Turbulent solutions of the binormal flow and the 1D cubic Schrödinger equation
Valeria Banica, Luis Vega · 2024
Read Turbulent solutions of the binormal flow and the 1D cubic Schrödinger equation by Valeria Banica, Luis Vega (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-14013.txt.
Introduction
Valeria Banica and Luis Vega survey six years of work on wild, turbulent behavior in a classic wave equation and a linked geometric flow. The equation, the one-dimensional cubic Schrödinger equation, was long set aside in turbulence studies because it can be solved exactly. The linked flow describes how a curve moves along its own normal direction.
Working with others, the authors proved a wide range of complex behavior in these systems. Their results cover the birth of sharp points, growth of wave energy, fractal patterns, and bursts of activity. Some findings confirm what computer simulations had suggested earlier. The survey, written for a special issue, gathers the story in one place for mathematicians.
Catalog record
| CREATOR | Valeria Banica, Luis Vega |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 93 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-14013
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Turbulent solutions of the binormal flow and the 1D cubic Schrödinger equation Valeria Banica, Luis Vega Abstract . In the last three decades there has been an intense activity on the exploration of turbulent phenomena of dispersive equations, as for instance the growth of Sobolev norms since the work of Bourgain in the 90s. In general the 1D cubic Schrödinger equation has been left aside because of its complete integrability. In a series of papers of the last six years that we survey here for the special issue of the ICMP 2024 ( [ 12 ] , [ 13 ] , [ 14 ] , [ 15 ] , [ 16 ] , [ 7 ] , [ 8 ] ), we considered, together with the 1D cubic Schrödinger equation, the binormal flow, which is a geometric flow explicitly related to it. We displayed rigorously a large range of complex behavior as creation of singularities and unique continuation, Fourier growth, Talbot effects, intermittency and multifractality, justifying in particular some previous numerical observations. To do so we constructed a class of well-posedness for the 1D cubic Schrödinger equation included in the critical Fourier-Lebesgue space ℱ L ∞ \mathcal{F}L^{\infty} and in supercritical Sobolev spaces with respect to scaling. Last but not least we recall that the binormal flow is a classical model for the dynamics of a vortex filament in a 3D fluid or superfluid, and that vortex motions are a key element of turbulence. August 24, 2026 1. Introduction 1.1. Vortex filaments dynamics and the binormal flow The evolution of vortex filaments is a key element in fluid and superfluid turbulence. We consider the classical model of the binormal flow (BF, or LIA from “local induction approximation” or VFE from “vortex filament equation”), which is the formally derived model for one vortex filament dynamics in a 3D fluid governed by Euler equations, also used in superfluids. If the vorticity at time t is concentrated along an arclength-parametrized curve \chi(t) in \mathbb{R}^{3} , its evolution in time is expected to evolve according to the binormal flow: (1) \chi_{t}=\chi_{x}\wedge\chi_{xx}. By using the Frenet system (see Appendix 7 ), that characterizes, in terms of the curvature and torsion, the tangent and normal vectors, together with their vectorial product which is called the binormal vector, the equation writes as: \chi_{t}=c\,b, where c stands for the curvature and b for the binormal vector. This explains the binormal flow name. The model was derived formally by using Biot-Savart integral by Da Rios in 1906 following the works of Levi-Civita ( [ 47 ] ). This was justified rigorously by the “if” theorem of Jerrard and Seis in 2017 ( [ 85 ] ). More precisely, under the assumption that the vorticity is concentrated along a smooth curve \tilde{\chi}(t) in \mathbb{R}^{3} for all times t\in[0,T] for some T , and for initial velocity field having controlled excess of the kinetic energy relative to the initial curve, they prove that \tilde{\chi}(t) evolves to leading order by binormal curvature flow. The proof relies on estimates on the distance of the Hamiltonian-Poisson structures of Euler equation and of the binormal flow, together with stability