🔭 astrophysics

What would it take for dark matter to be literally warm?

The paper argues that achieving a literally "warm" dark matter scenario consistent with current mass constraints requires highly coincidental and unnatural initial conditions, suggesting that the field should move beyond standard thermal relic benchmarks toward more expressive parameterizations and simulation-based methods to better interpret upcoming data.

Original authors: Katelin Schutz

Published 2026-08-27

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Katelin Schutz

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

The universe is filled with a mysterious substance called dark matter. We cannot see it, and it does not emit light, but we know it is there because its gravity holds galaxies together. For decades, the leading theory has been that this dark matter is "cold," meaning the particles that make it up move very slowly. This idea works incredibly well to explain the large-scale structure of the cosmos, from the way galaxies cluster to the vast cosmic web connecting them. However, when scientists zoom in to look at the smallest scales, such as the tiny dwarf galaxies orbiting our own Milky Way, the cold theory predicts there should be far more of these small galaxies than we actually observe. This discrepancy has led researchers to consider an alternative: what if dark matter is "warm"? In this scenario, the particles would move faster, washing out the formation of the tiniest structures and leaving behind a universe that matches our telescopes better.

For years, the concept of warm dark matter has served as a useful tool for scientists, acting as a simple benchmark to test their theories against observations. The idea was straightforward: imagine a particle that was once hot and moving at the speed of light in the early universe, but cooled down as the cosmos expanded. By adjusting the mass of this particle, scientists could dial the amount of small-scale structure up or down, creating a smooth bridge between the hot, fast-moving particles that erase too much structure and the cold, slow ones that create too much. This single parameter, the mass of the particle, became the standard way to report limits on the nature of dark matter. Recent observations, including data from the James Webb Space Telescope and deep surveys of the early universe, have pushed the lower limit for this mass to around 10,000 electron volts, or 10 keV. This number suggests that if dark matter is indeed a warm particle, it must be significantly heavier than a neutrino but much lighter than the atoms that make up our world.

Katelin Schutz, a physicist at McGill University, recently asked a deceptively simple question: what would it actually take for dark matter to be warm in this literal sense? She examined whether a universe filled with a thermal relic—a particle that was once in perfect thermal equilibrium with normal matter and then decoupled while still moving at relativistic speeds—could actually exist without breaking the laws of physics or requiring impossible coincidences. Her investigation reveals that for dark matter to be literally warm, the early universe would have needed to be filled with an enormous number of invisible, light particles. Specifically, to satisfy the current mass constraints, there would have to be roughly 10,000 different types of these light particles interacting with the visible universe at the moment the dark matter stopped interacting with it. This is far more than the roughly 100 particles known in the Standard Model of particle physics, and it is far more than even the most generous extensions of that model, such as supersymmetry, could provide.

The paper explores three ways this scenario might be forced to work, and finds each one deeply problematic. The first possibility is that the early universe contained a hidden sector of thousands of new particles. The second is that a period of early matter domination, driven by a heavy, long-lived particle, diluted the dark matter density after it decoupled. The third is that the dark sector was never in contact with our visible universe and was simply born colder through a process called asymmetric reheating. Schutz argues that all three options rely on strong, unexplained coincidences. In the second case, for instance, the timing of the decay of the heavy particle and the temperature at which dark matter decouples must be tuned with extreme precision to produce the correct amount of dark matter we see today. These two physical processes have no inherent connection, yet they must align perfectly to fit the warm dark matter model. Furthermore, if such a dilution event occurred, it would change the way gravity worked on small scales in a way that contradicts the standard mathematical formulas currently used to analyze the data.

The author concludes that the literal interpretation of warm dark matter is likely not the correct description of reality. Instead, the constraints we place on the "warm dark matter mass" are actually constraining a much broader and more complex landscape of theories. Many modern models of dark matter, such as those involving sterile neutrinos, self-interacting particles, or fuzzy dark matter, can suppress small-scale structures in ways that look similar to warm dark matter but have different underlying physics. These models often produce unique signatures, such as specific patterns of waves or steps in the distribution of matter, which a simple warm dark mass cannot capture. By continuing to use a single number to describe these diverse possibilities, scientists risk missing the true nature of the dark sector.

Schutz advocates for a shift in how the field reports its findings. Rather than squeezing all new data into a single parameter, researchers should move toward more expressive methods that can describe the shape of the suppression in the matter distribution. This includes using advanced computer simulations and machine learning techniques to compare observational data directly against a wide variety of specific models. These tools can detect subtle differences in how structures form, allowing scientists to distinguish between a warm particle, a self-interacting fluid, or a quantum wave. The paper suggests that the era of using a simple benchmark is ending. As our telescopes and simulations become more powerful, they will be able to see not just whether small structures are missing, but exactly how they are missing. This detailed view will be essential for uncovering the true identity of dark matter, moving us from a simple approximation to a precise understanding of the invisible universe.

