Laguerre polynomials and transitional asymptotics of the modified Korteweg–de Vries equation for step-like initial data


We consider the compressive wave for the modified Korteweg–de Vries equation with background constants \(c>0\) for \(x\rightarrow -\infty \) and 0 for \(x\rightarrow +\infty \). We study the asymptotics of solutions in the transition zone \(4c^2t-\varepsilon t<x<4c^2t-\beta t^{\sigma }\ln t\) for \(\varepsilon >0,\)\(\sigma \in (0,1),\)\(\beta >0.\) In this region we have a bulk of nonvanishing oscillations, the number of which grows as \(\frac{\varepsilon t}{\ln t}.\) Also we show how to obtain Khruslov–Kotlyarov’s asymptotics in the domain \(4c^2t-\rho \ln t<x<4c^2t\) with the help of parametrices constructed out of Laguerre polynomials in the corresponding Riemann–Hilbert problem.

This is a preview of subscription content, log in to check access.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5


  1. 1.

    Bertola, M., Buckingham, R., Lee, S.Y., Pierce, V.: Spectra of random Hermitian matrices with a small-rank external source: the supercritical and subcritical regimes. J. Stat. Phys. 153(4), 654–697 (2013).

    Google Scholar 

  2. 2.

    Bertola, M., Lee, S.Y., Mo, M.Y.: Mesoscopic colonization in a spectral band. J. Phys. A 42, 41, 415204,17 (2009)

    Google Scholar 

  3. 3.

    Bertola, M., Lee, S.Y.: First colonization of a spectral outpost in random matrix theory. Constr. Approx. 30(2), 225–263 (2009)

    Google Scholar 

  4. 4.

    Bertola, M., Tovbis, A.: Universality for the focusing Schroedinger equation at the gradient catastrophe point: rational breathers and poles of the tritronquee solution to Painleve I. Commun. Pure Appl. Math. 66(5), 678–752 (2013)

    Google Scholar 

  5. 5.

    Bothner, T., Deift, P., Its, A., Krasovsky, I.: On the asymptotic behavior of a log gas in the bulk scaling limit in the presence of a varying external potential I. Commun. Math. Phys. 337(3), 1397–1463 (2015). arxiv1407.2910

    Google Scholar 

  6. 6.

    Bikbaev, R.F., Novokshenov, V.Yu.: The Korteveg–de Vries equation with finite gap boundary conditions and self-similar solutions of whitham equations Proc. In: III International Workshop “Nonlinear and Turbulent Processes in Physics” Kiev 1, 32–35 (1988)

  7. 7.

    Bikbaev, R.F., Novokshenov, Yu.V.: Existence and uniqueness of the solution of the Whitham equation (Russian). Asymptotic methods for solving problems in mathematical physics Akad. Nauk SSSR Ural. Otdel. Bashkir. Nauchn. Tsentr Ufa 81–95 (1989)

  8. 8.

    Bikbaev, R.F.: Structure of a shock wave in the theory of the Korteweg–de Vries equation. Phys. Lett. A 141(5–6), 289–293 (1989)

    Google Scholar 

  9. 9.

    Bikbaev, R.F., Sharipov, R.A.: The asymptotic behaviour, as \(t\rightarrow \infty \), of the solution of the Cauchy problem for the Korteweg-de Vries equation in a class of potentials with finite-gap behaviour as \(x\rightarrow \pm \infty \). Teoret. Mat. Fiz. 78/3, 345–356 (1989). (translation in Theoret. and Math. Phys.78/3 244–252)

    Google Scholar 

  10. 10.

    Bikbaev, R.F.: The Korteweg–de Vries equation with finite-gap boundary conditions and Whitham deformations of Riemann surfaces (Russian). Funktsional. Anal. i Prilozhen. 23/4, 1–10 (1990). (translation in Funct. Anal. Appl.23/4 257–266)

    Google Scholar 

  11. 11.

    Bikbaev, R.F.: The influence of viscosity on the structure of shock waves in the MKdV model (Russian). Zap. Nauchn. Sem. S.-Peterburg. Otdel. Mat. Inst. Steklov. (POMI) Voprosy Kvant. Teor. Polya Statist. Fiz. 11, 37–42 (1995). (184 translation in J. Math. Sci.772 3042–3045)

    Google Scholar 

  12. 12.

