Singularities of solutions to chemotaxis systems

DOI10.1515/9783110599534MaRDI QIDQ2332801zbMATH OpenOpenAlexFDO

Publication date 5 November 2019

Published in De Gruyter Series in Mathematics and Life Sciences (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1515/9783110599534

finite-time blow-up fractional diffusion global existence parabolic-elliptic Keller-Segel system singular steady state

Research exposition (monographs, survey articles) pertaining to partial differential equations (35-02) Singularity in context of PDEs (35A21) Asymptotic behavior of solutions to PDEs (35B40) Blow-up in context of PDEs (35B44) Comparison principles in context of PDEs (35B51) Initial value problems for second-order parabolic systems (35K45) Nonlinear parabolic equations (35K55) PDEs in connection with biology, chemistry and other natural sciences (35Q92) Fractional partial differential equations (35R11) Cell movement (chemotaxis, etc.) (92C17)

Review text

In this book the author mainly studies the chemotaxis system \[ \begin{cases} u_t + (-\Delta)^{\frac{\alpha}{2}} u + \nabla \cdot (u \nabla v) =0, & \quad (x,t) \in \mathbb{R}^d \times (0,\infty), \\ \Delta v + u =0, & \quad (x,t) \in \mathbb{R}^d \times (0,\infty), \\ u(x,0) = u_0(x), & \quad x \in \mathbb{R}^d, \end{cases}\tag{1} \] where \(d \ge 2\) and the diffusion term is either the linear Laplacian with \(\alpha =2\) (then (1) is the minimal parabolic-elliptic Keller-Segel system) or the fractional Laplacian with \(\alpha \in (0,2)\). The introduction contains a review of some important properties of (1) along with a lot of references where the reader can find more details concerning these aspects. Moreover, basic properties of the semigroups generated by the (fractional) Laplacian are presented. The forthcoming five chapters contain a presentation of a number of recent results on (1) and the chapters are written such that each can be read independently. Chapter 1 is focused on the existence and uniqueness of local solutions as well as on the global-in-time existence of solutions with small initial data. For \(\alpha =2\) this is proved for mild solutions in appropriate Morrey spaces of the type \(M^p_1 (\mathbb{R}^d)\) along with parabolic regularization properties in the first section. In the forthcoming sections, the case \(\alpha \in (1,2)\) is discussed in a similar setting. Therein, the optimality of the Morrey spaces in the existence results is shown with the properties of the singular steady state \(u_C \in M_1^{d/\alpha} (\mathbb{R}^d)\), its existence being proved as well. The chapter is closed with global existence results for initial data being small either in Morrey spaces \(M^p_q\) with \(q>1\) or in appropriate weighted spaces with slowly decaying initial data as \(|x| \to \infty\). In Chapter 2 the global existence of radially symmetric solutions with large initial data is studied. For \(\alpha=2\) and \(d \ge 3\) this is proved in the first section for initial data being strictly below the singular steady state \(u_C\) in the sense of radial concentrations. The proof particularly relies on the derivation of a single PDE fulfilled by the radial distribution function \(M(r,t) = \int_{B_r(0)} u(x,t) dx\). In addition, several decay estimates implying the convergence to zero of \(u(t)\) as \(t \to \infty\) in different norms are proved. In the forthcoming six sections the case \(\alpha < 2\) and \(d \ge 2\) is studied. Again the global existence is proved for initial data being strictly below the singular steady state, now in the pointwise sense for \(\alpha \in (0,1)\) and in the sense of certain averages for \(\alpha \in (1,2)\). For the proof, in particular comparison principles in the pointwise sense (in the former case) and in an averaged sense (in the latter case) are established, and for \(\alpha \in (0,1)\) also a regularization of (1) is used. The chapter is closed with a global existence result for the parabolic-parabolic Keller-Segel system with \(\alpha=2\) and \(d=2\), the second equation in (1) being replaced by \(\varepsilon v_t = \Delta v -\gamma v +u\) with \(\varepsilon >0\) and \(\gamma \ge 0\), which for given \(u_0\) is valid for any \(\varepsilon\) large enough without any restriction on the mass of \(u_0\). Finally, comments on stationary and selfsimilar solutions for (1) with \(\alpha=2\) are given. Chapters 3 and 4 are concerned with results for finite-time blow-up which show in particular that the results of Chapter 2 are nearly optimal. The proofs in Chapter 3 are mainly based on the analysis of specific moments of \(u\). In the first section, finite-time blow-up for initial data having either large mass or being sufficiently concentrated in the sense of a small moment \(\int_{\mathbb{R}^d} |x|^\gamma u_0(x) dx\) with \(\gamma \in (1,\alpha)\) is proved for a more general problem containing only the first PDE of (1) with \(\nabla ((-\Delta)^{-\beta/2} u)\) instead of \(\nabla v\), where \(\alpha \in (1,2]\) and \(\beta \in (1,d]\) (\(\beta=2\) corresponds to (1)). The proof is based on a new virial identity for a regularization of the mentioned moment. In the forthcoming four sections some new criteria for finite-time blow-up with \(d \ge 2\) and new results concerning the amount of mass concentration of such solutions with \(d=2\) along with some new and simple proofs for already known results are presented for (1) with \(\alpha =2\). The proofs rely on the analysis of certain local moments of \(u\) instead of usual moments in previous proofs. These methods are then extended in Sections 3.6 and 3.8 to the case \(\alpha \in (0,2)\) and a criterion for finite-time blow-up in terms of a large Morrey norm \(M^{d/ \alpha}(\mathbb{R}^d)\) is proved which is nearly complementary to the corresponding boundedness results. In Section 3.7, for \(d=\alpha=2\) and (1) with chemoattractant consumption, i.e. \(-\gamma v\) with \(\gamma \ge 0\) is added on the left-hand side of the second equation in (1), a finite-time blow-up result for sufficiently concentrated initial data as well as a global existence result for \(\gamma\) large are proved. In Chapter 4, further finite-time blow-up results are proved based on Fujita's idea for the semilinear heat equation. In the first section for \(\alpha=2\) blow-up conditions depending either on the largeness of the solution to the heat equation with initial data \(u_0\) or on the radial distribution function of \(u_0\) are proved. These results are then extended to the case \(\alpha \in (0,2)\) in the second section. The final Chapter 5 contains several additional remarks and comments, including further results for a more general problem as well as conjectures and open problems.

Recommendations

Cited in

(28)

This page was built for publication: Singularities of solutions to chemotaxis systems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2332801)