Wave-driven geostrophic vortices in rotating turbulence
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Physics. The day in general physics: instruments, methods and results across the field.
Kai: We're looking at the paper: "Wave-driven geostrophic vortices in rotating turbulence".
Mira: Here's the gist of the paper. Wave-driven geostrophic vortices in rotating turbulence investigates how large-scale geostrophic vortices are maintained in rotating turbulence by measuring Reynolds stresses and analyzing wave interactions Mechanism of Vortex Maintenance The maintenance of coherent geostrophic vortices requires a continuous transfer of momentum from surrounding turbulent fluctuations, which is provided by the Reynolds stresses generated by inertial waves The authors confirm that the Reynolds stresses generated by inertial waves are consistent with the Reynolds-Averaged Navier-Stokes balance of the large-scale flow Wave-field analysis shows that geostrophic vortices selectively reflect and absorb inertial waves according to the sign of their angular momentum Wave absorption occurs either through the formation of a critical layer ahead of the reflection point or through enhanced viscous dissipation associated with increasing local wavenumber As rotation increases, fewer inertial waves undergo resonant absorption, which reduces angular-momentum transfer to the vortices and weakens the geostrophic flow Cyclone-Anticyclone Asymmetry The simulations demonstrate a predicted cyclone-anticyclone asymmetry where cyclones exhibit stronger and more compact cores than anticyclones Specifically, cyclones exhibit larger peak azimuthal velocities, with the velocity maximum located closer to the vortex center than in anticyclones This asymmetry is observed at finite Rossby numbers Wave-Vortex Interaction Details The selective nature of wave propagation is due to absorption occurring within critical layers where the Doppler-shifted frequency of an inertial wave vanishes, or through a subcritical mechanism arising from enhanced viscous dissipation associated with increasing local wavenumber Absorption is selective with respect to the sign of the azimuthal wavenumber m, as only harmonics satisfying ωmU > 0 are absorbed This asymmetry generates a nonzero off-diagonal component ⟨uφu r⟩Z,Φ,T of the Reynolds stress tensor, which sustains the geostrophic vortices in accordance with Eq.
Kai: (16) Effect of Rotation Rate The fraction of energy contained in geostrophic modes depends non-monotonically on the rotation rate Increasing rotation initially promotes two-dimensionalization, but sufficiently rapid rotation suppresses the large-scale geostrophic flow because the fraction of forced inertial-wave spectrum capable of transferring angular momentum decreases The system develops a single columnar vortex rather than a cyclone-anticyclone pair in the regime of rapid rotation (omega ≥ 60) Strong rotation does not necessarily enhance geostrophic organization; instead, it can ultimately suppress the formation and maintenance of large-scale vortices because increasing rotation enlarges the total interval of excited frequencies while leaving the resonant frequency range essentially unchanged The suppression arises from a different mechanism than in laboratory experiments involving noslip boundaries, as it is due to the decreasing fraction of forced inertial waves that satisfy the resonance condition Suppression at Rapid Rotation The geostrophic energy fraction first increases with rotation rate, reaches a maximum, and then decreases as the rotation becomes sufficiently rapid The suppression mechanism is related to how the forcing excites inertial waves with frequencies ω < 2omega and an approximately uniform spectral density The characteristic velocity of the geostrophic flow is bounded by Ug ≲ pε/α, where ε is the energy injection rate and α is the linear friction coefficient Structure of Geostrophic Vortices The local shear rate Σ(r) = r∂r(U/r) transfers energy from three-dimensional fluctuations to the coherent vortex, sustaining its existence The equation governing the structure in the statistically stationary state is derived by balancing viscous and frictional dissipation with the off-diagonal component of the Reynolds stress tensor The results for both cyclone and anticyclone show quantitative agreement with Eq.