    Bikbaev, R.F.: Complex Whitham deformations in problems with “integrable instability” (Russian). Teoret. Mat. Fiz. 104/3, 393–419 (1996). (translation in Theoret. and Math. Phys.104/3 1078–1097)

    Google Scholar 

  13. 13.

    Bikbaev, R.F.: Modulational instability stabilization via complex Whitham deformations: nonlinear Schrodinger equation Zap. Nauchn. Sem. S.-Peterburg. Otdel. Mat. Inst. Steklov. (POMI) 215 Differentsialnaya Geom. Gruppy Li i Mekh. 14 65–76 310 translation in J. Math. Sci. (New York)85 (1997) 1 1596–1604

  14. 14.

    Boutet de Monvel, A., Kotlyarov, V.P.: Focusing nonlinear Schrodinger equation on the quarter plane with time-periodic boundary condition: a Riemann–Hilbert approach. J. Inst. Math. Jussieu 6(4), 579–611 (2007)

    Google Scholar 

  15. 15.

    Boutet de Monvel, A., Its, A.R., Kotlyarov, V.P.: Long-time asymptotics for the focusing NLS equation with time-periodic boundary condition. C. R. Math. Acad. Sci. Paris 345(11), 615–620 (2007)

    Google Scholar 

  16. 16.

    Boutet de Monvel, A., Its, A.R., Kotlyarov, V.P.: Long -time asymptotics for the focusing NLS equation with time—periodic boundary condition on the half line. Commun. Math. Phys. 290(2), 479–522 (2009)

    Google Scholar 

  17. 17.

    Boutet de Monvel, A., Kotlyarov, V.P., Shepelsky, D.G.: Focusing NLS equation: long-time dynamics of step-like initial data. Int. Math. Res. Not. 7, 1613–1653 (2011)

    Google Scholar 

  18. 18.

    Buckingham, R., Venakides, S.: Long-time asymptotics of the non-linear Schrodinger equation shock problem. Commun. Pure Appl. Math. 60(9), 1349–1414 (2007)

    Google Scholar 

  19. 19.

    Claeys, T.: Birth of a cut in unitary random matrix ensembles. Int. Math. Res. Not. (IMRN) 2008(6), 40 (2008).

    Google Scholar 

  20. 20.

    Claeys, T., Grava, T.: Solitonic asymptotics for the Korteweg–de Vries equation in the small dispersion limit. SIAM J. Math. Anal. 42(5), 2132–2154 (2010)

    Google Scholar 

  21. 21.

    Cohen, A., Kappeler, T.: Scattering and inverse scattering for steplike potentials in the Schrdinger equation. Indiana Univ. Math. J. 34(1), 121–180 (1985)

    Google Scholar 

  22. 22.

    Corless, R.M., Gonnet, G.H., Hare, D.E.G., Jeffrey, D.J., Knuth, D.E.: On the Lambert W function. Adv. Comput. Math. 5, 329–359 (1996)

    Google Scholar 

  23. 23.

    Deift, P., Kriecherbauer, T., McLaughlin, K.T.-R., Venakides, S., Zhou, X.: Uniform asymptotics for polynomials orthogonal with respect to varying exponential weights and applications to universality questions in random matrix theory. Commun. Pure Appl. Math. 52(11), 1335–1425 (1999)

    Google Scholar 

  24. 24.

    Deift, P., Trubowitz, E.: Inverse scattering on the line. Commun. Pure Appl. Math. 32(2), 121–251 (1979)

    Google Scholar 

  25. 25.

    Deift, P., Zhou, X.: A steepest descent method for oscillatory Riemann–Hilbert problems. Asymptot. MKdV Equ. Ann. Math. 137(2), 295–368 (1993)

    Google Scholar 

  26. 26.

    Deift, P., Venakides, S., Zhou, X.: The collisionless shock region for the long-time behavior of solutions of the KdV equation. Commun. Pure Appl. Math. 47(2), 199–206 (1994)

    Google Scholar 

  27. 27.

