Physicists Reverse Turbulent Energy Flow, Challenging Kolmogorov's 1941 Framework

A research team led by the University of Pittsburgh has demonstrated that the direction of energy flow inside a turbulent fluid is not a fixed property of the flow's geometry -- it can be actively controlled. The finding, published this week in the journal Science Advances, presses against a framework that has organized fluid dynamics research since 1941.

For more than eighty years, the standard picture, rooted in Andrey Kolmogorov's cascade model, held that energy in three-dimensional turbulence moves in one direction: from large eddies down to progressively smaller ones, where it dissipates as heat. In thin, quasi-two-dimensional layers, the cascade was thought to run the other way. Either way, the direction was considered a consequence of the flow's dimensionality -- not something an experimenter could tune.

The new work pulls that assumption apart. According to reporting based on the paper reviewed in Science Advances, the Pitt team, working with collaborators at the University of Turin in Italy, found that what actually sets the direction of energy transfer is the geometric alignment between two tensor quantities at every point in the fluid: the stress acting on the fluid and the deformation the fluid undergoes in response. Alter that alignment, and you can redirect the energy cascade regardless of whether the flow is two-dimensional or three-dimensional.

Lei Fang, assistant professor in civil and environmental engineering at Pitt's Swanson School of Engineering, led the study alongside PhD student Xinyu Si and Turin collaborators Filippo De Lillo and Guido Boffetta. The formal title of the paper is "Manipulating the direction of turbulent energy flux via tensor geometry in a two-dimensional flow."

The practical scope of what was demonstrated matters here. The experiments were conducted in a controlled, electromagnetically driven thin-layer setup -- essentially a quasi-two-dimensional system. As the Martin Cid Magazine summary of the paper noted, the demonstration has not yet been replicated in fully three-dimensional, high-energy flows, and the gap between a laboratory tray and an open ocean current is substantial. Whether tensor-alignment control survives in thicker, more energetic regimes remains an open question; the team's stated next step is to probe exactly that boundary.

Still, the conceptual shift is not trivial. The Kolmogorov cascade has underpinned weather forecasting models, climate simulations, and engineering designs for turbomachinery and biomedical devices for decades. A mechanism that can redirect rather than merely observe that cascade would, if it generalizes, have implications for coastal current management, drug delivery in the bloodstream, and climate model parameterization -- all areas cited by the authors in the Science Advances paper.

The study adds to a small but growing body of work questioning whether Kolmogorov's scaling laws fully describe turbulence in every regime. A 2023 paper in Phys.org coverage noted that those principles were already recognized as incompletely describing certain turbulent flows. The new work goes further, proposing a geometric handle on directional control rather than just identifying where the old model breaks down.

As always with a single experimental demonstration in a controlled geometry, independent replication across different fluids and Reynolds numbers will be the necessary next test before the finding reshapes any downstream application. The underlying mechanism, however, is now on the table in a peer-reviewed venue, and that makes it worth watching.

Sources cited:
- Science Advances (via ScienceDaily) (https://www.sciencedaily.com/releases/2026/06/260602021655.htm)
- Mirage News (https://www.miragenews.com/new-discovery-shatters-80-year-turbulence-theory-1685740/)
- Martin Cid Magazine (https://www.martincid.com/science/turbulence-energy-flux-reversed-tensor-geometry-2/)
- SciTechDaily (https://scitechdaily.com/new-discovery-challenges-80-year-old-theory-about-turbulence/)