Mira: (16) Wave Propagation Analysis The analysis of inertial waves shows that inside cyclones, the m = 1 mode can develop a critical layer at r∗ ≈ 0.2, leading to absorption for r > r∗ Inside anticyclones, the m = −1 mode encounters a turning point at rt ≈ 0.35 instead of a critical layer The absence of a critical layer does not imply that the anticyclone cannot selectively absorb angular momentum through distributed viscous attenuation Selective absorption preferentially attenuates waves whose azimuthal wavenumbers have the appropriate sign, generating the nonzero off-diagonal component of the Reynolds stress tensor Frequency Spectra Analysis At low frequencies, the spectrum is dominated by the geostrophic contribution E2D, and superimposed is the three-dimensional component E3D associated with inertial waves For frequencies exceeding ω > 2omega, the three-dimensional component E3D decreases sharply, consistent with a cutoff imposed by the inertial-wave dispersion relation The total spectrum is determined almost entirely by the three-dimensional modes at these higher frequencies Grid Convergence and Spectral Computation The comparison of energy spectra computed using FFT (blue), FFT combined with DC/ST (yellow), and FFT with Hann windowing (green) shows that the yellow curve represents the reference solution employed throughout the main text The results obtained on all grids are in good agreement over the range of wavenumbers k ≲ 100 A spatial resolution of 5123 is sufficient for simulations with the chosen parameters The coarsest simulation exhibits an unphysical increase in energy density at large wavenumbers k ≳ 100, which can be attributed to insufficient spatial resolution Reynolds Stress Balance Validation The resulting balance for the anticyclone shows that the sum of all terms remains close to zero The residual imbalance near the vortex center is larger than for the cyclone, which is primarily attributable to greater uncertainty in locating the anticyclone axis The results support qualitative predictions of Ref.
Kai: 28 regarding velocity profiles of cyclones and anticyclones in a rotating turbulent flow Profiles of Vortices The radial profiles show that the azimuthal velocity is higher in cyclone than in anticyclone, and its maximum is located closer to the vortex axis in cyclone The local shear rate Σ(r) transfers energy from the three-dimensional fluctuations to the coherent vortex, thereby sustaining its existence The vertical vorticity profiles show a clear asymmetry between cyclones and anticyclones: the cyclone is more compact and exhibits higher vorticity at its core The DNS results support the theoretical picture of inertial-wave propagation, reflection, and selective absorption by geostrophic vortices The selection can be implemented due to presence of critical layer or subcritically due to action of viscosity This asymmetry generates a nonzero off-diagonal component ⟨uφu r⟩Z,Φ,T of the Reynolds stress tensor, which sustains the geostrophic vortices in accordance with Eq.
Title and authors: Kai: The title itself hints at the core idea, "Wave-driven geostrophic vortices in rotating turbulence," suggesting the waves aren't just background noise but are actively driving this organization. And we have Vladimir Parfenyev and Sergey Vergeles as the authors.
Mira: I looked at their work, and they’re tackling something really fundamental about how rotation affects these large scale structures, specifically looking at things like the cyclone-anticyclone asymmetry they predicted.
Lev: If you’re trying to run this on hardware, the whole idea hinges on accurately capturing that interaction between the inertial waves and the vortex structure, which is a big computational challenge.
Kai: They demonstrate that cyclones exhibit stronger and more compact cores than anticyclones, specifically showing larger peak azimuthal velocities located closer to their centers.
Mira: That asymmetry is predicted at finite Rossby numbers, which means it shows up in flows that aren't perfectly idealized, which is important because real flows are rarely perfectly symmetric.
Lev: For me, the fact that they confirm this asymmetry even at these finite numbers tells us we need to build simulations that can handle these subtle differences in structure accurately on the actual hardware.
The paper's summary: Kai: So, the summary boils down to how these vortices maintain their existence through a continuous transfer of momentum from surrounding turbulent fluctuations, which they link directly to the Reynolds stresses generated by inertial waves.
Mira: The key mechanism they uncover is that absorption happens either when an inertial wave hits a critical layer or through enhanced viscous dissipation as the local wavenumber gets bigger.
Lev: That selective absorption is what sustains the flow, but we have to ask how much momentum actually gets transferred across that boundary, because if it’s too small, the vortex weakens.
Kai: And there's another interesting point: as rotation rate increases, fewer inertial waves get absorbed resonantly, which actually reduces this angular momentum transfer to the vortices and weakens the geostrophic flow.