    Egorova, I., Gladka, Z., Kotlyarov, V., Teschl, G.: Long-time asymptotics for the Korteweg–de Vries equation with steplike initial data. Nonlinearity 26(7), 1839–1864 (2012)

    Google Scholar 

  28. 28.

    Grunert, K., Teschl, G.: Long-time asymptotics for the Korteweg–de Vries equation via nonlinear steepest descent. Math. Phys. Anal. Geom. 12(3), 287324 (2009)

    Google Scholar 

  29. 29.

    Gurevich, A.V., Pitaevskii, L.P.: Decay of initial discontinuity in the Korteweg–de Vries equation. JETP Lett. 17/5, 193 (1975)

    Google Scholar 

  30. 30.

    Jakovleva, A.I.: Master thesis “Application of inverse scattering transform method to a Cauchy problem for the modified Korteweg-de Vries equation”, Kharkov (1980) (Russian)

  31. 31.

    Kamvissis, S.: Long time behavior for the focusing nonlinear Schroedinger equation with real spectral singularities. Commun. Math. Phys. 180(2), 325–341 (1996)

    Google Scholar 

  32. 32.

    Khruslov, E.Ya.: Splitting of an initial step-like perturbation for the KdV equation. Lett. JETP 21(4), 469–472 (1975)

    Google Scholar 

  33. 33.

    Khruslov, EYa.: Asymptotics of the solution of the Cauchy problem for the Korteweg de Vries equation with initial data of step type. Matem. Sbornik (New Series) 99(141):2, 261–281 (1976)

    Google Scholar 

  34. 34.

    Khruslov, E.Ya., Kotlyarov, V.P.: Asymptotic solitons of the modified Korteweg–de Vries equation. Inverse Probl. 5(6), 1075–1088 (1989)

    Google Scholar 

  35. 35.

    Khruslov, E. Ya., Kotlyarov, V. P.: Soliton asymptotics of nondecreasing solutions of nonlinear completely integrable evolution equations. Spectral operator theory and related topics. Adv. Soviet Math. 19 American Mathematical Society, Providence, RI 129–180 (1994)

  36. 36.

    Khruslov, E.Ya., Kotlyarov, V.P.: Generation of asymptotic solitons in an integrable model of stimulated Raman scattering by periodic boundary data. Mat. Fiz. Anal. Geom. 10(3), 366–384 (2003)

    Google Scholar 

  37. 37.

    Kotlyarov, V., Minakov, A.: Riemann-Hilbert problem to the modified Korteveg de Vries equation: long-time dynamics of the steplike initial data. J. Math. Phys. 51, 093506 (2010)

    Google Scholar 

  38. 38.

    Kotlyarov, V., Minakov, A.: Step-initial function to the MKdV equation: hyper-elliptic long-time asymptotics of the solution. J. Math. Phys. Anal. Geom. 8(1), 37–61 (2011)

    Google Scholar 

  39. 39.

    Kotlyarov, V., Minakov, A.: Modulated elliptic wave and asymptotic solitons in a shock problem to the modified Kortweg–de Vries equation. J. Phys. A 48(30), 305201, 35 (2015)

    Google Scholar 

  40. 40.

    Lavrent’ev, M.A., Sabat, B.V.: Metody teorii funkcii kompleksnogo peremennogo. (Russian) Methods of the theory of functions of a complex variable Gosudarstv. Izdat. Tehn.-Teor. Lit. Moscow-Leningrad (1951)

  41. 41.

    Leach, J.A.: An initial-value problem for the modified Korteweg–de Vries equation. IMA J. Appl. Math. 78(6), 1196–1213 (2013)

    Google Scholar 

  42. 42.

    Marchant, T.R.: Undular bores and the initial-boundary value problem for the modified Korteweg–de Vries equation. Wave Motion 45(4), 540–555 (2008)

    Google Scholar 

  43. 43.

    Marchenko, V.A.: Operatory Shturma-Liuvillya i ikh prilozheniya. (Russian) [Sturm-Liouville operators and their applications] Izdat. “Naukova Dumka”, Kiev (1977) 331 pp

  44. 44.

    Minakov, A.: Long-time behaviour of the solution to the mKdV equation with step-like initial data. J. Phys. A: Math. Theor. 44, 085206 (2011)

    Google Scholar 

  45. 45.

