arxiv: 2605.08849 · v1 · submitted 2026-05-09 · ⚛️ physics.flu-dyn

Recognition: 2 theorem links

· Lean Theorem

Disentangling coherent structures and the origin of swirl-switching

Eman Bagheri, Philipp Schlatter, Riccardo Casali, Stefan Becker

Pith reviewed 2026-05-12 01:20 UTC · model grok-4.3

classification ⚛️ physics.flu-dyn

keywords swirl switchingturbulent flowbent pipemodal decompositioncoherent structuresstability analysisHilbert transformPOD

0 comments

The pith

Swirl-switching modes in bent pipes arise from distinct mechanisms in the bend versus downstream, rather than a single universal instability.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

A new decomposition method using the Hilbert transform and band-pass filtering separates mixed modes in turbulent bent-pipe flow. It shows that the swirl-switching observed in the bend and downstream sections stem from separate physical processes. The bend mode matches an intrinsic instability identified by local stability analysis of the mean cross-flow, suggesting turbulence excites but does not originate it. Downstream modes connect instead to shear layers in the modified base flow.

Core claim

The filtered Hilbert POD (FHPOD) applied to DNS of flow in a 180° bend at Re_D=10,000 separates four distinct mode families. A swirl-switching mode at Strouhal number 0.13 is localized in the curved section. Local linear stability analysis on the cross-sectional mean flow finds unstable eigenmodes at matching wavenumbers and frequencies, supporting that the phenomenon is an intrinsic instability of the curved-pipe flow excited by incoming turbulence but not caused by it. Downstream modes link to local shear layers of the modified base flow.

What carries the argument

Filtered Hilbert POD (FHPOD), which combines the Hilbert transform with band-pass filtering to prevent mode mixing in classical POD decompositions of turbulent flows.

If this is right

Previous interpretations of a single swirl-switching instability throughout the pipe are incorrect.
The bend swirl-switching can be analyzed and potentially controlled via stability theory of the mean flow.
Downstream coherent structures are independent and tied to post-bend shear layers.
Modal decompositions of similar flows should use filtering to avoid distributing one structure over multiple modes.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

This separation technique may resolve mode-mixing issues in other complex turbulent flows like jets or wakes.
It suggests targeted interventions in pipe bends could damp the local instability without affecting downstream dynamics.
The approach enables parameter-free identification of instability origins in geometry-driven flows.

Load-bearing premise

The band-pass filter bands and the local linear stability analysis on the mean cross-sectional flow correctly isolate the global nonlinear structures without artifacts from filtering choices or non-local turbulent effects.

What would settle it

A simulation of the bent pipe with laminar inlet conditions would show whether the swirl-switching mode still appears at the same frequency and location, or if its presence requires upstream turbulence.

Figures

Figures reproduced from arXiv: 2605.08849 by Eman Bagheri, Philipp Schlatter, Riccardo Casali, Stefan Becker.

**Figure 1.** Figure 1: Coherent structures identified byBr¨ucker(1998) as the manifestation of the swirl-switching phenomenon, shown using the spatio-temporal reconstruction of streamwise vorticity isosurfaces at 1.5 𝐷 downstream of a 90◦ bend measured over 50 seconds. Adapted from figure 2(b) of Br¨ucker (1998), with permission from Christoph Br¨ucker. 0 X0-2 [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗

**Figure 2.** Figure 2: (a) Geometry of the 180◦ bent pipe with curvature of 𝛾 = 0.2. Frenet–Serret frame with sˆ: tangent (streamwise); rˆ: radial (centrifugal); yˆ: binormal (lateral) unit vectors. The streamwise position is given by the non-dimensional arc length 𝑆 = 𝑠/𝐷, where 𝑠 is the dimensional arc length along the pipe centreline and 𝐷 is the pipe diameter. The arrows show the flow direction. condition allows a fully deve… view at source ↗

**Figure 3.** Figure 3: Comparison of the mean flow (a) and the normal Reynolds stresses (b) at 𝑆 = 14 with the reference data (El Khoury et al. 2013; Veenman 2004). 3.2. Filtered Hilbert Proper Orthogonal Decomposition (FHPOD) The FHPOD method is designed to address two limitations of classical POD, namely, the imperfect pairing of degenerate modes associated with a convecting structure and the mixing of multiple frequency bands… view at source ↗

