Critical threshold for a two-species chemotaxis system with the energy critical exponent Journal: Journal de Mathématiques Pures et Appliquées
Shen Bian · 2025
Read Critical threshold for a two-species chemotaxis system with the energy critical exponent Journal: Journal de Mathématiques Pures et Appliquées by Shen Bian (2025) 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-2511-06343.txt.
Introduction
Shen Bian finds the dividing line between calm spread and collapse in a two-species chemotaxis model. Chemotaxis describes creatures or cells drifting toward chemical signals, here with two species attracting each other through a power-law force. Diffusion follows a porous medium style, tuned so the system's free energy keeps a conformal symmetry.
Special steady states, symmetric and spread out rather than compact, set the threshold. When both species start below the steady states in the stated norms, solutions exist for all time. When both start above them, solutions blow up in finite time. The result gives a clean critical test for global existence in three or more dimensions.
Catalog record
| CREATOR | Shen Bian |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 64 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2511-06343
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Critical threshold for a two-species chemotaxis system with the energy critical exponent Journal: Journal de Mathématiques Pures et Appliquées Shen Bian Abstract We consider a two-species chemotaxis model in ℝ d ( d ≥ 3 ) {\mathbb{R}}^{d}(d\geq 3) featuring nonlinear porous medium-type diffusion and nonlocal attractive power-law interaction. Here, the nonlinear diffusion is chosen to be 1 / m 1 + 1 / m 2 = ( d + 2 ) / d 1/m_{1}+1/m_{2}=(d+2)/d in such a way that the associated free energy is conformal invariant, and there are radially symmetric, non-increasing and non-compactly supported stationary solutions ( U s ( x ) , V s ( x ) ) (U_{s}(x),V_{s}(x)) . We analyze the conditions on initial data ( u 0 , v 0 ) (u_{0},v_{0}) under which attractive forces dominate over diffusion, and further classify the global existence and finite time blow-up of dynamical solutions by virtue of these stationary solutions. Specifically, the solution ( u , v ) ( x , t ) (u,v)(x,t) exists globally in time if the initial data satisfy ‖ u 0 ‖ L m 1 ( ℝ d ) < ‖ U s ‖ L m 1 ( ℝ d ) \|u_{0}\|_{L^{m_{1}}({\mathbb{R}}^{d})}<\|U_{s}\|_{L^{m_{1}}({\mathbb{R}}^{d})} and ‖ v 0 ‖ L m 2 ( ℝ d ) < ‖ V s ‖ L m 2 ( ℝ d ) \|v_{0}\|_{L^{m_{2}}({\mathbb{R}}^{d})}<\|V_{s}\|_{L^{m_{2}}({\mathbb{R}}^{d})} . In contrast, there are blowing-up solutions when ‖ u 0 ‖ L m 1 ( ℝ d ) > ‖ U s ‖ L m 1 ( ℝ d ) \|u_{0}\|_{L^{m_{1}}({\mathbb{R}}^{d})}>\|U_{s}\|_{L^{m_{1}}({\mathbb{R}}^{d})} and ‖ v 0 ‖ L m 2 ( ℝ d ) > ‖ V s ‖ L m 2 ( ℝ d ) \|v_{0}\|_{L^{m_{2}}({\mathbb{R}}^{d})}>\|V_{s}\|_{L^{m_{2}}({\mathbb{R}}^{d})} . 1 Introduction In ecological systems, the interplay between multiple species is often intricately linked to their chemotactic behaviors, where organisms respond to chemical signals in their environment to navigate, interact, or survive. For two-species communities, such dynamics are particularly pronounced: each species may secrete specific signaling molecules, while simultaneously exhibiting directed movement toward the signals produced by themselves or the other species. This coupling of interspecific interactions and chemotactic response forms the core of two-species chemotaxis models. In numerous symbiotic and cooperative ecosystems, a key phenomenon emerges: two species mutually secrete chemical signals and exhibit positive directional movement toward one another. For instance, in ecology, pollinating insects are drawn to flowering plants in response to floral scent signals secreted by the plants; meanwhile, the insects release