Mira: That leads into their findings on rotation rate dependence; they show that increasing rotation initially promotes two-dimensionalization, but at sufficiently rapid speeds, it suppresses the large scale geostrophic flow because less of that forced inertial wave spectrum can actually transfer angular momentum.
Lev: I wonder if that suppression happens in a way that’s easy to manage with error correction; is it a gradual damping or a sudden cut-off?
The paper's improvements: Kai: The paper suggests an improvement in understanding the wave interaction by focusing on the sign of angular momentum for selective reflection and absorption. This gives us a clearer picture of how the vortex filters what waves it lets through.
Mira: And they highlight that this selective nature is due to two distinct processes: absorption at a critical layer or that subcritical mechanism involving viscous dissipation as the local wavenumber grows.
Lev: That distinction between the two absorption mechanisms is something we need to get right if we’re going to design any algorithm for modeling these flows with high fidelity on real systems.
Kai: They also show that this selective absorption generates a nonzero off-diagonal component of the Reynolds stress tensor, which is what actually sustains those vortices according to their equations.
Mira: That off-diagonal component is what keeps the vortex alive, and it’s directly related to the asymmetry they found between cyclones and anticyclones.
Lev: If we’re building a model for this, that off-diagonal term is going to be a massive source of numerical error if we don't capture that selective absorption correctly.
Conclusion: Kai: To wrap up, the paper confirms the theoretical picture of inertial wave propagation and selective absorption by geostrophic vortices. The key result they highlight is that this selection happens either via a critical layer or through subcritical viscous dissipation as the local wavenumber increases.
Mira: They also confirm that this asymmetry generates a nonzero off-diagonal component of the Reynolds stress tensor, which sustains the vortices in line with their equations. And they show that increasing rotation rate eventually suppresses organization because fewer inertial waves satisfy the resonance condition.
Lev: What this means for us is that strong rotation doesn't automatically mean better organization; instead, it can suppress large scale vortices because less of the forced inertial wave spectrum is available to transfer angular momentum when rotation gets too high.
Kai: It’s a nuanced point: increasing rotation actually reduces the fraction of forced inertial waves capable of doing the momentum transfer, which is different from what you might see in lab experiments with noslip boundaries.
Mira: They also show that at very rapid rotation, like when the Rossby number is high, the system settles into a single columnar vortex rather than that cyclone-anticyclone pair we saw earlier.
Lev: So, while they nail the dynamics of how they work in simulations, we still need to figure out how to translate these findings into real hardware constraints for error correction.
Kai: We’ll keep an eye on this paper and see if we can build a system that can predict exactly when you're going to get that single columnar vortex versus the pair.
Landau Institute for Theoretical Physics, Russian Academy of Sciences · HSE University
physics.flu-dyn
Submitted: 2026-10-11
Updated: 2026-11-13
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 88/100
The gist: Wave-driven geostrophic vortices in rotating turbulence investigates how large-scale geostrophic vortices are maintained in rotating turbulence by measuring Reynolds stresses and analyzing wave…
Key concepts
- Geostrophic Vortices
- These are large-scale rotating structures in a fluid flow that maintain their shape through a balance between the Coriolis force and pressure gradients. They are sustained by continuous momentum transfer from surrounding turbulent fluctuations, which is provided by inertial waves.
- Inertial Waves
- These are specific types of waves present in rotating turbulence. The vortices interact with these waves; absorption of these waves is key to maintaining the vortex structure and transferring angular momentum within the system.
- Cyclone-Anticyclone Asymmetry
- Simulations show that cyclones (low pressure centers) have stronger, more compact cores with higher peak azimuthal velocities than anticyclones (high pressure centers). This asymmetry arises from how vortices selectively interact with and absorb different types of inertial waves.
- Reynolds Stress Tensor Component <uφu_r>_Z,Φ,T
- This specific component of the Reynolds stress tensor is crucial because it generates the necessary momentum transfer to sustain the geostrophic vortices. The paper derives an equation linking this term to viscous and frictional dissipation, which governs vortex structure.
