When Chaos Meets Quantum: The Algorithm That Learned to Simulate Turbulence
26 Apr 2026, Yanjiang
What if the most chaotic things in the universe — the churn of a hurricane, the swirl of a vortex, the unpredictable dance of molecules in a storm — could be tamed not by a bigger classical computer, but by a quantum one? That’s the question posed by a new preprint (arXiv:2507.06198) from a team at IBM Quantum: Sergey Bravyi, Robert Manson-Sawko, Mykhaylo Zayats, and corresponding author Sergiy Zhuk. Their work proposes something that sounds almost like a contradiction in terms: a quantum algorithm for simulating nonlinear, noisy, classical dynamics.
This is not a small ambition. Nonlinear systems — the kind where effects don’t scale linearly with causes — are famously hard to simulate. They are the reason weather forecasts fail beyond a week, the reason turbulence remains one of the great unsolved problems in physics, and the reason that, for all our computational power, we still cannot predict the flow of air over an airplane wing with perfect accuracy. Classical computers struggle with them because the number of variables grows explosively, and because small errors in initial conditions — the famous “butterfly effect” — amplify exponentially.
Enter the quantum algorithm.
The Architecture of a Quantum Storm
The team’s approach is built on a deceptively simple observation: many nonlinear systems of practical interest can be described by a specific mathematical structure — a system of stochastic dissipative differential equations with a quadratic nonlinearity. That mouthful of terms breaks down into three key ingredients:
Stochastic means the system includes noise — randomness baked into the equations themselves. Real-world systems are never perfectly isolated; they are buffeted by thermal fluctuations, measurement uncertainties, and environmental disturbances. A useful simulation must account for this.
Dissipative means energy leaks out of the system over time. Think of a spinning top that gradually slows down, or a storm that dissipates after the fuel supply runs out. Dissipation provides a natural damping mechanism — and, as it turns out, a crucial computational advantage.
Quadratic nonlinearity is the mathematical heart of the beast. It means that the equations contain terms where variables multiply each other — like x·y rather than just x or y. This is the simplest form of nonlinearity that still captures genuinely complex behavior. It is also, mathematically, the hardest to handle.
Here’s where the quantum magic enters. The team shows that this class of nonlinear systems can be encoded into a quantum circuit using a technique called block encoding — a way of representing a matrix as a sub-block of a larger unitary operator that a quantum computer can implement. The key insight is that the quadratic nonlinearity, which would normally require exponential resources to simulate classically, can be compressed into a quantum representation that scales only polynomially with the number of variables.
The Cost of Chaos
But there is a catch — and it is a significant one. The algorithm’s runtime scales exponentially with a parameter that quantifies the inverse relative error in the initial conditions. In plain language: if you don’t know the starting state of your system very precisely, the quantum advantage evaporates.
This is not a bug; it is a fundamental feature of nonlinear dynamics. The butterfly effect is not something that clever algorithms can wish away. The team’s result makes this limitation explicit: the exponential sensitivity to initial conditions becomes an exponential cost in simulation time.
Yet within that constraint, the algorithm delivers something remarkable. For systems where the dissipation rate is large compared to the nonlinearity — meaning the system’s natural damping dominates over its chaotic tendencies — the runtime scales polynomially with every other parameter: the number of variables, the evolution time, the nonlinearity strength. This is, as the authors note, the first rigorous quantum algorithm capable of simulating strongly nonlinear systems with such efficiency.
A Proof of Quantum Supremacy
Perhaps the most provocative finding in the paper is not the algorithm itself, but the proof that the problem it solves is BQP-complete. BQP — Bounded-Error Quantum Polynomial Time — is the complexity class of problems that can be solved efficiently by a quantum computer. A problem is BQP-complete if solving it efficiently with a quantum computer would imply that any problem in BQP can be solved efficiently — in other words, it captures the full power of quantum computation.
This is significant. It means that simulating noisy nonlinear dynamics is not just something quantum computers happen to be good at; it is a problem that is inherently quantum in its complexity. If a classical computer could solve it efficiently, that would imply that classical computers can simulate any quantum computation efficiently — a result that would upend our understanding of computational complexity.
The team provides strong evidence that this is not the case. The problem they have identified sits in that sweet spot of computational difficulty: hard enough that classical computers will likely never solve it efficiently, but structured enough that quantum computers can.
Dancing with the Vortex
To demonstrate the algorithm’s practical potential, the team simulated a vortex flow in the two-dimensional Navier-Stokes equation — the fundamental equation describing fluid motion. This is not a toy problem. The Navier-Stokes equation is one of the Clay Millennium Problems, carrying a million-dollar prize for anyone who can prove that solutions always exist and are smooth. Simulating its behavior, even in simplified cases, is a benchmark for any computational method.
The results show that the quantum algorithm can reproduce the expected behavior of a decaying vortex — the swirling structure gradually dissipates as energy leaks out of the system. The simulation matches classical results where they are available, and extends beyond them into regimes where classical methods struggle.
There is something deeply satisfying about this. A vortex — that most classical of phenomena, the kind of thing you see when you pull the plug in a bathtub — being simulated by a quantum computer, a machine built on principles that seem to defy common sense. The universe, it turns out, is stranger and more connected than we imagined.
The Philosophical Payoff
What does this mean for our understanding of computation and reality? The result suggests that the boundary between classical and quantum computation is not where we thought it was. Nonlinear dynamics — the kind of behavior that governs weather, turbulence, and biological systems — may be fundamentally quantum in the computational sense. Not because the systems themselves are quantum, but because simulating them efficiently requires quantum resources.
This is a subtle but profound shift in perspective. We tend to think of quantum computers as being useful for quantum problems — simulating molecules, cracking codes, understanding materials. But this work suggests that some of the most classically difficult problems — the ones that have resisted our best efforts for decades — may yield to quantum approaches not despite their classical nature, but because of it.
The team’s algorithm does not solve the Navier-Stokes problem. It does not predict the weather a month in advance. It does not tame turbulence. What it does is open a door — a narrow door, perhaps, with a carefully specified set of conditions and limitations — but a door nonetheless. Behind it lies the possibility that the most chaotic, the most unpredictable, the most stubbornly classical phenomena in the universe may someday be simulated by machines that operate on principles we are only beginning to understand.
And that, perhaps, is the most humbling thought of all: that the quantum world, which we once thought of as a strange and distant realm, may be the key to understanding the everyday chaos we have lived with all along.
Yanjiang is an online editor of Loom Science
References
- Sergey Bravyi et al., Quantum simulation of a noisy classical nonlinear dynamics, arXiv:2507.06198