**Figure 4.** Figure 4: (a) Isosurfaces of the mean flow streamfunction𝛹 = ±1.5 × 10−3 𝑈b𝐷. Panels (b) and (c) separate the two vortex pairs for clarity. (b) The isolated Dean vortices rapidly decay downstream of the bend. (c) A new vortex pair with opposite rotation to the Dean vortices emerges downstream of the bend. For visualisation purposes, we perform a simulation with a Poiseuille profile prescribed at the inflow, in addit… view at source ↗

**Figure 5.** Figure 5: Four instantaneous snapshots of 𝜆2 = −15𝑈 2 𝑏 /𝐷 2 isosurfaces coloured by streamwise vorticity 𝜔𝑠 . Reynolds number is 𝑅𝑒𝐷 = 10, 000, and a Poiseuille profile is prescribed as the inflow condition. The arrow shows the direction of the flow. Panels (a)–(d) show the temporal evolution from initial vortex formations to a fully turbulent state within the bend. 4.1. Identification of modes After establishing t… view at source ↗

**Figure 6.** Figure 6 [PITH_FULL_IMAGE:figures/full_fig_p014_6.png] view at source ↗

**Figure 7.** Figure 7: PSD of the time-coefficients of POD modes 2 and 3 and the real part of HPOD mode 1. HPOD collapses the two degenerate POD modes into a single complex mode and reduces the leakage of the low-frequency instability at 𝑆𝑡 ≈ 0.03. (a) (b) (c) (d) [PITH_FULL_IMAGE:figures/full_fig_p015_7.png] view at source ↗

**Figure 8.** Figure 8: Isosurfaces of 𝛹 = ±0.03𝑈b𝐷. (a) HPOD mode 1 (real part). (b) HPOD mode 1 (imaginary part). (c) POD mode 2. (d) POD mode 3. exhibits nearly equal contributions from 𝑆𝑡 ≈ 0.03 and 𝑆𝑡 ≈ 0.1, along with a smaller peak at 𝑆𝑡 ≈ 0.25. This spectral mixing is also visible in the spatial structure shown in figure 9(b), where isosurfaces of the streamwise velocity 𝑈𝑠 are shown. In this figure, two features indicate… view at source ↗

**Figure 9.** Figure 9: (a) PSD of the time-coefficients for the real part of the third HPOD mode. (b) Isosurface of the streamwise velocity 𝑈𝑠 = ±0.3𝑈b for the real part of the third HPOD mode 0 X0-15 [PITH_FULL_IMAGE:figures/full_fig_p015_9.png] view at source ↗

**Figure 10.** Figure 10: PSD of time-coefficients for the FHPOD modes. 𝐸⊥, 𝑗(𝑆) := ∬ A (𝑆) [PITH_FULL_IMAGE:figures/full_fig_p016_10.png] view at source ↗

**Figure 11.** Figure 11: Streamwise distribution of the cross-stream modal energy 𝐸⊥ (𝑆) of the FHPOD modes shows the spatial support of each mode along the pipe centreline 𝑆. comparing with the classical POD in figure 8. In the classical POD results, downstream structures appear phase-shifted, while the phase relation of the structures within the bend is inconsistent. Furthermore, the streamwise length of these structures inside… view at source ↗

**Figure 12.** Figure 12: shows mode AX, the lowest-frequency mode at 𝑆𝑡 ≈ 0.03, which is the second most energetic mode in the present study (see [PITH_FULL_IMAGE:figures/full_fig_p017_12.png] view at source ↗

**Figure 13.** Figure 13: Isosurfaces of the streamfunction 𝛹 = ±0.03𝑈b𝐷 for (a) mode SS and (b) mode SB [PITH_FULL_IMAGE:figures/full_fig_p018_13.png] view at source ↗

**Figure 14.** Figure 14: Isosurfaces of 𝛹 = ±0.03𝑈b𝐷 for (a) mode DS1 and (b) mode DS2 [PITH_FULL_IMAGE:figures/full_fig_p018_14.png] view at source ↗