Terminology
Summary
Wave-driven geostrophic vortices in rotating turbulence investigates how large-scale geostrophic vortices are maintained in rotating turbulence by measuring Reynolds stresses and analyzing wave interactions
Mechanism of Vortex Maintenance
The maintenance of coherent geostrophic vortices requires a continuous transfer of momentum from surrounding turbulent fluctuations, which is provided by the Reynolds stresses generated by inertial waves The authors confirm that the Reynolds stresses generated by inertial waves are consistent with the Reynolds-Averaged Navier-Stokes balance of the large-scale flow Wave-field analysis shows that geostrophic vortices selectively reflect and absorb inertial waves according to the sign of their angular momentum Wave absorption occurs either through the formation of a critical layer ahead of the reflection point or through enhanced viscous dissipation associated with increasing local wavenumber As rotation increases, fewer inertial waves undergo resonant absorption, which reduces angular-momentum transfer to the vortices and weakens the geostrophic flow
Cyclone-Anticyclone Asymmetry
The simulations demonstrate a predicted cyclone-anticyclone asymmetry where cyclones exhibit stronger and more compact cores than anticyclones Specifically, cyclones exhibit larger peak azimuthal velocities, with the velocity maximum located closer to the vortex center than in anticyclones This asymmetry is observed at finite Rossby numbers
Wave-Vortex Interaction Details
The selective nature of wave propagation is due to absorption occurring within critical layers where the Doppler-shifted frequency of an inertial wave vanishes, or through a subcritical mechanism arising from enhanced viscous dissipation associated with increasing local wavenumber Absorption is selective with respect to the sign of the azimuthal wavenumber m, as only harmonics satisfying ωmU > 0 are absorbed This asymmetry generates a nonzero off-diagonal component ⟨uφu r⟩Z,Φ,T of the Reynolds stress tensor, which sustains the geostrophic vortices in accordance with Eq. (16)
Effect of Rotation Rate
The fraction of energy contained in geostrophic modes depends non-monotonically on the rotation rate Increasing rotation initially promotes two-dimensionalization, but sufficiently rapid rotation suppresses the large-scale geostrophic flow because the fraction of forced inertial-wave spectrum capable of transferring angular momentum decreases The system develops a single columnar vortex rather than a cyclone-anticyclone pair in the regime of rapid rotation (omega ≥ 60)
Conclusion on Suppression
Strong rotation does not necessarily enhance geostrophic organization; instead, it can ultimately suppress the formation and maintenance of large-scale vortices because increasing rotation enlarges the total interval of excited frequencies while leaving the resonant frequency range essentially unchanged The suppression arises from a different mechanism than in laboratory experiments involving noslip boundaries, as it is due to the decreasing fraction of forced inertial waves that satisfy the resonance condition
Suppression at Rapid Rotation
The geostrophic energy fraction first increases with rotation rate, reaches a maximum, and then decreases as the rotation becomes sufficiently rapid The suppression mechanism is related to how the forcing excites inertial waves with frequencies ω < 2omega and an approximately uniform spectral density The characteristic velocity of the geostrophic flow is bounded by Ug ≲ pε/α, where ε is the energy injection rate and α is the linear friction coefficient
Structure of Geostrophic Vortices
The local shear rate Σ(r) = r∂r(U/r) transfers energy from three-dimensional fluctuations to the coherent vortex, sustaining its existence The equation governing the structure in the statistically stationary state is derived by balancing viscous and frictional dissipation with the off-diagonal component of the Reynolds stress tensor The results for both cyclone and anticyclone show quantitative agreement with Eq. (16)
Wave Propagation Analysis
The analysis of inertial waves shows that inside cyclones, the m = 1 mode can develop a critical layer at r∗ ≈ 0.2, leading to absorption for r > r∗ Inside anticyclones, the m = −1 mode encounters a turning point at rt ≈ 0. p. 7 Inside the anticyclone, the m = −1 mode has no critical layer for the chosen ω and kz, according to Eq. (19),
but instead encounters a turning point at rt ≈ 0.Where the paper says it
Frequency Spectra Analysis
At low frequencies, the spectrum is dominated by the geostrophic contribution E2D, and superimposed is the three-dimensional component E3D associated with inertial waves For frequencies exceeding ω > 2omega, the three-dimensional component E3D decreases sharply, consistent with a cutoff imposed by the inertial-wave dispersion relation The total spectrum is determined almost entirely by the three-dimensional modes at these higher frequencies
Grid Convergence and Spectral Computation
