Abstract
In this paper, we consider the Cauchy problem for the Burgers equation in the line. We shall prove that this problem is ill-posed in the Sobolev space \(H^s({\mathbb {R}})\) with \(1\le s<\frac{3}{2}\) in the sense of “norm inflation” by constructing an explicit example of initial data.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
1.1 The Concept of Well-posedness
We say that the Cauchy problem
is locally well-posed in a Banach space X if the following three conditions hold
-
1.
(Local existence) For any initial data \(u_0\in X\), there exists a short time \(T = T(u_0) > 0\) and a solution \(\textbf{S}_{t}(u_0)\in \mathcal {C}([0,T),X)\) to the Cauchy problem (1.1);
-
2.
(Uniqueness) This solution \(\textbf{S}_{t}(u_0)\) is unique in the space \(\mathcal {C}([0,T),X)\);
-
3.
(Continuous Dependence) The data-to-solution map \(u_0 \mapsto \textbf{S}_{t}(u_0)\) is continuous in the following sense: for any \(T_1 < T\) and \(\varepsilon > 0\), there exists \( \delta > 0\), such that if \(\Vert u_0-\widetilde{u}_0\Vert _{X}\le \delta \), then \(\textbf{S}_{t}(\widetilde{u}_0)\) exists up to \(T_1\) and
$$\begin{aligned} \Vert \textbf{S}_{t}(u_0)-\textbf{S}_{t}(\widetilde{u}_0)\Vert _{\mathcal {C}([0,T),X)}\le \varepsilon . \end{aligned}$$
The problem is said to be ill-posed in X if it is not well-posed in the above sense. Based on the definition of well-posedness, at least three types of ill-posedness were studied in the literature: nonexistence, non-uniqueness, and discontinuous dependence on the data. In this paper we are interested in discontinuity with respect to the data.
1.2 The Burgers equation
The Burgers equation with fractional dissipation is written as
where \(\gamma \in [0,2]\) and the fractional power operator \(\Lambda ^{\gamma }\) is defined by Fourier multiplier with the symbol \(|\xi |^{\gamma }\)
The Burgers equation (1.2) with \(\gamma =0\) and \(\gamma =2\) has received an extensive amount of attention since the studies by Burgers in the 1940s. If \(\gamma =0\), the equation is perhaps the most basic example of a PDE evolution leading to shocks. If \(\gamma =2\), it provides an accessible model for studying the interaction between nonlinear and dissipative phenomena. Kiselev et al. [8] gave a complete study for general \(\gamma \in [0,2]\) for the periodic case. In particular, for the case \(\gamma =1\), they proved the global well-posedness of the equation in the critical Hilbert space \(H^{\frac{1}{2}}({\mathbb {T}})\) by using the method of modulus of continuity. Subsequently, Miao-Wu [12] proved the global well-posedness of the critical Burgers equation in critical Besov spaces \(B^{1/p}_{p,1}({\mathbb {R}})\) with \(p\in [1,\infty )\) with the help of Fourier localization technique and the method of modulus of continuity. For more results on the fractional Burgers equation and dispersive perturbations of Burgers equations, we refer the readers to see [1, 3, 9,10,11] and the references therein. We should mention that Molinet et al. [11] proved that the Cauchy problem for a class of dispersive perturbations of Burgers equations is locally well-posed in \(H^s({\mathbb {R}})\).
In this paper, we focus on the well-posedness problem of the following Burgers equation as the most simple quasilinear symmetric hyperbolic equation.
Kato [6] demonstrated that the flow map cannot be of Hölder continuous type. Roughly speaking, (1.3) can be viewed as the simplest in the family of partial differential equations modeling the Euler and Navier-Stokes equation nonlinearity. The local well-posedness of the Burgers equation (1.3) for data in \(H^s({\mathbb {R}})\) with any \(s>3/2\) can be proved by combining the Sobolev embedding \(H^{s-1}({\mathbb {R}})\hookrightarrow L^\infty ({\mathbb {R}})\) and the classical energy estimate
Also, in [13] it is obtained that the solution map is continuous dependence while not uniformly continuous dependence on initial data for the Burgers equation (1.3) in the same space \(H^s({\mathbb {R}})\) with \(s>3/2\). For the endpoint case, Linares et al. in [10] proved that the Cauchy problem for (1.3) is ill-posed in \(H^{3/2}({\mathbb {R}})\), where the key point is that the available local well-posedness theory in \(H^s({\mathbb {R}})\) with any \(s>3/2\) have be used, for more details see Remark 1.6. Using the idea developed in [10], Guo et al. in [4] proved the ill-posedness for the Camassa-Holm equation in the critical Sobolev space \(H^{3/2}({\mathbb {R}})\) and even in the Besov space \(B^{1+1/p}_{p,r}({\mathbb {R}})\) with \(r>1\).
