When fluid flow transitions from smooth, predictable laminar motion to chaotic, swirling motion, it becomes turbulent. Turbulence is one of the great unsolved challenges in classical physics: it is deterministic (governed by Newton's laws), yet its behavior is so complex and sensitive to initial conditions that it appears random. Understanding turbulence is essential for engineering (aircraft drag, pipeline efficiency, weather prediction) and has profound implications for energy dissipation and mixing.
Laminar to turbulent transition: Reynolds number revisited
Recall from §14.6 that the Reynolds number predicts flow regime:
\[ \text{Re} = \frac{\rho v D}{\eta} \]
- Re << 1: Stokes flow (viscosity dominates, drag ∝ v)
- 1 < Re < 1000: Laminar flow (organized layers, nonlinear drag)
- Re ~ 1000–10,000: Transition (flow becomes unsteady, vortices form)
- Re >> 10,000: Turbulence (chaotic, drag ∝ v²)
At high Reynolds numbers, inertial forces (which tend to create motion) far outweigh viscous forces (which dampen it). Small disturbances grow into large-scale eddies, causing the flow to become chaotic.
Why does turbulence occur?
A smooth laminar flow is unstable at high Re. Tiny perturbations—microscopic roughness, vibrations, or thermal fluctuations—are amplified by the flow itself. Faster-moving fluid elements outpace slower ones, creating shear layers. These shear layers roll up into vortices (spinning eddies). The vortices interact, merge, and cascade into smaller and smaller scales—a process called the energy cascade.
Energy cascade and dissipation
In turbulent flow, energy enters at large scales (e.g., a river flowing around a boulder creates large eddies). The energy is transferred to smaller eddies through inertial subrange, where neither viscosity nor large-scale forcing dominates. Eventually, at the smallest scales (the Kolmogorov scale, typically micrometers to millimeters), viscosity becomes significant and dissipates energy as heat.
\[ \ell_K = \left(\frac{\eta^3}{\rho \epsilon}\right)^{1/4} \]
where \( \epsilon \) is the energy dissipation rate per unit mass. This cascade creates self-similarity: the structure of eddies at different scales is statistically similar, leading to fractal-like geometry in turbulent fields.
Worked example: Drag coefficient in turbulent flow
For a sphere at high Re (turbulent wake), the drag force is approximately:
\[ F_D = \frac{1}{2} C_D \rho A v^2 \]
where \( C_D \) is the drag coefficient (dimensionless, typically 0.4–0.5 for a sphere, depending on surface roughness). Compare this to Stokes' law (\( F_D = 6\pi\eta r v \)):
- Stokes (Re << 1): \( F_D \propto v \), independent of \( \rho \)
- Turbulent (Re >> 1): \( F_D \propto v^2 \), proportional to \( \rho \)
Numerical example: Consider a baseball (m ≈ 0.145 kg, r ≈ 0.037 m) thrown at v = 40 m/s through air (\( \rho_{\text{air}} = 1.2 \) kg/m³, \( \eta = 1.8 × 10^{-5} \) Pa·s).
Reynolds number: \[ \text{Re} = \frac{1.2 \times 40 \times 2 \times 0.037}{1.8 \times 10^{-5}} \approx 500,000 \]
At this Re, turbulent drag dominates: \[ F_D = \frac{1}{2} \times 0.4 \times 1.2 \times (\pi \times 0.037^2) \times 40^2 \approx 1.4 \text{ N} \]
This matches experimental observations: air resistance causes a baseball to slow from 40 m/s to ~20 m/s over a typical 60-foot pitch.
Vortex formation and the Magnus effect
A spinning ball experiences sideways force because the rotation creates asymmetric flow. On one side, ball rotation and air flow are in the same direction (additive, higher relative velocity, lower pressure by Bernoulli). On the other side, they oppose (subtractive, lower relative velocity, higher pressure). The pressure difference creates a sideways force—the Magnus effect.
\[ F_{\text{Magnus}} = \frac{1}{2} C_L \rho A v^2 \]
where \( C_L \) is the lift coefficient. A pitcher throwing a curveball exploits this: spin creates lift, making the ball curve unexpectedly.
Turbulent mixing and diffusion
One practical benefit of turbulence is enhanced mixing. In laminar flow, mixing occurs slowly by molecular diffusion. In turbulence, eddies of different scales mix fluids rapidly, spreading contaminants, heat, and nutrients much faster than diffusion alone.
The effective turbulent diffusion coefficient is much larger than molecular diffusion: \[ D_{\text{eff}} \sim v' \ell' \]
where \( v' \) and \( \ell' \) are typical velocity and length fluctuations in the turbulent flow. For a river, turbulence disperses pollutants downstream far faster than still water would.
Friction factor and pipe flow
For turbulent flow in a pipe, the Darcy-Weisbach equation relates pressure drop to flow:
\[ \Delta P = f \frac{L}{D} \frac{1}{2} \rho v^2 \]
where \( f \) is the friction factor (depends on Re and pipe roughness). The friction factor is not constant; for smooth pipes, it decreases with Re (more inertia, less viscous effect relative to inertia):
\[ f \propto \text{Re}^{-0.25} \quad (\text{turbulent, smooth pipe}) \]
This is why larger pipes are more efficient: drag per unit length decreases as the pipe diameter increases (at the same flow speed).
Real-world consequences: weather, aviation, and engineering
Weather patterns: Turbulence in the atmosphere creates weather. Laminar air flow over mountains would produce smooth, predictable air motion. Instead, turbulence generates vortices, low-pressure systems, and storms.
Airplane design: Aircraft engineers must account for both drag (quadratic in speed) and skin friction (turbulent boundary layer). Modern aircraft are designed to minimize both, using streamlined shapes and special surface coatings (riblets inspired by shark skin).
Fluid transport: Pumping viscous or turbulent fluids through pipes requires substantial energy. Reducing turbulence (using larger pipes, lower flow rates, or smoother surfaces) saves fuel and money.
Climate and diffusion: Atmospheric and oceanic turbulence determines how quickly heat, carbon, and pollutants spread globally.
Outstanding questions in turbulence
Despite decades of research, many questions remain:
- Can turbulence be fully predicted or only statistically characterized?
- What determines the energy cascade structure at all scales?
- How do vortex breakdown and reconnection occur?
These questions drive research in computational fluid dynamics (CFD), a field that uses supercomputers to simulate turbulence at high resolution.
What you should be able to do
- Explain the transition from laminar to turbulent flow in terms of Reynolds number
- Describe the energy cascade and how turbulence dissipates energy
- Compare drag at low Re (viscous, Stokes) and high Re (turbulent, quadratic)
- Recognize drag coefficients and apply them to moving objects
- Explain the Magnus effect and its role in spinning objects
- Understand enhanced diffusion and mixing in turbulent flow
- Apply the Darcy-Weisbach equation to pipe flow
- Appreciate the complexity of turbulence and its importance in nature and engineering
Summary
Turbulence emerges from the instability of laminar flow at high Reynolds numbers. It features a cascade of eddies from large to small scales, with energy dissipated at the smallest (viscous) scales. Turbulent drag is quadratic in velocity and much larger than viscous drag, dominating macroscopic fluid transport. Understanding and predicting turbulence remains one of the grand challenges in physics and engineering.
📚 See also: Halliday Vol 2, Ch 14, §14.7.
Have a question? 🤔
If something isn't clear or you have a question, ask it here. The answer will be published on this page.