Could dark matter’s hidden properties change the cosmic web?

What would the universe’s sprawling network of matter look like if dark matter behaved slightly differently? In a preprint, Rafael C. Nunes investigates two possibilities: dark matter exchanging energy with dark energy, and dark matter departing from its usual pressureless description. The simulations produce contrasting signatures, but interpreting them requires separating changes to dark matter from changes to the wider cosmological model.

The study does not detect either possibility in observations. It explores which features of simulated matter distributions and motions might help future measurements distinguish competing descriptions.

Two departures from the standard picture

The standard cosmological model, called Lambda cold dark matter, treats dark matter as pressureless and dark energy as a cosmological constant. Nunes modifies that picture in two separate ways.

In the first scenario, dark matter and dark energy exchange energy through a prescribed interaction. The study tests two negative values of the coupling parameter, −0.1 and −0.05, corresponding to energy flowing from dark matter to dark energy. This particular interaction does not introduce an additional gravitational-like “fifth force.” Instead, it changes the evolution of the dark components and the growth of matter concentrations.

The second scenario assigns dark matter a small nonzero equation-of-state parameter: a number describing pressure relative to energy density. The chosen value is 0.001, rather than the standard zero. The setup also specifies a squared rest-frame sound-speed parameter of 0.000001. These are particular modeling choices, not a test of every possible form of nonstandard dark matter.

Following matter and motion

Nunes extends the particle-mesh simulation code pmwd to handle the alternative models. Each simulation contains 512 cubed particles representing dark matter, with gravity calculated using a computational mesh. Two box sizes provide complementary views: a larger volume samples broader patterns, while a smaller volume gives finer spatial resolution.

The simulations start at redshift 49, when cosmic structure was much less developed, and evolve to the present day. Paired initial conditions with specially chosen fluctuations reduce the random variation that would otherwise obscure comparisons.

The analysis measures the matter power spectrum, which describes how strongly matter clusters on different spatial scales. It also measures velocity divergence—whether local motions converge or spread apart—and the statistical relationship between density and velocity divergence.

Those motion statistics matter because density alone does not fully describe structure formation. Once matter streams cross and flows become more complicated, the simple relationship between density and motion predicted by linear theory breaks down.

Large differences, but not a coupling-only experiment

For the interacting dark-sector models, the paper reports matter-power differences of roughly 50–100% at the largest scales sampled, with substantial, scale-dependent differences continuing into the nonlinear regime. These percentages describe differences in clustering power, not percentages of matter disappearing.

A crucial qualification is that the interacting models and the standard reference do not share identical baseline cosmological parameters. Their matter content, expansion parameters and fluctuation properties are chosen using observational constraints within the respective frameworks.

The comparison therefore asks how selected, observationally motivated models differ as a whole. It does not show that switching on the interaction alone causes a 50–100% change. The paper explicitly distinguishes its approach from an experiment that holds every other parameter fixed.

The velocity statistics also differ between models. Nunes argues that combining density and motion could provide complementary ways to distinguish them, although the study does not establish a survey’s eventual detection significance.

For the nonzero equation-of-state scenario, the baseline cosmological parameters are held fixed relative to the standard reference. Reported differences in both density and normalized velocity statistics remain within approximately 5%, with large-scale matter-power suppression of about 2–4%.

Where numerical precision matters

The standard-model simulations are checked against established predictions. The finer-resolution run agrees with nonlinear matter-power references to within about 4% out to the principal small-scale limit discussed. Velocity reconstruction is more sensitive to particle sampling, especially in sparsely populated regions, making the finer run the preferred basis for nonlinear velocity comparisons.

That accuracy context matters particularly for the smaller signals: agreement between simulations does not by itself establish that a few-percent effect will be distinguishable in observations.

The preprint calls for broader parameter studies, calibrated nonlinear predictions and the inclusion of massive neutrinos, whose effects could partly resemble the proposed signatures. Its contribution is a set of simulated targets for further investigation—not evidence that dark matter exchanges energy or has nonzero pressure.