    Minakov, A.: Asymptotics of rarefaction wave solution to the mKdV equation. J. Math. Phys. Anal. Geom. 7(1), 59–86 (2011)

    Google Scholar 

  46. 46.

    Moskovchenko, E.A., Kotlyarov, V.P.: A new Riemann–Hilbert problem in a model of stimulated Raman scattering. J. Phys. A: Math. Gen. 39, 014591 (2006)

    Google Scholar 

  47. 47.

    Moskovchenko, E.A.: Simple periodic boundary data and Riemann–Hilbert problem for integrable model of the stimulated Raman scattering. J. Math. Phys. Anal. Geom. 5/1, 82–103 (2009)

    Google Scholar 

  48. 48.

    Moskovchenko, E.A., Kotlyarov, V.P.: Periodic boundary data for an integrable model of stimulated Raman scattering: long-time asymptotic behavior. J. Phys. A: Math. Theor. 43(5), 31 (2010).

    Google Scholar 

  49. 49.

    Minakov, A.: Ph.D. thesis. Riemann–Hilbert problems and the modified Korteweg de Vries equation: asymptotic analysis of solutions with step-like initial data (2013)

  50. 50.

    Mo, M.Y.: The Riemann–Hilbert approach to double scaling limit of random matrix eigenvalues near the “birth of a cut” transition. Int. Math. Res. Not (2008).

  51. 51.

    Novokshenov, VYu.: Time asymptotics for soliton equations in problems with step initial conditions (Russian) Sovrem. Mat. Prilozh., Asimptot. Metody Funkts. Anal. 5, 138–168 (2005). (translation in J. Math. Sci. (N. Y.) 1255 717–749)

  52. 52.

    Novoksenov, VYu.: Asymptotic behavior as \(t\rightarrow \infty \) of the solution of the Cauchy problem for a nonlinear Schrodinger equation (Russian). Dokl. Akad. Nauk SSSR 251(4), 799–802 (1980)

    Google Scholar 

  53. 53.

    Novokshenov, VYu.: Asymptotic formulae for the solutions of the system of nonlinear Schrodinger equations. Uspekhi Matem Nauk 37(2), 215–216 (1982)

    Google Scholar 

  54. 54.

    Novokshenov, VYu.: Asymptotics as \(t\rightarrow \infty \) of the Solution to a Two-Dimentional Generalisation of the Toda Lattice. Doklady AN SSSR 265/6, 1320–1324 (1982). (translation in Soviet Math. Dokl26/1 264–268)

    Google Scholar 

  55. 55.

    Shabat, A.B.: An inverse scattering problem. (Russian). Differentsialnye Uravneniya 15(10), 1824–1834 (1918)

    Google Scholar 

  56. 56.

    Trogdon, Thomas, Olver, Sheehan, Deconinck, Bernard: Numerical inverse scattering for the Kortewegde Vries and modified Kortewegde Vries equations. Physica D 241(11), 10031025 (2012)

    Google Scholar 

  57. 57.

    Vanlessen, M.: Strong asymptotics of Laguerre-type orthogonal polynomials and applications in random matrix theory. Constr. Approx. 25, 125–175 (2007)

    Google Scholar 

  58. 58.

    Wadati, M.: The modified Korteweg–de Vries equation. J. Phys. Soc. Jpn. 34, 1289–1296 (1973)

    Google Scholar 

  59. 59.

    Zaharov, V.E., Shabat, A.B.: A plan for integrating the nonlinear equations of mathematical physics by the method of the inverse scattering problem. I. (Russian). Funkcional. Anal. i Prilozen. 8(3), 43–53 (1974)

    Google Scholar 

  60. 60.

    Zhou, X.: The Riemann–Hilbert problem and inverse scattering. SIAM J. Math. Anal. 20, 966–986 (1989)

    Google Scholar 

Download references

Author information



Corresponding author

Correspondence to A. Minakov.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix: A Asymptotics of modified Laguerre polynomials

Appendix: A Asymptotics of modified Laguerre polynomials

In this section we prove Lemma 5.1 and hence Corollary 5.1. Large \(n\rightarrow \infty \) asymptotics of the RHP as in Corollary 5.1 with \(\Lambda = n\) was studied in [57], but that treatment is not suitable for our present purposes.