cuticular pheromones that in turn attract more conspecifics or other pollinators. Through bidirectional chemotaxis, coordination between plant pollen dispersal and insect food acquisition is achieved. In oncology, cancer-associated fibroblasts (CAFs) secrete chemokines to attract T cells to aggregate in tumor regions, and activated T cells also secrete signals that reciprocally attract CAFs. This bidirectional chemotactic dynamic influences the balance between tumor immune killing and immune evasion. In microbiology, nitrogen-fixing bacteria in soil secrete organic acids to attract phosphate-solubilizing bacteria, which in turn release amino acids that attract nitrogen-fixing bacteria. By virtue of bidirectional chemotaxis, these two types of bacteria form symbiotic aggregated patches, enabling the efficient completion of nitrogen and phosphorus nutrient cycling in soil. The core commonalities of these phenomena are as follows: first, the two species each secrete chemical attractants, facilitating mutual directional chemotaxis; second, the diffusion behavior of the species depends on their own density—diffusion weakens at low density to avoid extinction (degenerate diffusion). To quantitatively describe such dynamics, the following system is proposed \displaystyle\left\{\begin{array}[]{ll}u_{t}=\Delta u^{m_{1}}-\nabla\cdot(u\nabla c),&x\in{\mathbb{R}}^{d},t\geq 0,\\ -\Delta c=v,&x\in{\mathbb{R}}^{d},t\geq 0,\\ v_{t}=\Delta v^{m_{2}}-\nabla\cdot(v\nabla z),&x\in{\mathbb{R}}^{d},t\geq 0,\\ -\Delta z=u,&x\in{\mathbb{R}}^{d},t\geq 0,\\ (u,v)(x,0)=(u_{0},v_{0})(x)\geq 0,&x\in{\mathbb{R}}^{d}.\end{array}\right. Here, u(x,t) and v(x,t) represent the densities of two interacting agent types. The diffusion exponents m_{1}>1 and m_{2}>1 determine the diffusion capacity of the agents, which depends on their densities. The bidirectional chemotaxis is captured by the terms -\nabla\cdot(u\nabla c) and -\nabla\cdot(v\nabla z) , reflecting mutual attraction. The chemical concentrations c(x,t) and z(x,t) , emitted by v and u respectively, are given by the nonlocal expressions \displaystyle c=c_{d}\int_{{\mathbb{R}}^{d}}\frac{v(y)}{|x-y|^{d-2}}dy,\quad z=c_{d}\int_{{\mathbb{R}}^{d}}\frac{u(y)}{|x-y|^{d-2}}dy, (6) where the constant is defined by \displaystyle c_{d}=\frac{1}{d(d-2)\alpha_{d}}\quad\text{with}\quad\alpha_{d}=\frac{\pi^{d/2}}{\Gamma(d/2+1)}. (7) The system ( 1 ) admits several a priori estimates. Specifically, its solutions remain nonnegative, and mass conservation holds: \displaystyle M_{1}:=\int_{{\mathbb{R}}^{d}}u(x,t)dx=\int_{{\mathbb{R}}^{d}}u_{0}(x)dx,\quad M_{2}:=\int_{{\mathbb{R}}^{d}}v(x,t)dx=\int_{{\mathbb{R}}^{d}}v_{0}(x)dx. (8) Throughout this paper, the initial data are assumed to satisfy \displaystyle(u_{0},v_{0})\in L_{+}^{1}\cap L^{\infty}({\mathbb{R}}^{d}),\quad\int_{{\mathbb{R}}^{d}}(|x|^{2}u_{0}(x)+|x|^{2}v_{0}(x))dx<\infty. (9) The competition between diffusion and nonlocal aggregation is represented by the free energy \displaystyle F(u,v)=\frac{1}{m_{1}-1}\int_{{\mathbb{R}}^{d}}u^{m_{1}}dx+\frac{1}{m_{2}-1}\int_{{\mathbb{R}}^{d}}v^{m_{2}}dx-c_{d}\iint_{{\mathbb{R}}^{d}\times{\mathbb{R}}^{d}}\frac{u(x)v(y)}{|x-y|^{d-2}}dxdy. (10) This functional satisfies a dissipation principle: \displaystyle\frac{dF(u,v)}{dt}=-\int_{{\mathbb{R}}^{d}}u\left|\frac{m_{1}}{m_{1}-1}\nabla u^{m_{1}-1}-\nabla c\right|^{2}dx-\int_{{\mathbb{R}}^{d}}v\left|\frac{m_{2}}{m_{2}-1}\nabla v^{m_{2}-1}-\nabla z\right|^{2}dx\leq 0. Note that system ( 1 ) possesses two scaling invariances. Indeed, if u and v solve ( 1 ), then the scaled functions \displaystyle\left\{\begin{array}[]{ll}u_{\lambda}(x,t)=\lambda^{\frac{2m_{2}}{m_{1}+m_{2}-m_{1}m_{2}}}u(\lambda x,\lambda^{\frac{2m_{1}}{m_{1}+m_{2}-m_{1}m_{2}}}t),\\ v_{\lambda}(x,t)=\lambda^{\frac{2m_{1}}{m_{1}+m_{2}-m_{1}m_{2}}}v(\lambda