The comparison of energy spectra computed using FFT (blue), FFT combined with DC/ST (yellow), and FFT with Hann windowing (green) shows that the yellow curve represents the reference solution employed throughout the main text The results obtained on all grids are in good agreement over the range of wavenumbers k ≲ 100 A spatial resolution of 5123 is sufficient for simulations with the chosen parameters The coarsest simulation exhibits an unphysical increase in energy density at large wavenumbers k ≳ 100, which can be attributed to insufficient spatial resolution
Reynolds Stress Balance Validation
The resulting balance for the anticyclone shows that the sum of all terms remains close to zero The residual imbalance near the vortex center is larger than for the cyclone, which is primarily attributable to greater uncertainty in locating the anticyclone axis The results support qualitative predictions of Ref. 28 regarding velocity profiles of cyclones and anticyclones in a rotating turbulent flow
Profiles of Vortices
The radial profiles show that the azimuthal velocity is higher in cyclone than in anticyclone, and its maximum is located closer to the vortex axis in cyclone The local shear rate Σ(r) transfers energy from the three-dimensional fluctuations to the coherent vortex, thereby sustaining its existence The vertical vorticity profiles show a clear asymmetry between cyclones and anticyclones: the cyclone is more compact and exhibits higher vorticity at its core
Summary of Key Findings
The DNS results support the theoretical picture of inertial-wave propagation, reflection, and selective absorption by geostrophic vortices The selection can be implemented due to presence of critical layer or subcritically due to action of viscosity This asymmetry generates a nonzero off-diagonal component ⟨uφu r⟩Z,Φ,T of the Reynolds stress tensor, which sustains the geostrophic vortices in accordance with Eq. (16) The fraction of forced inertial-wave spectrum capable of transferring angular momentum decreases with increasing rotation rate Consequently, strong rotation can ultimately suppress rather than enhance two-dimensionalization
Final State Observation
In the limit of small local Rossby number on the vortex axis, the velocity profiles of cyclones and anticyclones become nearly identical For the DNS run with omega = 100, the local Rossby number is approximately 0.2, and the corresponding profiles are presented in SM36 The system develops a single columnar vortex rather than a cyclone-anticyclone pair for omega ≥ 60 The rotation direction of this single vortex can be either cyclonic or anticyclonic, with both cases observed in independent simulations with identical parameters
Supplementary Material Details
The supplementary material includes grid-convergence tests, examines the effects of vertical spectral treatment and Hann windowing on the energy spectra It also provides radial profiles of inertial-wave modes with azimuthal wavenumbers m = ±2 and m = ±3 The analysis reports the velocity profiles of geostrophic vortices at omega = 100 The Reynolds-stress balance for the anticyclone is shown in Fig. S3 The radial dependence of the spectral intensity I(r, m) for azimuthal harmonics m = ±2 in cyclone and anticyclone are presented in Fig. p. 7 u
nit
s
)
10 − 6
10 − 5
10 − 4
10 − 3
a) Cyclone
r
10 − 1 100
b) Anticyclone
m = 0
theory
m = 0
theory
FIG. Radial profiles of the spectral intensity (21) of the axisymmetric inertial-wave field (m = 0, kz = 4, ω = 10. u
nit
s
)
10 − 6
10 − 5
10 − 4
10 − 3
a) Cyclone
r
10 − 1 100
b) Anticyclone
m = 0
theory
m = 0
theory
FIG. Radial profiles of the spectral intensity (21) of the axisymmetric inertial-wave field (m = 0, kz = 4, ω = 10. Snapshots of the velocity field are saved every ∆t = 20 time
units throughout this initial transient stage. Figure 1 shows the time evolution of the kinetic energy
E = ⟨u
2+v
2+w
2
⟩S/2. Normalized energy fluxes associated with the total velocity field Π(k)/ε and with the vertically averaged
modes Π( ¯ k)/ε shown alongside with the normalized cumulative power F(k)/ε = Π(k)/ε + D(k)/ε injected into modes
with wavenumbers ≤ k. geostrophic flow is quantified separately through the twodimensional energy spectrum
E2D(kh) = X
kh−1<|kh|≤kh
|uˆ¯(kh)|
2 + |vˆ¯(kh)|
2
/2, (9)
where kh = (kx, ky) is the horizontal wave vector, u¯(x, y)
and v¯(x, y) are the vertically averaged horizontal velocity
components, and the corresponding spectral coefficients
uˆ¯(kh) and vˆ¯(kh) are obtained via a standard 2D Fast
Fourier Transform.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Acknowledgments and Data Availability
The research presented in Sections III-VI was supported by the Basic Research Program of HSE University (project HSE-BR-2025-57) while the work reported in Sections VII-VIII was supported by the Ministry of Science and Higher Education of the Russian Federation (within the state assignment FFWR-2024-0014 of the Landau Institute for Theoretical Physics) The data that support the findings of this study are available from the corresponding author upon reasonable request The paper is 2610.15863.