**Figure 15.** Figure 15: Isosurfaces of the streamfunction 𝛹 = ±0.015𝑈b𝐷 for (a) mode DS1 and (b) mode DS2, shown in semi-transparent white and overlaid on the mean flow isosurfaces of 𝛹 = ±5.0 × 10−4 𝑈b𝐷 in the lower equatorial half of the bend (𝑦 ≤ 0). Red indicates a rotation axis aligned with sˆ, whereas blue indicates the opposite orientation. Both modes are located near the interface between the counter-rotating four-vortex… view at source ↗

**Figure 16.** Figure 16: Local stability analysis of mean cross-sectional base flows extracted at different angular positions 𝜃 along the 180◦ bend. (a) corresponding growth rate 𝜎𝐷/𝑈𝑏 and (b) Strouhal number of the unstable eigenmodes. Two unstable branches are identified: an upstream branch (𝑈1), dominant over the first half of the bend, and a downstream branch (𝑈2), which becomes unstable in the second half of the bend. emerge… view at source ↗

**Figure 17.** Figure 17: Representative unstable eigenmodes from the local stability analysis, visualised by isosurfaces of the cross-sectional streamfunction at |𝛹 | = 0.5 |𝛹 |max for four angular positions along the bend: (a) 𝜃 = 10◦ , (b) 𝜃 = 40◦ , (c) 𝜃 = 115◦ , and (d) 𝜃 = 165◦ . Red and blue show opposite signs of𝛹. The first two panels correspond to the upstream unstable branch (𝑈1), with contours concentrated closer to th… view at source ↗

**Figure 18.** Figure 18: Comparison between the DNS mode SS and the mode reconstructed from the local stability analysis. The arrows show the direction of the flow. (a) Isosurfaces of the streamfunction for the imaginary part of mode SS from FHPOD and for (b) phase-aligned reconstruction of the unstable local eigenmodes. The uniformly signed isosurface in panel (b) is a consequence of the phase-alignment procedure, which is used … view at source ↗

read the original abstract

Modal decomposition of turbulent flows using classical proper orthogonal decomposition (POD) often suffers from mode mixing, in which a distinct coherent structure may be distributed over several POD modes. We propose a decomposition method based on the Hilbert transform and band-pass filtering to address this issue (filtered Hilbert POD -- FHPOD). We apply this approach to the turbulent flow through a 180 bent pipe at $Re_D=10,000$ (based on bulk velocity ($U_b$) and pipe diameter ($D$)) and curvature $\gamma=0.2$, simulated using direct numerical simulation. The FHPOD results in four distinct mode families, including a swirl-switching mode at Strouhal number of 0.13 localised in the curved section. Our novel modal decomposition shows that the modes observed in the bend and downstream correspond to distinct physical mechanisms rather than to a single universal swirl-switching instability throughout the pipe, as previous work implied. To further examine the origin of the swirl-switching mode, we perform a local stability analysis of the cross-sectional mean flow along the bend. We find unstable eigenmodes at the same streamwise wavenumber and within the same range of Strouhal numbers as the swirl-switching mode found in the modal decomposition. The result supports the interpretation that the swirl-switching phenomenon is an intrinsic instability of the curved-pipe flow that can be excited and potentially enhanced by incoming turbulent structures, but is ultimately not caused by them. Finally, we also establish a link of the downstream modes to the local shear layers of the modified base flow, highlighting the different nature of these modes.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

Swirl-switching is a local instability of the bend rather than one universal mode, shown by separating structures with FHPOD and matching them to stability eigenmodes.

read the letter

The main thing to take away is that this work separates the swirl-switching into something that lives in the bend at St around 0.13 and treats the downstream structures as a different set tied to shear layers. The FHPOD approach, which layers band-pass filtering and the Hilbert transform on top of POD, is what lets them pull the families apart without the usual mixing. They run this on DNS of a 180-degree bend at Re_D=10,000 and curvature 0.2, then compare the extracted mode to local linear stability on the cross-sectional mean flow and get matching wavenumber and frequency. That match is the concrete support for calling the swirl-switching an intrinsic instability of the curved section that turbulence can excite but does not create. It also pushes back on earlier pictures of a single pipe-wide mechanism. The separation and the stability link are the clear advances here, and they line up with engineering interest in bent ducts and turbomachinery. The local analysis is a reasonable check, but the flow is developing through the bend, so the parallel-flow assumption leaves room for non-local or convective effects that a global treatment might catch. The filter cutoffs are free parameters that produce the clean split, and the St value is identified from the data before the stability step, which adds modest dependence. These are standard caveats for the approach and do not break the central comparison. This paper is aimed at people who work on coherent structures in curved or non-straight internal flows and on modal tools that reduce mixing. A reader who needs better separation methods or cares about the origin of swirl-switching would get usable ideas from it. The claim is specific enough and the evidence is direct enough that it deserves a serious referee to check the implementation details.