In order to study large \(n\rightarrow \infty \) asymptotics, first we introduce a new matrix-valued function

$$\begin{aligned} W(\zeta )=\mathrm {e}^{n\ell \sigma _3}Y(\zeta )\mathrm {e}^{n(g(\zeta )-\ell )\sigma _3}, \end{aligned}$$

where \(g(\zeta )\) is the function analytic in \({\mathbb {C}}{\setminus }(-\infty ,a]\) defined by:

$$\begin{aligned} g(\zeta )= & {} \frac{2}{\pi a}\int \limits _{0}^a\ln (\zeta -s)\sqrt{\frac{a-s}{s}}\mathrm {d}s=\frac{-2}{a}\sqrt{\zeta (\zeta -a)} - \ln a\nonumber \\&+\ln \left( 2\zeta -a+2\sqrt{\zeta (\zeta -a)}\right) +\frac{2\zeta }{a}+\ell , \end{aligned}$$

A simple computation shows that \(g(\zeta )\) has the asymptotic expansion

$$\begin{aligned} g(\zeta )=\ln \zeta -\frac{a}{4\zeta }-\frac{a^2}{16\zeta ^2}-\frac{5a^3}{192\zeta ^3}+\cdots =\ln \zeta -\sum \limits _{j=1}^{\infty }\frac{2\ (2n-1)!}{(n+1)\ n!^2}\frac{a^n}{4^n \zeta ^n}. \end{aligned}$$

The constant \(a, \ell \) are given by

$$\begin{aligned} a=\frac{4n}{\Lambda },\quad \ell :=\ln \frac{a}{4\mathrm {e}}. \end{aligned}$$

At \(\zeta =\infty \) we have \(W(\zeta )=\left( I+{\mathcal {O}}(\zeta ^{-1})\right) , \) and W satisfies the boundary value condition

$$\begin{aligned}&W_+=W_-\begin{pmatrix}\mathrm {e}^{n(g_+-g_-)}&{}0\\ \sqrt{\zeta }\ \mathrm {e}^{n(g_-+g_+-2l-\frac{\Lambda \zeta }{n})}&{}\mathrm {e}^{-n(g_+-g_-)}\end{pmatrix},\zeta \in (0,+\infty ),\\&\qquad \qquad W_+=W_-\mathrm {e}^{n(g_+-g_-)\sigma _3},\zeta \in (-\infty ,0). \end{aligned}$$

Further, we introduce the “effective” potential

$$\begin{aligned} \varphi (\zeta ):=-\frac{\Lambda }{2n}\zeta +g(\zeta )-\ell , \end{aligned}$$

so that the jump condition can be written in the following form:

$$\begin{aligned} W_+=W_-\begin{pmatrix}\mathrm {e}^{n(\varphi _+-\varphi _-)}&{}0\\ \sqrt{\zeta }\ \mathrm {e}^{n(\varphi _-+\varphi _+)}&{}\mathrm {e}^{-n(\varphi _+-\varphi _-)}\end{pmatrix}. \end{aligned}$$

It is convenient now to scale the interval (0, a) into (0, 1),  i.e.

$$\begin{aligned} W^{(1)}\left( \lambda \right) :=W\left( \zeta \right) ,\qquad \lambda :=\frac{\zeta }{a}=\frac{\zeta \ \Lambda }{4n}. \end{aligned}$$


$$\begin{aligned}&\psi \left( \lambda \right) := \varphi (\zeta )=-2\int \limits _{1}^{\lambda }\sqrt{\frac{s-1}{s}}\ \mathrm {d}s=-2\sqrt{\lambda (\lambda -1)}+\ln \left( -1+2\lambda +2\sqrt{\lambda (\lambda -1)}\right) \\&\quad =-2\lambda +\ln \lambda +\ln 4e-h(\lambda )=-2\lambda +\ln \lambda +\ln 4e+{\mathcal {O}}(\lambda ^{-1}), \end{aligned}$$

where h(.) is as in (92). The scaled effective potential \(\psi (\lambda ),\) which is analytic in \(\lambda \in {\mathbb {C}}{\setminus }(-\infty ,1],\) has the following properties on the interval \((-\infty ,1]\):

$$\begin{aligned} \psi _+-\psi _-=2\pi \mathrm {i},\quad \lambda \in (-\infty ,0),\qquad \qquad \psi _-+\psi _+ = 0,\ \ \ \lambda \in (0,1). \end{aligned}$$