x,\lambda^{\frac{2m_{2}}{m_{1}+m_{2}-m_{1}m_{2}}}t)\end{array}\right. also solve ( 1 ). These scalings preserve the L^{p_{1}} and L^{p_{2}} norms: \displaystyle\|u_{\lambda}\|_{L^{p_{1}}({\mathbb{R}}^{d})}=\|u\|_{L^{p_{1}}({\mathbb{R}}^{d})},\quad\|v_{\lambda}\|_{L^{p_{2}}({\mathbb{R}}^{d})}=\|v\|_{L^{p_{2}}({\mathbb{R}}^{d})}, (13) where \displaystyle p_{1}=\frac{(m_{1}+m_{2}-m_{1}m_{2})d}{2m_{2}},\quad p_{2}=\frac{(m_{1}+m_{2}-m_{1}m_{2})d}{2m_{1}}. (14) The above scalings become mass invariant for both u and v if m_{1}=m_{2}=2-2/d . When m_{1}m_{2}+2m_{1}/d=m_{1}+m_{2} , then p_{2}=1 and mass conservation holds for v , whereas p_{1}=1 and mass conservation holds for u when m_{1}m_{2}+2m_{2}/d=m_{1}+m_{2} . It is worth noting that single-species chemotaxis systems have been intensively studied in recent years. For m_{1}=m_{2}>1 and d\geq 2 , the global existence and blow-up of dynamic solutions have been analyzed in [ 4 , 5 , 6 , 21 , 25 , 30 , 36 , 37 ] . Additionally, significant efforts have focused on investigating steady states, as documented in [ 7 , 8 , 9 , 10 , 11 , 13 , 19 ] . For two-species systems, [ 12 ] established results on global existence and finite-time blow-up for m_{1}>1 , m_{2}>1 , and d\geq 3 . Specifically, for m_{1}=m_{2}=2-2/d , there exists a number M_{c}>0 such that solutions exist globally if M_{1}M_{2}<M_{c}^{2} and blow up in finite time if \frac{M_{1}M_{2}}{M_{1}^{2-2/d}+M_{2}^{2-2/d}}>\frac{M_{c}^{2/d}}{2}. For m_{1}m_{2}+2m_{1}/d<m_{1}+m_{2} and m_{1}m_{2}+2m_{2}/d<m_{1}+m_{2} , one can construct large initial data ensuring blow-up in finite time. When m_{1}=m_{2}=1 and d=2, [ 20 ] showed that solutions exist globally if M_{1}M_{2}<4\pi(M_{1}+M_{2}) and blow-up occurs if M_{1}M_{2}>4\pi(M_{1}+M_{2}) . For multi-species populations, [ 23 , 24 ] proved global existence and infinite-time aggregation in subcritical and critical cases. For multi-species populations in {\mathbb{R}}^{2} , [ 23 , 24 ] proved global existence and infinite-time aggregation in subcritical and critical cases. Our main goal in this paper is to classify the global existence and finite-time blow-up of solutions based on the initial data. The main tool for analyzing ( 1 ) is the stationary system, which can be interpreted in the distributional sense as \displaystyle\left\{\begin{array}[]{ll}\frac{m_{1}}{m_{1}-1}U_{s}^{m_{1}-1}-C_{s}=\overline{C}_{1},\text{ in }\Omega_{1},\\ \ \frac{m_{2}}{m_{2}-1}V_{s}^{m_{2}-1}-Z_{s}=\overline{C}_{2},\text{ in }\Omega_{2},\\ \ U_{s}=0\text{ in }{\mathbb{R}}^{d}\setminus\Omega_{1},\quad U_{s}>0\text{ in }\Omega_{1},\\ \ V_{s}=0\text{ in }{\mathbb{R}}^{d}\setminus\Omega_{2},\quad V_{s}>0\text{ in }\Omega_{2},\\ \ C_{s}=c_{d}\int_{{\mathbb{R}}^{d}}\frac{V_{s}(y)}{|x-y|^{d-2}}dy,\quad Z_{s}=c_{d}\int_{{\mathbb{R}}^{d}}\frac{U_{s}(y)}{|x-y|^{d-2}}dy.\end{array}\right. Here, \overline{C}_{1} and \overline{C}_{2} are two constant chemical potentials, and \Omega_{1}=\{x\in{\mathbb{R}}^{d}\,|U_{s}>0\},\Omega_{2}=\{x\in{\mathbb{R}}^{d}\,|V_{s}>0\}. Letting \phi_{1}=\frac{m_{1}-1}{m_{1}}(C_{s}+\overline{C}_{1}) and \phi_{2}=\frac{m_{2}-1}{m_{2}}(Z_{s}+\overline{C}_{2}) , the stationary system reduces to \displaystyle\left\{\begin{array}[]{ll}-\Delta\phi_{1}=\frac{m_{1}-1}{m_{1}}\phi_{2}^{\frac{1}{m_{2}-1}},&\text{ in }\Omega_{2},\\ -\Delta\phi_{2}=\frac{m_{2}-1}{m_{2}}\phi_{1}^{\frac{1}{m_{1}-1}},&\text{ in }\Omega_{1}.\end{array}\right. System ( 1 ) is the well-known Hénon-Lane-Emden system, which has been extensively studied by many authors. The Hénon-Lane-Emden conjecture states that there is no positive classical solution to ( 1 ) in \Omega_{1}=\Omega_{2}={\mathbb{R}}^{d} if and only if 1/m_{1}+1/m_{2}<(d+2)/d . This conjecture is known to be true for radial solutions in all dimensions [ 28 , 33 ] . For non-radial solutions, the conjecture is only fully answered when d\leq 4