Referee Report

3 major / 2 minor

Summary. The manuscript introduces filtered Hilbert POD (FHPOD), which combines band-pass filtering and the Hilbert transform to reduce mode mixing in classical POD. Applied to DNS data of turbulent flow in a 180° bent pipe (Re_D = 10,000, γ = 0.2), FHPOD yields four distinct mode families. One family is a swirl-switching mode localized in the bend at St ≈ 0.13. The authors perform local linear stability analysis on the time-averaged cross-sectional velocity profiles and report unstable eigenmodes whose Strouhal numbers and streamwise wavenumbers match those of the FHPOD swirl-switching mode. They conclude that the bend and downstream modes arise from separate physical mechanisms and that swirl-switching is an intrinsic instability of the curved-pipe mean flow that can be excited by turbulence but is not caused by it.

Significance. If the FHPOD separation proves robust and the local stability results remain valid under the parallel-flow approximation, the work would provide concrete evidence that swirl-switching is not a single universal mechanism throughout the pipe and would supply a practical tool for disentangling coherent structures in flows with overlapping scales. The direct frequency/wavenumber match between data-driven modes and linear eigenmodes is a clear strength that supplies an external check on the decomposition. The result also highlights the utility of combining modal analysis with stability theory in curved-wall turbulence.

major comments (3)

[FHPOD method description] The band-pass filter cutoffs that define the four FHPOD mode families are free parameters. The manuscript does not report a sensitivity study showing that the separation into distinct bend and downstream families, or the isolation of the St = 0.13 swirl-switching mode, remains qualitatively unchanged under reasonable variations of these cutoffs. Because the central claim rests on this separation, quantitative robustness metrics are required.
[Local stability analysis] Local linear stability analysis is performed on the cross-sectional mean flow under the parallel-flow assumption. In a strongly developing curved pipe with γ = 0.2 the base flow varies rapidly in the streamwise direction; non-parallel effects, convective transport, and possible global-mode selection are therefore expected. The manuscript should quantify the streamwise development length relative to the wavelength of the unstable eigenmodes and discuss whether the observed frequency/wavenumber match survives a non-parallel or weakly non-parallel formulation.
[Downstream mode discussion] The claim that the downstream modes are linked to local shear layers of the modified base flow is supported only by visual inspection of the mode shapes. A quantitative comparison (e.g., overlap integrals or energy budgets between the FHPOD modes and the shear-layer eigenmodes) would strengthen the assertion that these modes are mechanistically distinct from the bend swirl-switching family.

minor comments (2)

[Abstract] The abstract states that the swirl-switching mode is 'localised in the curved section' while the stability analysis is performed 'along the bend.' Clarify whether the eigenmode comparison is performed at a single station or integrated over the bend length.
[Introduction] Notation for the curvature parameter γ and the Strouhal number definition should be stated explicitly once in the main text even if they are standard in the field.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive and positive report. The comments highlight important aspects of robustness and interpretation that we address below. We have revised the manuscript to incorporate additional analyses and discussions as detailed in the point-by-point responses.

read point-by-point responses

Referee: [FHPOD method description] The band-pass filter cutoffs that define the four FHPOD mode families are free parameters. The manuscript does not report a sensitivity study showing that the separation into distinct bend and downstream families, or the isolation of the St = 0.13 swirl-switching mode, remains qualitatively unchanged under reasonable variations of these cutoffs. Because the central claim rests on this separation, quantitative robustness metrics are required.