Due to the above properties (115), function

$$\begin{aligned} W^{(2)}(\lambda ) = a^{\frac{\sigma _3}{4}}W^{(1)}(\lambda )a^{\frac{-\sigma _3}{4}} \end{aligned}$$

does not have jump along \(\lambda \in (-\infty ,0),\) and solves a RHP of the form

$$\begin{aligned}&W^{(2)}_+=W^{(2)}_-\begin{pmatrix}\mathrm {e}^{n(\psi _+-\psi _-)} &{} 0 \\ \sqrt{\lambda } &{} \mathrm {e}^{-n(\psi _+-\psi _-)}\end{pmatrix},\quad \lambda \in (0,1),\\&W^{(2)}_+=W^{(2)}_-\begin{pmatrix}1 &{} 0 \\ \sqrt{\lambda }\ \mathrm {e}^{2n\psi } &{} 1\end{pmatrix},\quad \lambda \in (1,\infty ), \\&W^{(2)}(\lambda )=I+\mathrm {O}\left( \frac{1}{\lambda }\right) . \end{aligned}$$

Tracking back the relation between \(W^{(2)}(\lambda )\) and \(W^{(1)}(\lambda ),\)\(W(\zeta )\), \(Y(\zeta ),\)\(L(\zeta ),\) we see that

$$\begin{aligned} W^{(2)}(\lambda ) = \left( \frac{\sqrt{2}\ n^{n+\frac{1}{4}}}{\mathrm {e}^n}\right) ^{\sigma _3}L(4n\lambda )(4n\lambda )^{n\sigma _3}\mathrm {e}^{-nh(\lambda )\sigma _3}\left( \frac{\mathrm {e}^n}{\sqrt{2}\ n^{n+\frac{1}{4}}}\right) ^{\sigma _3}, \end{aligned}$$

Important role now is played by the signature table of \({\text {Re}}\psi .\) It is shown in the Fig. 6.

Fig. 6

Distribution table of signs for \(\mathrm {Re}\ \psi (\lambda )\)

Now we introduce the \(\delta -\) transformation, which removes the term \(\sqrt{\lambda }\) from the jump matrix. It must solve the following scalar conjugation problem:

$$\begin{aligned} \delta _+\delta _-=\sqrt{\lambda },\ \lambda \in (0,1). \end{aligned}$$

The solution is given by the formula

$$\begin{aligned}&\delta (\lambda )=\exp \left\{ \frac{R(\lambda )}{2\pi \mathrm {i}}\int \limits _{0}^{1}\frac{\ln \sqrt{s}\ \mathrm {d}s}{(s-\zeta )\ R_+(s)}\right\} ,\quad R(\lambda )=\sqrt{\lambda (\lambda -1)},\\&\delta (\lambda )=\left( \frac{\lambda }{2\lambda -1+2\sqrt{\lambda (\lambda -1)}}\right) ^{1/4},\quad \delta (\infty ) = \frac{1}{\sqrt{2}}. \end{aligned}$$

With a series of transformations, we now show how to remove the oscillating jump from the interval (0, 1). First of them is

$$\begin{aligned} W^{(3)}(\lambda ):=\delta ^{\sigma _3}(\infty )W^{(2)}(\lambda )\delta ^{-\sigma _3}(\lambda ). \end{aligned}$$

The jump matrix \(W^{(3)}_+=W^{(3)}_+J^{(3)}\) is

$$\begin{aligned} J^{(3)}=\begin{pmatrix}\frac{\delta _-}{\delta _+}\mathrm {e}^{n(\psi _+-\psi _-)}&{}0\\ \mathrm {e}^{n(\psi _-+\psi _+)}&{}\frac{\delta _+}{\delta _-}\mathrm {e}^{-n(\psi _+-\psi _-)}\end{pmatrix},\ \lambda \in (0,1), \qquad J^{(3)}=\begin{pmatrix}1&{}0\\ \frac{\sqrt{\lambda }}{\delta ^2}\ \mathrm {e}^{2n\psi }&{}1\end{pmatrix},\ \lambda \in (1,+\infty ). \end{aligned}$$

