Vortex density fluctuations in quantum turbulence

We compute the frequency spectrum of turbulent superfluid vortex density fluctuations and obtain the same Kolmogorov scaling which has been observed in a recent experiment in Helium-4. We show that the scaling can be interpreted in terms of the spectrum of reconnecting material lines. The calculation is performed using a vortex tree algorithm which considerably speeds up the evaluation of Biot-Savart integrals.

💡 Research Summary

This paper addresses a striking observation in quantum turbulence: the frequency spectrum of vortex line density fluctuations in superfluid helium‑4 follows a f⁻⁵ᐟ³ scaling, identical to the Kolmogorov scaling seen in classical turbulence. The authors set out to reproduce this result numerically and to understand its physical origin.

The dynamics of vortex lines are modeled by the Schwarz equation (Eq. 1), which includes mutual‑friction terms (α, α′) coupling the superfluid to a prescribed normal‑fluid velocity field vₙ. The superfluid velocity induced by the vortex configuration is given by the Biot–Savart law (Eq. 2). Direct evaluation of the Biot–Savart integral scales as O(N²) and becomes prohibitive for realistic vortex tangles containing hundreds of thousands of discretization points.

To overcome this bottleneck, the authors develop a three‑dimensional octree “vortex‑tree” algorithm. The computational domain is recursively subdivided into cubes; each node stores the total circulation and the “center of vorticity” of its contents. During the velocity evaluation a cell is accepted as a distant source if its distance d satisfies d > w/θ + δ, where w is the cell width, δ the offset of the vorticity centre, and θ the opening angle. With θ = 0.7 the relative error in the velocity is ≤ 0.25 % while the algorithm scales as N log N. This enables simulations with up to N ≈ 4 × 10⁵ points.

The normal fluid is generated using a Kinematic Simulation (KS) that sums a finite set of random Fourier modes. By choosing mode amplitudes Aₘ, Bₘ such that the modal energy spectrum follows E(k) ∝ k⁻⁵ᐟ³, the normal fluid reproduces the Kolmogorov inertial‑range statistics (solenoidal, homogeneous, isotropic). The temperature is set to 2.164 K, giving realistic mutual‑friction coefficients while keeping the back‑reaction of the vortex lines on the normal fluid negligible.

Starting from 16 randomly oriented straight vortices, the system evolves under the coupled dynamics. For three values of the effective Reynolds number (Reₙ ≈ 23, 57, 113) the vortex line density L grows, then saturates to a statistically steady state. Snapshots of the saturated tangle (Fig. 3) show a dense, tangled network. Energy spectra computed on a 512² grid in the central plane reveal a clear k⁻⁵ᐟ³ range for the normal fluid, and the superfluid follows the same scaling in the same inertial range. At wavenumbers larger than the forcing cutoff k_M, the normal fluid is essentially at rest and mutual friction damps Kelvin waves, leading to a k⁻¹ spectrum characteristic of isolated straight vortex lines.

Velocity statistics of the superfluid are non‑Gaussian. The probability density functions of the three velocity components exhibit power‑law tails PDF(v_i) ∝ v_i^{−3.1}, consistent with previous helium experiments and with theoretical predictions based on the singular nature of vortex lines.

The central result concerns the power spectral density (PSD) of L(t). In the saturated regime the PSD scales as f⁻⁵ᐟ³ over a broad frequency band (Fig. 6), reproducing the experimental finding. To interpret this, the authors replace the active vortex filaments with passive material lines that are advected solely by the prescribed normal‑fluid field (no reconnection). The length of these material lines also fluctuates and its PSD again follows f⁻⁵ᐟ³ (Fig. 7). This supports the hypothesis that randomly oriented vortex segments behave like passive scalars, and that the observed density fluctuations stem from the statistics of reconnecting material lines rather than from the vorticity spectrum itself.

In summary, the paper demonstrates three key aspects of quantum turbulence: (i) a classical Kolmogorov k⁻⁵ᐟ³ energy cascade is present in both normal and superfluid components; (ii) superfluid velocity statistics are strongly non‑Gaussian, reflecting the singular vortex structure; (iii) vortex‑line‑density fluctuations exhibit an f⁻⁵ᐟ³ spectrum, which can be understood as the signature of reconnecting material‑line dynamics. The vortex‑tree algorithm, especially when parallelized, opens the way to much larger simulations, enabling future studies of zero‑temperature limits, Kelvin‑wave cascades, and anomalous scaling phenomena.