Authors: We agree that the band-pass cutoffs are parameters whose influence should be quantified. The cutoffs were chosen to align with distinct peaks in the frequency spectra of the classical POD modes, separating the low-frequency swirl-switching content from higher-frequency structures. In the revised manuscript we have added a sensitivity study in which the cutoffs are varied by ±10 % and ±20 % around the nominal values. The separation into four mode families and the isolation of the St ≈ 0.13 mode remain qualitatively unchanged; quantitative metrics (mode-shape overlap integrals and energy fractions) are reported and show variations below 8 %. This analysis is included as a new appendix. revision: yes
Referee: [Local stability analysis] Local linear stability analysis is performed on the cross-sectional mean flow under the parallel-flow assumption. In a strongly developing curved pipe with γ = 0.2 the base flow varies rapidly in the streamwise direction; non-parallel effects, convective transport, and possible global-mode selection are therefore expected. The manuscript should quantify the streamwise development length relative to the wavelength of the unstable eigenmodes and discuss whether the observed frequency/wavenumber match survives a non-parallel or weakly non-parallel formulation.

Authors: The parallel-flow assumption is indeed an approximation. In the revised manuscript we quantify the streamwise development by computing the streamwise gradient of the mean velocity profiles along the bend. The characteristic wavelength of the unstable eigenmodes (k ≈ 2–3) corresponds to roughly 2–3D, while the mean-flow profiles change appreciably over 5–8D. This scale separation supports the local approximation over one wavelength. We have added a dedicated paragraph discussing the limitations of the parallel-flow assumption, the expected magnitude of non-parallel corrections, and the fact that a global stability analysis lies beyond the present scope. The reported frequency/wavenumber agreement with the FHPOD modes is retained as supporting evidence under the stated approximation. revision: yes
Referee: [Downstream mode discussion] The claim that the downstream modes are linked to local shear layers of the modified base flow is supported only by visual inspection of the mode shapes. A quantitative comparison (e.g., overlap integrals or energy budgets between the FHPOD modes and the shear-layer eigenmodes) would strengthen the assertion that these modes are mechanistically distinct from the bend swirl-switching family.

Authors: While the spatial alignment of the downstream FHPOD modes with the shear layers of the modified base flow is visually evident, we acknowledge that quantitative metrics strengthen the mechanistic interpretation. In the revised manuscript we have computed overlap integrals between the FHPOD downstream modes and the eigenmodes obtained from local stability analysis of the shear-layer profiles extracted at the same streamwise stations. These integrals exceed 0.75 for the dominant downstream family, confirming the association. The integrals are now reported together with the mode-shape comparisons, reinforcing the distinction from the bend swirl-switching family. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation chain is self-contained

full rationale

The paper proposes a new FHPOD method (Hilbert transform plus band-pass filtering) and applies it to DNS data of bent-pipe flow, yielding four mode families with a swirl-switching structure at St=0.13 localized to the bend. It then performs an independent local linear stability analysis directly on the time-averaged cross-sectional mean velocity profiles extracted from the same DNS. Unstable eigenmodes are reported at matching streamwise wavenumber and Strouhal range. This match is not equivalent to the input by construction: the stability calculation uses only the mean flow field and the parallel-flow assumption; it does not incorporate the filtered POD modes, the chosen filter bands, or the Hilbert phase information. The claim that bend and downstream modes represent distinct mechanisms follows from the separation produced by the proposed decomposition, while the intrinsic-instability interpretation is supported by the external stability result rather than by re-deriving the observed frequency. No self-citations of prior author work are invoked as load-bearing uniqueness theorems, no parameters are fitted to a data subset and then relabeled as predictions, and no known empirical pattern is merely renamed. The parallel-flow assumption and filter robustness are questions of correctness and sensitivity, not circularity.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

The work rests on standard fluid-dynamics assumptions plus a small number of filter parameters chosen to isolate observed frequencies; no new physical entities are postulated.

free parameters (1)

band-pass filter cutoffs
Selected to isolate specific Strouhal numbers including the swirl-switching frequency of 0.13

axioms (2)

standard math Incompressible Navier-Stokes equations govern the flow
Basis for the direct numerical simulation at the stated Reynolds number
domain assumption Local linear stability analysis on the time-averaged cross-sectional flow can identify the origin of observed global coherent structures
Invoked to link the stability eigenmodes to the swirl-switching phenomenon

pith-pipeline@v0.9.0 · 5588 in / 1455 out tokens · 58398 ms · 2026-05-12T01:20:24.997056+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

88 extracted references · 88 canonical work pages · 1 internal anchor

[1]