Now we need to factorize the jump matrix on the interval [0, 1] according to the signature table, namely

$$\begin{aligned} \begin{pmatrix}\frac{\delta _-}{\delta _+}\mathrm {e}^{n(\psi _+-\psi _-)}&{}0\\ \mathrm {e}^{n(\psi _-+\psi _+)}&{}\frac{\delta _+}{\delta _-}\mathrm {e}^{-n(\psi _+-\psi _-)}\end{pmatrix} = \begin{pmatrix} 1 &{} \frac{\delta _-^2\ \mathrm {e}^{-2n\psi _-}}{\sqrt{\lambda }} \\ 0 &{} 1 \end{pmatrix} \begin{pmatrix} 0 &{} -\mathrm {e}^{-n(\psi _-+\psi _+)} \\ \mathrm {e}^{n(\psi _-+\psi _+)} &{} 0 \end{pmatrix} \begin{pmatrix} 1 &{} \frac{\delta _+^2\ \mathrm {e}^{-2n\psi _+}}{\sqrt{\lambda }} \\ 0 &{} 1 \end{pmatrix}. \end{aligned}$$

The next step is to “open lenses”, we transform the RH problem, “moving” the left and the right factor in the above formula to the lower and upper half-planes, respectively. For a new RH problem \(W^{(4)}\) the jumps \(W^{(4)}_+=W^{(4)}_-J^{(4)}\) are

$$\begin{aligned} J^{(4)} = {\left\{ \begin{array}{ll}\begin{pmatrix} 1 &{} \frac{\delta ^2\ \mathrm {e}^{-2n\psi }}{\sqrt{\lambda }} \\ 0 &{} 1 \end{pmatrix},\quad \lambda \in l_u (\text {the upper lense}), \quad \begin{pmatrix} 1 &{} \frac{\delta ^2\ \mathrm {e}^{-2n\psi }}{\sqrt{\lambda }} \\ 0 &{} 1 \end{pmatrix},\quad \lambda \in l_d (\text {the lower lense}), \\ \begin{pmatrix} 0 &{} -1 \\ 1 &{} 0 \end{pmatrix},\quad \lambda \in (0,1), \quad \begin{pmatrix}1&{}0\\ \frac{\sqrt{\lambda }}{\delta ^2}\ \mathrm {e}^{2n\psi }&{}1\end{pmatrix},\quad \lambda \in (1,+\infty ). \end{array}\right. } \end{aligned}$$

The RHP for \(W^{(4)}\) gives us the first statement of Lemma 5.1 concerning function

$$\begin{aligned} {\mathcal {Q}}_n(\lambda ):=W^{(4)}(\lambda )\left( \delta (\lambda )\delta ^{-1}(\infty )\right) ^{\sigma _3}. \end{aligned}$$

Indeed, the fact that \({\mathcal {Q}}_n(\lambda )\) decomposees into an asymptotic series follows, for example, from an explicit representation (62) of function L(.). Then RHP for \(W^{(4)}\) gives us uniformity with respect to \(n\ge 1\) of all the ingredients in this asymptotic series. Further, since for large \(n\rightarrow \infty \) the jump matrix is concentrated mainly on the interval (0, 1),  the solution to the RH problem for \(W^{(4)}\) is close everywhere except for a neighborhood of the points 0,  1 to the solution

$$\begin{aligned}&M_{mod}=\begin{pmatrix}a(\lambda )&{}\mathrm {i}b(\lambda )\\ i b(\lambda )&{}a(\lambda )\end{pmatrix},\quad a(\lambda )=\frac{1}{2}\left( \gamma (\lambda )+\gamma ^{-1}(\lambda )\right) ,\ b(\lambda )\\&\quad =\frac{1}{2}\left( \gamma (\lambda )-\gamma ^{-1}(\lambda )\right) ,\ \gamma (\lambda )=\root 4 \of {\frac{\lambda -1}{\lambda }} \end{aligned}$$

to the model problem with jumps only on (0, 1). This gives us the second statement of Lemma 5.1 concerning function

$$\begin{aligned} {\mathcal {E}}_n(\lambda ) := W^{(4)}(\lambda )M_{mod}^{-1}(\lambda ). \end{aligned}$$

However, rigorous proof of the fact that ingredients in asymptotic series for \({\mathcal {E}}_n\) tend to 0 as \(n\rightarrow \infty \) involves construction of local parametrices around the points 0, 1,  since jump matrix for \(W^{(4)}\) is not uniformly close to I in vicinities of these points. The details can be found for example in [57].