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Precise prediction of vortex core characteristics based on rotation rate: The improved AI system can predict whether a rotating flow will organize into a single columnar vortex or a cyclone-anticyclone pair based on the background rotation rate Omega, as shown by the observation that "for Omega >= 60, the system develops a single columnar geostrophic vortex rather than a cyclone-anticyclone pair."
-
Selective wave interaction modeling: The system can model how vortices selectively absorb inertial waves by determining whether absorption occurs
at critical layers or, when no critical layer is present, through enhanced viscous dissipation associated with the growth of the local wavenumber.
-
Asymmetric vortex core dynamics: The AI can predict the asymmetry in vortex structure by quantifying that
cyclones exhibit stronger and more compact cores
and thatthe cyclone is more compact and exhibits higher vorticity at its core
compared to anticyclones. -
Reynolds stress tensor component estimation: The system can estimate the off-diagonal component of the Reynolds stress tensor, which is crucial for maintaining vortices, by using the equation derived from DNS:
⟨uφu r⟩ Z,Φ,T - νΣ + α Z r0 dx U(x)x 2/r squared = 0.
-
Optimal rotation rate determination: The AI can determine the optimal rotation rate for two-dimensionalization by identifying the point where
the geostrophic energy fraction in our DNS initially increases with the rotation rate but decreases at sufficiently rapid rotation.
Abstract
We perform direct numerical simulations of rotating turbulence to investigate the mechanisms responsible for the maintenance of large-scale geostrophic vortices. We confirm that the Reynolds stresses generated by inertial waves are consistent with the Reynolds-Averaged Navier-Stokes balance of the large-scale flow and demonstrate the predicted cyclone-anticyclone asymmetry, with cyclones exhibiting stronger azimuthal velocities and more compact cores. Wave-field analysis provides evidence that geostrophic vortices selectively absorb and reflect inertial waves according to the sign of their angular momentum. Wave absorption occurs either at a critical layer encountered as waves propagate radially inward toward the vortex core, before reaching the reflection point, or through enhanced viscous dissipation associated with increasing local wavenumber. As rotation rate increases, inertial waves with opposite angular wavenumbers exhibit increasingly similar dispersion relations, reducing the difference in their absorption. This weakens the net angular-momentum transfer to the vortices and diminishes the strength of the geostrophic flow. Consequently, the fraction of energy contained in geostrophic modes depends non-monotonically on the rotation rate, demonstrating that sufficiently rapid rotation can ultimately suppress, rather than enhance, turbulence two-dimensionalization.
Sources
- Confirming Wave Turbulence Predictions in Rotating Turbulence
- High-level, high-resolution ocean modeling at all scales with Oceananigans
Related papers
- Data-driven impeller model for efficient large eddy simulations of metastable von K'arm'an flows
- Numerical study of nitrogen oxides chemistry during plasma assisted combustion in a sequential combustor
- Emergent Transfer of a Physics Foundation Model from Simulation to Laboratory Turbulence
- Early onset of secondary shear instability in Kelvin-Helmholtz braids at high Reynolds number
- Rotational covariance, continuous spectrum and non-modal growth of three-dimensional planar weak viscous shock layers
- Deep Learning-based Algebraic Reynolds Stress Closures for RANS Simulations of Turbulent Flows