Problem Statement
The paper addresses the theoretical viability of "Warm Dark Matter" (WDM) interpreted literally as a thermal relic that decoupled from the Standard Model (SM) thermal bath while relativistic. While WDM has served as a valuable one-parameter benchmark for constraining small-scale structure formation over the past decades, current observational constraints push the WDM mass ( $m_{\text{WDM}}$ ) toward the $\sim 10$ keV scale. The author investigates whether a self-consistent cosmological and particle physics framework exists that can produce a thermal relic with these specific mass and temperature properties without relying on unexplained coincidences between physically unrelated parameters.

Methodology and Theoretical Analysis
The author analyzes the relic abundance equation for a fermionic thermal relic decoupling while relativistic. By equating the required relic density ( $\Omega_{\text{DM}}h^2 \approx 0.12$ ) with the mass constraints ( $m_X \sim 10$ keV), the paper derives the necessary temperature ratio ( $T_X/T_\gamma$ ) between the dark sector and the SM. The analysis then evaluates three distinct mechanisms that could theoretically satisfy these conditions:

Entropy Transfer via Relativistic Degrees of Freedom: The paper calculates the number of relativistic degrees of freedom ( $g_*(T_D)$ ) required at the time of decoupling to achieve the necessary temperature ratio via standard entropy transfer (analogous to the neutrino-photon decoupling).
Entropy Dilution via Early Matter Domination (EMDE): The author examines scenarios where a massive, long-lived particle ( $\phi$ ) dominates the energy density after the dark matter decouples, subsequently decaying and injecting entropy to dilute the relic density. This requires analyzing the interplay between the decoupling temperature ( $T_D$ ), the domination temperature ( $T_{\text{dom}}$ ), and the reheating temperature ( $T_{\text{RH}}$ ).
Asymmetric Reheating of a Decoupled Dark Sector: The paper considers a dark sector that never thermalized with the SM but was born colder via asymmetric reheating, where the final temperature ratio is determined by the initial reheating conditions and the entropy history of the dark sector.

The author also reviews the phenomenological implications of these scenarios, specifically comparing the transfer functions derived from these non-standard expansion histories against the standard WDM fitting functions (e.g., Bode et al. 2001) used in current observational analyses.

Key Results and Findings
The analysis concludes that a literally warm thermal relic is theoretically possible only under conditions of "strong coincidences" involving unconnected physical parameters:

Relativistic Degrees of Freedom: To satisfy current mass constraints ( $\sim 10$ keV) via standard entropy transfer, the early universe would require $\sim 10^4$ relativistic degrees of freedom at decoupling. This is two orders of magnitude beyond the Standard Model and its plausible extensions (such as SUSY or twin Higgs models), which typically offer only $\mathcal{O}(100)$ new degrees of freedom.
Early Matter Domination: While an EMDE can dilute the relic density without requiring excessive degrees of freedom, it necessitates a precise tuning of the duration of the matter-dominated era. The relationship between the decoupling temperature, the mass/lifetime of the decaying particle, and the reheating temperature must be fine-tuned to reproduce the observed abundance. Furthermore, the EMDE alters the expansion history, meaning the perturbations evolve through a matter-dominated epoch rather than the radiation-dominated epoch assumed in standard WDM transfer functions. This results in a transfer function that differs from the standard WDM benchmark on sub-horizon scales, rendering standard constraints potentially inconsistent with the actual cosmology.
Asymmetric Reheating: This scenario requires the initial temperature asymmetry and the entropy evolution of two independent sectors to conspire to produce the exact temperature ratio required for a specific WDM mass, despite having no intrinsic physical connection to the particle mass itself.

The paper further notes that the standard WDM fitting functions are already a source of systematic error at the current constraint frontier, as exact Boltzmann calculations for thermal relics at these masses yield "colder" transfer functions than the widely used fits.

Significance and Recommendations
The paper argues that the WDM benchmark, while historically useful for calibrating simulations and developing observational probes, is no longer a self-consistent literal hypothesis for dark matter. The constraints derived under the WDM assumption are likely constraining a broad class of self-consistent models (e.g., sterile neutrinos, interacting dark sectors, fuzzy DM, freeze-in production) that share the feature of small-scale suppression but possess distinct microphysical origins and transfer function shapes.

The author advocates for a transition away from the one-parameter WDM mass benchmark toward more expressive frameworks:

Phenomenological Parameterization: Adopting multi-parameter transfer functions (e.g., varying the sharpness and slope of the suppression) to capture the shape of the cutoff rather than just its scale.
Physical Parameterization: Using frameworks like the Effective Theory of Structure Formation (ETHOS) to map observables to effective dark sector parameters.
Simulation-Based Inference (SBI): Utilizing machine learning and emulators trained on diverse simulation suites to constrain specific dark matter models directly from data without relying on pre-defined functional forms.

The paper concludes that as observational data (from Lyman- $\alpha$ forests, strong lensing, and satellite counts) becomes more precise, the field must move beyond the "binary" question of whether suppression exists to a richer analysis of what the suppression looks like, as the shape of the power spectrum encodes the unique microphysics of the dark sector.

More like this