A.1 Airy Parametrix

For the benefit of the reader we recall the form of the so–called “Airy” parametrix, which is used to provide local solutions of a large class of Riemann Hilbert problems [23]. This is defined by

$$\begin{aligned} \Psi _{Ai}(\zeta ) = \left\{ \begin{array}{l} \begin{pmatrix}v_1&{}v_0\\ v'_1&{}v'_0\end{pmatrix},\arg \zeta \in (0,2\pi /3), \quad \begin{pmatrix}v_{-1}&{}v_0\\ v'_{-1}&{}v'_0\end{pmatrix},\arg \zeta \in (0,-2\pi /3), \\ \begin{pmatrix}v_1&{}-\mathrm {i}v_{-1}\\ v'_1&{}-\mathrm {i}v'_{-1}\end{pmatrix},\arg \zeta \in (2\pi /3,\pi ), \quad \begin{pmatrix}v_{-1}&{}\mathrm {i}v_{1}\\ v'_{-1}&{}\mathrm {i}v'_{1}\end{pmatrix},\arg \zeta \in (-\pi ,-2\pi /3), \end{array}\right\} \times \mathrm {e}^{\pi \mathrm {i}\sigma _3/4}. \end{aligned}$$

where we have used the notation

$$\begin{aligned} v_0(\zeta )=\sqrt{2\pi }Ai(\zeta ),\quad v_1(\zeta )=\sqrt{2\pi }\mathrm {e}^{-\pi \mathrm {i}/6}Ai(\zeta \mathrm {e}^{-2\pi \mathrm {i}/3}),\quad v_{-1}(\zeta )=\sqrt{2\pi }\mathrm {e}^{\pi \mathrm {i}/6}Ai(\zeta \mathrm {e}^{2\pi \mathrm {i}/3}). \end{aligned}$$

We have

$$\begin{aligned} v_0-\mathrm {i}v_1+\mathrm {i}v_{-1}\equiv 0. \end{aligned}$$

This function \(\Psi _{Ai}(\zeta )\) has the following jumps \(\Psi _{Ai,+}=\Psi _{Ai,-}J_{\Psi _{Ai}}\):

$$\begin{aligned} J_{\Psi _{Ai}}=\begin{pmatrix}1&{}0\\ 1&{}1\end{pmatrix},\arg \zeta =0,\ \begin{pmatrix}1&{}1\\ 0&{}1\end{pmatrix},\arg \zeta =\frac{2\pi }{3}, \ \begin{pmatrix}1&{}1\\ 0&{}1\end{pmatrix},\arg \zeta =\frac{-2\pi }{3},\ \begin{pmatrix}0&{}-1\\ 1&{}0\end{pmatrix},\arg \zeta =\pi . \end{aligned}$$

and the following behaviour at infinity:

$$\begin{aligned} \Psi _{Ai}(\zeta )=\zeta ^{-\sigma _3/4}\frac{1}{\sqrt{2}}\begin{pmatrix}1&{}1\\ 1&{}-1\end{pmatrix}\left( I+ \begin{pmatrix} \frac{-1}{48}&{}\frac{-1}{8}\\ \frac{1}{8} &{} \frac{1}{48} \end{pmatrix}\zeta ^{-3/2}+{\mathcal {O}}(\zeta ^{-3})\right) \mathrm {e}^{\frac{2}{3}\zeta ^{3/2}\sigma _3}\mathrm {e}^{\pi \mathrm {i}\sigma _3/4}. \end{aligned}$$

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Bertola, M., Minakov, A. Laguerre polynomials and transitional asymptotics of the modified Korteweg–de Vries equation for step-like initial data. Anal.Math.Phys. 9, 1761–1818 (2019).

Download citation