Vectors, tensors and the basic equations of fluid mechanics , author =

work page
[2]

and Becker, S

Bagheri, E. and Becker, S. and Schlatter, P. , title =. Journal of Fluid Mechanics , year =

work page
[3]

Journal of Computational Physics , volume =

Spurious modes in Spectral Collocation Methods with Two Non-periodic Directions , author =. Journal of Computational Physics , volume =

work page
[4]

and Katz, J

Bai, K. and Katz, J. , title =. Experiments in Fluids , year =

work page
[5]

Berger, S. A. and Talbot, L. and Yao, L.-S. , title =. Annual Review of Fluid Mechanics , year =

work page
[6]

Bilde, K. G. and S. Decay of secondary motion downstream bends in turbulent pipe flows , journal =. 2023 , volume =

work page 2023
[7]

Elements of Fluid Dynamics , author =

work page
[8]

2025 , Date-Added =

Laminar flow and linear stability in helical pipes , author =. 2025 , Date-Added =

work page 2025
[9]

and Trip, R

Camarri, S. and Trip, R. and Fransson, J. H. M. , title =. Journal of Fluid Mechanics , year =

work page
[10]

and Alenius, E

Carlsson, C. and Alenius, E. and Fuchs, L. , title =. Physics of Fluids , year =

work page
[11]

Journal of Fluid Mechanics , volume =

Modal instability of the flow in a toroidal pipe , author =. Journal of Fluid Mechanics , volume =. 2016 , publisher =

work page 2016
[12]

International Journal of Heat and Fluid Flow , volume =

Characterisation of the steady, laminar incompressible flow in toroidal pipes covering the entire curvature range , author =. International Journal of Heat and Fluid Flow , volume =. 2017 , publisher =

work page 2017
[13]

Spectral Methods: Fundamentals in Single Domains , author =

work page
[14]

Experimental Thermal and Fluid Science , volume =

An experimental investigation regarding the laminar to turbulent flow transition in helically coiled pipes , author =. Experimental Thermal and Fluid Science , volume =. 2006 , publisher =

work page 2006
[15]

A time-recording

Br. A time-recording. Proceedings of the Eighth International Symposium on Flow Visualization (ISFV8) , address =

work page
[16]

Hydrodynamic Stability , author =

work page
[17]

Dean, W. R. , title =. The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science , year =

work page
[18]

and Ball, K

Duggleby, A. and Ball, K. S. and Paul, M. R. , title =. 2006 , eprint =

work page 2006
[19]

and Ball, K

Duggleby, A. and Ball, K. S. and Paul, M. R. and Fischer, P. F. , title =. Journal of Turbulence , year =

work page
[20]

El Khoury, G. K. and Schlatter, P. and Noorani, A. and Fischer, P. F. and Brethouwer, G. and Johansson, A. V. , title =. Flow, Turbulence and Combustion , year =

work page
[21]

and Lieuwen, T

Emerson, B. and Lieuwen, T. C. and Juniper, M. P. , title =. Journal of Fluid Mechanics , year =

work page
[22]

Fischer, P. F. and Lottes, J. W. and Kerkemeier, S. G. , title =. 2008 , howpublished =

work page 2008
[23]

Physical Review Fluids , volume =

Instability of steady flows in helical pipes , author =. Physical Review Fluids , volume =. 2020 , publisher =

work page 2020
[24]

Fluid Dynamics Research , volume =

A comparative study on instability of steady flows in helical pipes , author =. Fluid Dynamics Research , volume =. 2020 , publisher =

work page 2020
[25]

and Colonius, T

Gudmundsson, K. and Colonius, T. , title =. Journal of Fluid Mechanics , year =

work page
[26]

Mathematical and Numerical Modeling of 1-D and 2-D Consolidation , author =

work page
[27]

Journal of Fluid Mechanics , volume =

On the effect of torsion on a helical pipe flow , author =. Journal of Fluid Mechanics , volume =. 1982 , publisher =

work page 1982
[28]

and Apte, S

He, X. and Apte, S. V. and Karra, S. K. and Do. An. Physics of Fluids , year =

work page
[29]

Turbulent pipe flow downstream of a 90 ^

Hellstr. Turbulent pipe flow downstream of a 90 ^. Journal of Fluid Mechanics , year =

work page
[30]

and Canton, J

Hufnagel, L. and Canton, J. and. The three-dimensional structure of swirl-switching in bent pipe flow , journal =. 2018 , volume =

work page 2018
[31]

