint linking pressure gradients to the 4-acceleration $A^a$ .
Cosmographic Propagation: The resulting non-geodesic matter flow is propagated into covariant local cosmography. The authors derive expressions for the Hubble rate and deceleration parameter as seen by an observer, accounting for the intrinsic dipole generated by the matter-frame 4-acceleration.

Key Contributions and Results

Exact Nonlinear 4-Acceleration: The paper derives an exact, nonlinear expression for the matter-frame 4-acceleration $A^a$ (Equation 4.9) in terms of the energy density, isotropic pressure, and anisotropic pressure gradients. This acceleration arises purely from velocity dispersion (kinetic stress) and vanishes in the dust limit.
Vorticity Generation: The authors demonstrate that velocity dispersion sources vorticity. Since the 4-acceleration is non-zero, its curl acts as a source term in the standard covariant vorticity evolution equation (Equation 4.11). The paper provides the specific nonlinear source terms for vorticity evolution at $O(v^2\rho)$ , showing that vorticity is generated by the curl of the divergence of anisotropic pressure and the antisymmetric product of density and pressure gradients.
Covariant Cosmographic Completion: The framework provides a "kinetic completion" to dust cosmography.
- The intrinsic Hubble dipole is identified directly with the matter-frame 4-acceleration.
- The 4-acceleration enters the deceleration multipoles (dipole, quadrupole, etc.) of the distance-redshift relation.
- The authors derive the transformation of these multipoles to a boosted observer frame, distinguishing the physical non-geodesicity of the matter flow from the kinematic dipole caused by observer motion.
Linearized Analysis: Extending the analysis to linear perturbations about an FLRW background, the authors show how the 4-acceleration evolves self-consistently (Equation 5.22) and how it modifies the deceleration multipoles. They note that at linear order, the anisotropic pressure is degenerate with the tidal field (Weyl tensor) in the quadrupole, making them indistinguishable via distance-redshift measurements alone at this order.

Significance and Claims
The paper claims to provide the first covariant, nonperturbative kinetic framework that connects the microscopic dynamics of a collisionless gas (via the Vlasov hierarchy) to macroscopic local cosmography without assuming a background spacetime.

Physical Origin of the Dipole: The work establishes a physical origin for the intrinsic Hubble dipole, attributing it to kinetic stresses (velocity dispersion) rather than arbitrary frame choices.
Testing the Dust Breakdown: The construction offers a specific theoretical tool to test whether the observed breakdown of the dust model in local data ( $z \lesssim 0.01$ ) can be explained by velocity dispersion effects.
Self-Consistency: Unlike previous phenomenological effective fluid models or effective field theory approaches that require matching to simulations or assume an FLRW background, this approach derives the fluid equations and acceleration self-consistently from the Vlasov equation within a controlled velocity-order truncation.

The authors note that the second-moment closure is an approximation that may fail if higher velocity cumulants become significant on highly nonlinear scales. They indicate that a companion paper [40] specializes this hierarchy to perturbed FLRW spacetime to predict the specific size of the resulting cosmographic signals and to calibrate the mean dispersion using N-body simulations.

More like this