In this paper, we shall prove that the Cauchy problem for the Burgers equation (1.3) in the Sobolev space \(H^s({\mathbb {R}})\) with \(1\le s<\frac{3}{2}\) is ill-posed in the sense of “norm inflation" by constructing an explicit example of initial data.
1.3 Main Result
Now let us state our main ill-posedness result of this paper.
Theorem 1.1
Let \(1\le s<\frac{3}{2}\). For any \(\delta >0\), there exists initial data satisfying
such that a solution \(u(t)\in \mathcal {C}([0,T_0];H^s)\) of the Cauchy problem (1.3) satisfies
Remark 1.1
Theorem 1.1 indicates that the Burgers equation is ill-posed in \(H^s({\mathbb {R}})\) with \(1\le s<\frac{3}{2}\) by exhibiting a strong discontinuity with respect to the initial data known as a norm inflation.
Strategies to Proof. We shall outline the main ideas in the proof of Theorem 1.1.
-
Firstly, we construct an explicit example for initial data \(u_0\), where the norm \(\Vert u_0\Vert _{H^s}\) is sufficiently small while \(\Vert u'_0\Vert _{L^\infty }\) can be large enough.
-
Secondly, we express the solution to the Burgers equation (1.3) by exploring fully the properties of the flow map and give the explicit blow-up time \(T^*\).
-
Lastly, we mainly observe that the transport term does cause growth of the \(L^2\)-norm of \(u_x\) as t tends to \(T^*\). Precisely speaking, we estimate the \(L^2\)-norm of \(u_x\) over \((-\psi (t,q_0),\psi (t,q_0))\) and obtain that its lower bound can be arbitrarily large as t tends to \(T^*\).
The structure of the paper. In Section 2 we provide several key Lemmas. In Section 3 we present the proof of Theorem 1.1.
Let us complete this section with some notations we shall use throughout this paper.
Notations. The notation \(A\le a\wedge b\) means that \(A \le a\) and \(A \le b\). \(a\approx b\) means \(C^{-1}b\le a\le Cb\) for some positive harmless constants C. Given a Banach space X, we denote its norm by \(\Vert \cdot \Vert _{X}\). For \(I\subset {\mathbb {R}}\), we denote by \(\mathcal {C}(I;X)\) the set of continuous functions on I with values in X. For all \(f\in \mathcal {S}'\), the Fourier transform \({\widehat{f}}\) is defined by
We denote \(J^{s}:=(1-\partial _x^2)^{\frac{s}{2}}\). For \(s\in {\mathbb {R}}\), the nonhomogeneous Sobolev space is defined by
2 Preliminary
In the section, we make some preparations for the proof of the main theorem.
2.1 Key Example for Initial Data
Firstly, we construct an explicit example as follows. Set
where two positive numbers \(p_0\) and \(q_0\in (0,1)\) will be fixed later.
It is easy to check that \(u_0(x)\) is an odd function. Furthermore, we can deduce that the following result holds:
Lemma 2.1
For every \(q_0\in (0,1)\) and \(s\in (\frac{1}{2},\frac{3}{2})\), there exists \(C=C_s>0\) such that
Proof
The proof essentially follows that of Lemma 3.1 in [2] or Proposition 1 in [5]. For the sake of readability, we sketch the proof here. Defining the function
then using the fact \(\widehat{e^{-|x|}}(\xi )=2 /\left( 1+\xi ^{2}\right) \), we have
Using the definition of the \(H^s\)-norm and the change of variable setup \(y=q_0\xi \), we have
where we have used \(s\in \left( \frac{1}{2}, \frac{3}{2}\right) \) and \(q_0 \in (0,1)\). This proves the lower bound.