International Journal for Numerical Methods in Fluids , volume =

H. International Journal for Numerical Methods in Fluids , volume =. 1999 , publisher =

work page 1999
[32]

International Journal of Heat and Fluid Flow , volume =

Influence of curvature and torsion on turbulent flow in helically coiled pipes , author =. International Journal of Heat and Fluid Flow , volume =. 2000 , doi =

work page 2000
[33]

and Ewing, D

Jain, A. and Ewing, D. and Ching, C. Y. , title =. Experimental Thermal and Fluid Science , year =

work page
[34]

and Ewing, D

Jain, A. and Ewing, D. and Ching, C. Y. , title =. International Journal of Heat and Fluid Flow , year =

work page
[35]

Computers & Fluids , volume =

Neko: A modern, portable, and scalable framework for high-fidelity computational fluid dynamics , author =. Computers & Fluids , volume =. 2024 , publisher =

work page 2024
[36]

and Hussain, F

Jeong, J. and Hussain, F. , title =. Journal of Fluid Mechanics , year =

work page
[37]

Juniper, M. P. and Pier, B. , title =. European Journal of Mechanics -- B/Fluids , year =

work page
[38]

Kalpakli, A. and. Turbulent pipe flow downstream a 90 ^. International Journal of Heat and Fluid Flow , year =

work page
[39]

Kalpakli, A. and. Experiments in Fluids , year =

work page
[40]

and Sattarzadeh, S

Kalpakli, A. and Sattarzadeh, S. S. and. Combined hot-wire and. Journal of Visualization , year =

work page
[41]

Journal of Fluid Mechanics , volume =

Pressure-driven flows in helical pipes: bounds on flow rate and friction factor , author =. Journal of Fluid Mechanics , volume =. 2020 , publisher =

work page 2020
[42]

and Canton, J

Lupi, V. and Canton, J. and Schlatter, P. , title =. International Journal of Heat and Fluid Flow , year =

work page
[43]

and Massaro, D

Lupi, V. and Massaro, D. and Peplinski, A. and Schlatter, P. , title =. Physical Review Fluids , year =

work page
[44]

and Canton, J

Lupi, V. and Canton, J. and Rinaldi, E. and. Modal stability analysis of toroidal pipe flow approaching zero curvature , journal =. 2024 , volume =

work page 2024
[45]

Lupi, V. and. Direct Numerical Simulations of Turbulent Flow in Helical Pipes , booktitle =

work page
[46]

Shapes of Pipes , author =

work page
[47]

and Demange, S

Li, X.-B. and Demange, S. and Chen, G. and Wang, J.-B. and Liang, X.-F. and Schmidt, O. T. and Oberleithner, K. , title =. Journal of Fluid Mechanics , year =

work page
[48]

and Lupi, V

Massaro, D. and Lupi, V. and Peplinski, A. and Schlatter, P. , title =. Physical Review Fluids , year =

work page
[49]

and Peplinski, A

Massaro, D. and Peplinski, A. and Schlatter, P. , title =. Journal of Fluid Mechanics , year =

work page
[50]

and Sipp, D

Mettot, C. and Sipp, D. and B. Quasi-laminar stability and sensitivity analyses for turbulent flows: Prediction of low-frequency unsteadiness and passive control , journal =. 2014 , volume =

work page 2014
[51]

2013 , issn =

Evolution of turbulence characteristics from straight to curved pipes , journal =. 2013 , issn =. doi:https://doi.org/10.1016/j.ijheatfluidflow.2013.03.005 , author =

work page doi:10.1016/j.ijheatfluidflow.2013.03.005 2013
[52]

and Schlatter, P

Noorani, A. and Schlatter, P. , title =. International Journal of Heat and Fluid Flow , year =

work page
[53]

and Sieber, M

Oberleithner, K. and Sieber, M. and Nayeri, C. N. and Paschereit, C. O. and Petz, C. and Hege, H.-C. and Noack, B. R. and Wygnanski, I. , title =. Journal of Fluid Mechanics , year =

work page
[54]

and Paschereit, C

Oberleithner, K. and Paschereit, C. O. and Wygnanski, I. , title =. Journal of Fluid Mechanics , year =

work page
[55]

