Global existence versus blow-up in a high dimensional Keller-Segel equation with degenerate diffusion and nonlocal aggregation (Q2258923): Difference between revisions

The authors consider weak solutions to the degenerate parabolic equation \[ \begin{cases} u_t = \Delta u^m - \nabla \cdot \left( u \nabla ((-\Delta)^{-\frac{\beta}{2}} u) \right), &\qquad (x,t) \in \mathbb R^d \times (0,\infty),\\ u(x,0) = u_0(x), & \qquad x \in \mathbb{R}^d.\end{cases} \tag{1} \] It is of Keller-Segel type and includes nonlocal cross-diffusion. It is assumed that \(d \geq 3\), \(\beta \in [2,d)\), \(1<m< 2- \frac{\beta}{d}\) and \(u_0 \in L^1 (\mathbb R^d)\) is nonnegative. The authors study whether weak solutions to (1) exist globally in time or blow up in finite time. Namely, if \(u_0 \in L^1 (\mathbb R^d) \cap L^p (\mathbb R^d) \cap L^m (\mathbb R^d)\) is such that \(\| u_0 \|_{L^p(\mathbb R^d)}\) is small enough with \(p := \frac{d (2-m)}{\beta}\), then the existence of a global weak solution \(u \in L^\infty((0,\infty); L^p (\mathbb R^d))\) to (1) is shown. Moreover, if \(u_0 \in L^1 (\mathbb R^d) \cap L^q (\mathbb R^d)\) with some \(q \geq m\) satisfying \(q \in (p, \frac{d}{\beta-2})\) has bounded second moment and is large in the sense that its free energy is negative, then the existence of a weak solution to (1) which blows up in finite time with respect to the \(L^q(\mathbb R^d)\)-norm is proved. In the smaller range \(m \in ( \frac{2d}{d+ \beta}, 2- \frac{\beta}{d})\), it is shown that the norm \(\| u_0 \|_{L^{\frac{2d}{d+\beta}} (\mathbb R^d)}\) is critical for the occurrence of finite time blow-up. More precisely, the authors prove the existence of a threshold \(\alpha >0\) such that weak solutions to (1) are global for \(\| u_0 \|_{L^{\frac{2d}{d+\beta}} (\mathbb R^d)} < \alpha\) and blow up in finite time for \(\| u_0 \|_{L^{\frac{2d}{d+\beta}} (\mathbb R^d)} > \alpha\). The proof of the latter result relies on the method of moments, while the authors treat the degenerate and nonlocal problem (1) with the help of regularized approximations for which they prove suitable compactness properties.