To get the upper bound, we split the domain of integration as
Due to the simple fact \(\sin ^{2}(q_0\xi )\le |q_0\xi |^2\wedge 1\), we have
which completes the proof of Lemma 2.1.
2.2 Existence and Blow-up criterion
Lemma 2.2
For every \(s\in [1,\frac{3}{2})\), there exists a solution \(u\in \mathcal {C}([0,T^*);H^s)\cap L^\infty ([0,T^*);\textrm{Lip})\) for the Burgers equation (1.3), where \(T^*<\infty \) is the maximal time for initial data \(u_0\). Furthermore, we have
Proof
Easy computations give that
from which and Lemma 2.1, we get that \(u_0\in H^s\cap \textrm{Lip}\).
Applying \(J^{s}\) to (1.3) and taking the \(L^2\) inner product of the resulting equality with \(J^{s}u\), then using the following commutator estimate (see [7])
we obtain
which implies that
Applying \(\partial _x\) to (1.3) and taking the inner product of the resulting equality with \(|\partial _xu|^{p-2}\partial _xu\) with \(p\ge 2\), we obtain
which reduces to
Using Gronwall’s inequality and letting \(p=\infty \), we get
This is enough to complete the proof of Lemma 2.2.
2.3 The Equation Along the Flow
Given a Lipschitz velocity field u, we may solve the following ODE to find the flow induced by u:
which is equivalent to the integral form
Furthermore, we get from (1.3) that
which means that
Thus we can give the explicit expression of the flow as
Let \(y=\psi (t,x)\), then we have
Differentiating (1.3) with respect to space variable x, we find
Combining the above and (2.6), we obtain
which reduces to
We should mention that the above can also be deduced from (2.7) and (2.8).
According to the definition of \(u_0\), we can deduce that the maximal existence time of the solution to (1.3) is
Because the velocity field is Lipschitz, then we get that for \(t\in [0,T^*)\)
This shows that \(\psi (t,\cdot )\) is an increasing diffeomorphism over \({\mathbb {R}}\), that is, for all \(x,y\in {\mathbb {R}},\) there holds that \(\psi (t,x)<\psi (t,y)\) if \(x< y\).
3 Proof of Main Theorem
In this section, we prove Theorem 1.1.
Proof of Theorem 1.1
By the definition of \(u'_0(x)\) and (2.10), we know that \(u_x(t,\psi (t,x))\) is continuous in \([0,T^*)\times (-q_0,q_0)\). We should emphasize that \(u_x(t,x)\) is discontinuous in \([0,T^*)\times {\mathbb {R}}\), but we can claim that \(u_x(t,x)\) is continuous in \([0,T^*)\times (-\psi (t,q_0),\psi (t,q_0))\). In fact, for any \(x,y\in (-\psi (t,q_0),\psi (t,q_0))\), we have \(\psi ^{-1}(t,x),\psi ^{-1}(t,y)\in (-q_0,q_0)\). Moreover, we deduce from (2.9) that for \(t\in [0,T^*)\)
Also, it follows from (2.9) and (1.3) that
Thus, we obtain from (2.5) and (3.11)-(3.12) for \(s,t\in [0,T^*)\) and \(x,y\in (-\psi (t,q_0),\psi (t,q_0))\)
By the Burgers equation \(\partial _tu=-uu_x\), we can deduce that \(\partial _tu(t,x)\) is continuous in \([0,T^*)\times (-\psi (t,q_0),\psi (t,q_0))\). That is \(u(t,x)\in \mathcal {C}^1([0,T^*)\times (-\psi (t,q_0),\psi (t,q_0)))\). Furthermore, one has
The similar argument shows that \(u_x(t,x)\in \mathcal {C}^1([0,T^*)\times (-\psi (t,q_0),\psi (t,q_0)))\).