Differential Geometry and its Applications , author =

work page
[56]

and Craft, T

Poletto, R. and Craft, T. and Revell, A. , title =. Flow, Turbulence and Combustion , year =

work page
[57]

Turbulent Flows , author =

work page
[58]

and Kinzel, M

Kriegseis, J. and Kinzel, M. and Nobach, H. , title =. Proceedings of ISPIV 2021 , year =

work page 2021
[59]

and Rezaeiravesh, S

Perez, A. and Rezaeiravesh, S. and Markidis, S. and Schlatter, P. , title =. 2025 , howpublished =. doi:10.2139/ssrn.5380203 , url =

work page doi:10.2139/ssrn.5380203 2025
[60]

and Paschereit, C

Qadri, J. and Paschereit, C. O. and Oberleithner, K. , title =. European Journal of Mechanics -- B/Fluids , year =

work page
[61]

and Kriegseis, J

Raiola, M. and Kriegseis, J. , title =. AIAA AVIATION 2024 Forum , year =

work page 2024
[62]

XXIX. An experimental investigation of the circumstances which determine whether the motion of water shall be direct or sinuous, and of the law of resistance in parallel channels , author =. Philosophical Transactions of the Royal Society of London , number =. 1883 , publisher =

work page
[63]

Hilbert Proper Orthogonal Decomposition: a tool for educing advective wavepackets from flow field data

Raiola, M. and Kriegseis, J. , title =. arXiv preprint arXiv:2507.02487 , year =

work page internal anchor Pith review Pith/arXiv arXiv
[64]

A study of linear wavepacket models for subsonic turbulent jets using local eigenmode decomposition of

Rodr. A study of linear wavepacket models for subsonic turbulent jets using local eigenmode decomposition of. European Journal of Mechanics -- B/Fluids , year =

work page
[65]

Large-eddy simulations of 90 ^

R. Large-eddy simulations of 90 ^. Journal of Turbulence , year =

work page
[66]

Large-eddy simulation of low-frequency oscillations of the

R. Large-eddy simulation of low-frequency oscillations of the. Physics of Fluids , year =

work page
[67]

and Sonobe, R

Sakakibara, J. and Sonobe, R. and Goto, H. and Tezuka, H. and Tada, H. and Tezuka, K. , title =. 14th International Symposium on Flow Visualization (ISFV14) , address =. 2010 , pages =

work page 2010
[68]

and Machida, N

Sakakibara, J. and Machida, N. , title =. Physics of Fluids , year =

work page
[69]

and Citro, V

Siconolfi, L. and Citro, V. and Giannetti, F. and Camarri, S. and Luchini, P. , title =. Journal of Fluid Mechanics , year =

work page
[70]

and Henningson, D

Schmid, P. and Henningson, D. , year =. Stability and Transition in Shear Flows , publisher =

work page
[71]

and Lebedev, A

Sipp, D. and Lebedev, A. , title =. Journal of Fluid Mechanics , year =

work page
[72]

and Sumida, M

Sudo, K. and Sumida, M. and Hibara, H. , title =. Experiments in Fluids , year =

work page
[73]

and Juniper, M

Tammisola, O. and Juniper, M. P. , title =. Journal of Fluid Mechanics , year =

work page
[74]

and Shwin, S

Takahashi, H. and Shwin, S. and Hamdani, A. and Fujisawa, N. and Kikura, H. , title =. Journal of Flow Control, Measurement & Visualization , year =

work page
[75]

Taylor, G. I. , title =. Proceedings of the Royal Society of London. Series A , year =

work page
[76]

and Kaario, O

Tirunagari, S. and Kaario, O. and Vuorinen, V. and Larmi, M. , title =. CSI Journal of Computing , year =

work page
[77]

Tunstall, M. J. and Harvey, J. K. , title =. Journal of Fluid Mechanics , year =

work page
[78]

van Doorne, C. W. H. and Westerweel, J. , title =. Experiments in Fluids , year =

work page
[79]

and Kumar, V

Vashisth, S. and Kumar, V. and Nigam, K. D. P. , title =. Industrial & Engineering Chemistry Research , year =

work page
[80]

Veenman, M. P. B. , title =. 2004 , volume =

work page 2004

Showing first 80 references.