Evidence for Hydrodynamic Attractors

Hydrodynamic attractors are well supported as model- and observable-specific reductions of far-from-equilibrium evolution. In Israel–Stewart hydrodynamics and relaxation-time kinetic theory, attraction of the selected Bjorken-flow observable can precede hydrodynamization; in the strongly coupled $\mathcal N=4$ supersymmetric Yang–Mills comparison, the two occur together. The cross-framework recurrence is significant, but it is not evidence for one universal curve, and heavy-ion data do not presently isolate an attractor independently of the full collision model.

Evidence cutoff. 11 August 2026. Reassess by 11 February 2027 or when a calculation tests less symmetric flow, realistic QCD, or an experimentally identifiable attractor observable.

Required background. Attractors and asymptotic gradients defines the reduction and its relation to divergent series; holographic hydrodynamization and attractors supplies the strong-coupling construction.

Helpful background. Kinetic-to-hydrodynamic maps fixes the weak-coupling interface; frame-invariant dissipative data identifies comparisons that survive field redefinitions; hydrodynamic poles and mode matching separates slow modes from transient ones.

The attractor statement

For a chosen microscopic or effective theory, symmetry class, and scaled observable $X(w)$, an attractor is a lower-dimensional curve or manifold toward which a family of solutions approaches as the scaled time $w$ evolves. In Bjorken expansion, a common choice is $w\sim\tau T(\tau)$ and $X$ is a pressure anisotropy or logarithmic derivative of the energy density.

Three notions must remain separate:

• hydrodynamization: nonhydrodynamic modes are sufficiently suppressed for a hydrodynamic constitutive description to predict selected observables;
• gradient convergence: a low-order derivative expansion approximates those observables;
• attractor approach: solutions lose sensitivity to some initial-condition directions.

They can occur at different times. The scope here is conformal or near-conformal expanding systems and bulk one-point observables studied in controlled effective, kinetic, or holographic models. It excludes a claim that every many-body system, every correlator, or realistic QCD has the same attractor.

Evidence matrix

Source and method	Relation to the bounded claim	Independence	Result and stated uncertainty	Main limitation
Heller and Spaliński, 2015, Müller–Israel–Stewart Bjorken flow and Borel analysis	establishes an attractor beyond a convergent gradient series	effective-theory calculation; not a microscopic test	many initial conditions approach one resurgent solution in the selected scaled observable	model coefficients and Bjorken symmetry control the curve; MIS is not QCD
Heller et al., 2018, relaxation-time kinetic theory	supports the phenomenon with quasiparticle dynamics	distinct microscopic picture from holography, though still conformal and highly symmetric	transient modes explain a divergent gradient expansion and approach to hydrodynamics	relaxation-time collision kernel is a closure, not full kinetic QCD
Kurkela et al., 2020, kinetic, causal-hydrodynamic, and holographic evolution	qualifies cross-model universality	compares several frameworks in common variables; theories are intentionally different	late-time attraction is shared, while early-time approach differs qualitatively between weak and strong coupling	common Bjorken-like reduction can make unlike dynamics appear closer than general flows would
Heller et al., 2022, flows beyond Bjorken symmetry	supports divergent gradient behavior outside the original geometry	extends the formal test rather than providing an independent experimental observation	divergence persists in more general nonlinear flows	divergence alone neither proves an attractor nor fixes its dimension
Successful heavy-ion hydrodynamic fits summarized by Soloviev, 2022	context only; supports early effective fluid behavior	experimental data are independent, but inference passes through initial-state, pre-equilibrium, transport, and hadronization models	collective observables are compatible with rapid hydrodynamization	no measured observable uniquely selects an attractor over alternative pre-equilibrium histories

A robust common core and model-dependent details

The robust claim is structural: transient information can decay while a low-dimensional relation among bulk observables becomes predictive, even when the formal gradient expansion diverges. Kinetic and holographic examples reduce the chance that this is an artifact of one closure. Their early-time behavior, nonhydrodynamic spectra, and preferred scaling variables nevertheless differ. Calling the curves “the attractor” without naming the theory and observable erases precisely the physics that could discriminate coupling regimes.

Apparent universality also depends on projection. Two high-dimensional trajectories may collapse in the pressure anisotropy while remaining distinguishable in higher moments or unequal-time correlators. Initialization on a one-parameter family can manufacture a narrow curve. A credible attractor claim therefore varies initial data in directions that the plotted observable can actually detect and reports transverse decay rates.

Experimental ceiling

Heavy-ion flow and spectra strongly support the usefulness of hydrodynamics, but they do not isolate the attractor stage. The same final observables depend on nuclear initial conditions, pre-equilibrium matching, viscosities, particlization, and hadronic rescattering. A posterior preference for one pre-equilibrium module would still not be an attractor observation unless an alternative without the proposed reduction is tested on held-out observables.

There is no direct laboratory measurement in this evidence set of the phase-space convergence defining the heavy-ion attractor. That negative-evidence statement is more informative than promoting general hydrodynamic success to a direct detection.

What would change the assessment

Evidence would strengthen with action-distinct simulations of QCD-like dynamics in less symmetric geometries, convergence in several independent observables, and quantitative predictions for an experimental correlation that competing pre-equilibrium models cannot mimic. It would weaken if attraction disappears when realistic longitudinal fluctuations, conserved charges, or nonconformal scales are included, or if the collapsed curve is shown to be a coordinate artifact.

Source selection

The finite set includes the original resurgent construction, a kinetic-theory realization, a direct weak/strong comparison, an extension beyond Bjorken flow, and a review used only to bound the experimental interpretation. Studies that rename ordinary late-time equilibration as an attractor without varying initial conditions were excluded.

Related research pages

• Schwinger–Keldysh, kinetic theory, and hydrodynamics for the hierarchy of real-time approximations.
• Hydrodynamic-attractor predictive domain for the wider normative question.
• Thermal and nonequilibrium field theory for neighboring programs and observables.

References

• Heller, Michał P., Aleksi Kurkela, Michał Spaliński, and Viktor Svensson. “Hydrodynamization in Kinetic Theory: Transient Modes and the Gradient Expansion.” Physical Review D 97 (2018): 091503. DOI.
• Heller, Michał P., Alexandre Serantes, Michał Spaliński, Viktor Svensson, and Benjamin Withers. “Hydrodynamic Gradient Expansion Diverges beyond Bjorken Flow.” Physical Review Letters 128 (2022): 122302. DOI.
• Heller, Michał P., and Michał Spaliński. “Hydrodynamics beyond the Gradient Expansion: Resurgence and Resummation.” Physical Review Letters 115 (2015): 072501. DOI.
• Kurkela, Aleksi, Wilke van der Schee, Urs Achim Wiedemann, and Bin Wu. “Early- and Late-Time Behavior of Attractors in Heavy-Ion Collisions.” Physical Review Letters 124 (2020): 102301. DOI.
• Soloviev, Alexander. “Hydrodynamic Attractors in Heavy Ion Collisions: A Review.” European Physical Journal C 82 (2022): 319. DOI.