For notational convenience we now set
Thus, we have
Set \(w(t,x):=u_x(t,x)\), then we obtain from (1.3)
which implies that for \((t,x)\in [0,T^*)\times (-\psi (t,q_0),\psi (t,q_0))\)
Integrating (3.14) with respect to space variable x over \([-\psi (t,q_0),\psi (t,q_0)]\), we have
As \(u_0(x)\) is odd, the solution of Burgers equation satisfies \(u(t,x)=-u(t,-x)\), which tells us that \(w(t,x)=w(t,-x)\). Thus we have
and
Inserting (3.16) and (3.17) into (3.15) yields
To simplify notation let
combining (3.13), then (3.18) reduces to
Solving the above differential inequality gives us that
which implies
Notice that
and
combining the above and (3.19) yields
where we have used that
and in the last step used
By Lemma 2.1, one has
but
if some large \(p_0\) and small \(q_0\) is chosen. In fact, we can take \(p_0\) and \(q_0\) such that
Hence, we can choose \(T_0\in [0,T^*)\) such that
This completes the proof of Theorem 1.1.
References
N. Alibaud, J. Droniou, J. Vovelle, Occurrence and non-appearance of shocks in fractal Burgers equations, J. Hyperbolic Differ. Equ., 4(3), 479-499 (2007).
P. Byers, Existence time for the Camassa-Holm equation and the critical Sobolev index, Indiana Univ. Math. J. 55, 941-954 (2006).
H. Dong, D. Du, D. Li, Finite time singularities and global well-posedness for fractal Burgers equation, Indiana Univ. Math. J. 58 (2009).
Z. Guo, X. Liu, L. Molinet, Z. Yin, Ill-posedness of the Camassa-Holm and related equations in the critical space, J. Differ. Equ., 266, 1698-1707 (2019).
A. Himonas, C. Holliman, K. Grayshan, Norm inflation and ill-posedness for the Degasperis-Procesi equation, Comm. Partial Differ. Equ., 39, 2198-2215 (2014).
T. Kato, The Cauchy problem for quasi-linear symmetric hyperbolic systems, Arch. Ration. Mech. Anal., 58, 181-205 (1975).
T. Kato, G. Ponce, Commutator estimates and the Euler and Navier-Stokes equations, Comm. Pure Appl. Math., 41 (1988), 891–907.
A. Kiselev, F. Nazarov, R. Shterenberg, Blow up and regularity for fractal Burgers equation, Dyn. Partial Differ. Equ., 5, 211-240 (2008).
G. Karch, C. Miao, X. Xu, On convergence of solutions of fractal Burgers equation to wardrare faction waves, SIAM J. Math. Anal., 39, 1536-1549 (2007).
F. Linares, D. Pilod, J.-C. Saut, Dispersive perturbations of Burgers and hyperbolic equations I: local theory, SIAM J. Math. Anal., 46(2), 1505-1537 (2014).
L. Molinet, D. Pilodb, S. Vento, On well-posedness for some dispersive perturbations of Burgers’equation, Ann. I. H. Poincaré-AN, 35, 1719-1756 (2018).
C. Miao, G. Wu, Global well-posedness of the critical Burgers equation in critical Besov spaces, J. Differ. Equ., 247, 1673-1693 (2009).
N. Tzvetkov, Ill-posedness issues for nonlinear dispersive equations, in Lectures on Nonlinear Dispersive Equations, Gakkõtosho, Tokyo, 2006, pp. 63-103.
Acknowledgements
The authors want to thank the anonymous reviewer for his/her careful reading and useful suggestions which greatly improved the presentation of this paper. J. Li is supported by the National Natural Science Foundation of China (11801090 and 12161004) and Jiangxi Provincial Natural Science Foundation (20212BAB211004). Y. Yu is supported by the National Natural Science Foundation of China (12101011). W. Zhu is supported by the National Natural Science Foundation of China (12201118) and Guangdong Basic and Applied Basic Research Foundation (2021A1515111018).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Data availability
Data sharing is not applicable to this article as no new data were created or analyzed in this study.
Conflict of interest
The authors declare that they have no conflict of interest.
Additional information
Communicated by K Sandeep.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Li, J., Yu, Y. & Zhu, W. Ill-posedness for the Burgers equation in Sobolev spaces. Indian J Pure Appl Math 55, 189–197 (2024). https://doi.org/10.1007/s13226-022-00357-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13226-022-00357-z