estimates obtained previously by Jerrard and Smets ( [ 86 ] ). Understanding when the vorticity propagates its initial structure of being concentrated along a curve is still a very difficult open problem. We shall recall here only a representative few advances in this direction in the last five years. Concerning the Navier-Stokes equation the Cauchy problem was proved by Bedrossian, Germain and Harrop-Griffiths ( [ 19 ] ) to be locally well-posed for an initial filament data with no symmetry assumptions, but for small times not allowing to observe the binormal flow and to pass to the vanishing viscosity limit. Vanishing viscosity limit was proved by Gallay and Sverak ( [ 62 ] ) for the particular case of axisymmetric vortex rings. The binormal flow dynamic was recovered by Fontelos and Vega ( [ 59 ] ) for Giga-Miyakawa solutions with initial filament data with no symmetry assumptions, but this regime does not allow for passing to the vanishing viscosity limit for positive times. In Euler equations Donati, Lacave and Miot ( [ 54 ] ), and previously Dávila, Del Pino, Musso and Wei ( [ 48 ] ) and Cao and Wan ( [ 36 ] ) by other methods, constructed solutions with vorticity concentrated on helices, which are particular solutions with helical symmetry of the binormal flow. Thus we are still facing a gap towards the case of no-symmetry vortex filaments in Euler equations. 1.2. Links between the binormal flow and Schrödinger equations We shall present now the link between the binormal flow and Schrödinger equations that is at the basis of our results. A detailed presentation is done in § 7 . It is easy to see that if \chi is a binormal flow solution then its tangent vector T=\chi_{x} satisfies (2) T_{t}=T_{x}\wedge T_{x}+T\wedge T_{xx}=T\wedge T_{xx}. This is the 1D Schrödinger map equation with values in the sphere \mathbb{S}^{2} , that coincides with the 1D Heisenberg continuous model derived in ferro-magnetic theory by Landau and Lifshitz in 1935 ( [ 99 ] ). Moreover, solutions of the Schrödinger map, and thus corresponding solutions of the binormal flow equation, are related, via the Hasimoto transformation, to the 1D focusing cubic Schrödinger equation (NLS) on the line: (3) iu_{t}+u_{xx}+|u|^{2}u=0. Indeed Hasimoto discovered in 1972 ( [ 82 ] ) the following transform based on the Frenet system of curves given by tangent, normal and binormal vector. If \chi is a binormal flow solution with non-vanishing curvature, it is easy to show 1 1 1 This is due to the fact that if \chi(t,x) is a binormal flow solution then, by computing and identifying the crossed second derivatives of the tangent and the normal vector T_{tx}=T_{xt},n_{tx}=n_{xt} , it follows that its curvature and torsion (c,\tau) satisfy the system (called intrinsic equations): \left\{\begin{array}[]{ll}c_{t}=-2c_{x}\,\tau-c\,\tau_{x},\\ \tau_{t}=\left(\frac{c_{xx}-c\,\tau^{2}}{c}+\frac{c^{2}}{2}\right)_{x}.\end{array}\right. that the following function, called filament function: u(t,x):=c(t,x)e^{i\int_{0}^{x}\tau(t,s)ds}, where c,\tau are the curvature and torsion of the curve, satisfies the 1D Schrödinger equation (4) iu_{t}+u_{xx}+\left(|u|^{2}-f\right)u=0, where f is a space independent function determined by (c,\tau)(t,0) . The Hasimoto transform is thus assigning a solution of 1D Schrödinger equation to a solution of the binormal flow. In particular Hasimoto’s transform can be seen as an inverse Madelung transform 2 2 2 From the intrinsic equations it follows that if \chi solves the binormal flow then c^{2} and 2\tau satisfy a Euler-Korteweg type equation. Thus Hasimoto’s transform assigns to such Euler-Korteweg solutions a solution of 1D cubic Schrödinger equation. In particular, by a change of phase, Hasimoto’s transform assigns a solution of ( 4 ) with potential f=1 , that is Gross-Pitaevskii equation. Conversely, Madelung transform ensures that if u is a solution of Gross-Pitaesvskii equation, then introducing \rho,v such that u(t,x)=\sqrt{\rho(t,x)}e^{i\theta(t)}e^{i\int_{0}^{x}v(t,s)ds} , i.e. \rho:=|u|^{2} and v:=2\nabla_{x}Arg\,u , then (\rho,v) solve an Euler type equation with the extra quantum pressure, i.e. a Euler-Korteweg type equation. . Conversely, for any function f depending only on the time variable, for instance f\equiv 0 , from a smooth solution u of ( 4 ), Hasimoto gave a method to construct frames whose first vector is a solution