• PDF
• Cite
• Share
Article Contents  Article Contents

# Hopf bifurcation and pattern formation in a delayed diffusive logistic model with spatial heterogeneity

• * Corresponding author: Junping Shi

Partially supported by a grant from China Scholarship Council, US-NSF grant DMS-1715651, National Natural Science Foundation of China (No.11571257), Science and Technology Commission of Shanghai Municipality (No. 18dz2271000)

• In this paper, we study the Hopf bifurcation and spatiotemporal pattern formation of a delayed diffusive logistic model under Neumann boundary condition with spatial heterogeneity. It is shown that for large diffusion coefficient, a supercritical Hopf bifurcation occurs near the non-homogeneous positive steady state at a critical time delay value, and the dependence of corresponding spatiotemporal patterns on the heterogeneous resource function is demonstrated via numerical simulations. Moreover, it is proved that the heterogeneous resource supply contributes to the increase of the temporal average of total biomass of the population even though the total biomass oscillates periodically in time.

Mathematics Subject Classification: Primary: 35K57; Secondary: 35B10, 35B32, 35B36, 35R10, 92B05, 92D40.

 Citation: • • Figure 1.  The non-homogeneous steady states of Eq (2) when $m(x)$ is a cosine function: (a) $m(x) = \cos(x)+2$; (b) $m(x) = \cos(1.5x)+2$; (c) $m(x) = \cos(2x)+2$. Here $d = 2$ (which is equivalent to $\lambda = 0.5$), $\tau = 0.71<\tau_{0\lambda }\approx0.785$ and initial value $u_{0} = 2$ for all three cases, and the solution converges to the non-homogeneous steady state

Figure 2.  The non-homogeneous steady states of Eq. (2) when $m(x)$ is a sine function: (a) $m(x) = \sin(x)+2$; (b) $m(x) = \sin(1.5x)+1.788$; (c) $m(x) = \sin(2x)+2$. The parameters are the same as in Figure 1, and here $\tau = 0.73<\tau_{0\lambda }\approx0.785$. The solution converges to the non-homogeneous steady state for each case

Figure 3.  The non-homogeneous steady states of Eq. (2) when $m(x)$ is a monotone linear function: (a) $m(x) = 1+x/\pi$; (b) $m(x) = 3-x/\pi$. Here $d = 2$ and $\tau = 0.73<\tau_{0\lambda }$. The solution converges to the positive monotone steady state

Figure 4.  The periodic orbits induced by Hopf bifurcation near the non-homogeneous steady state of Eq. (2) for case that $m(x)$ is cosine function: (a) $m(x) = \cos(x)+2$; (b) $m(x) = \cos(1.5x)+2$; (c) $m(x) = \cos(2x)+2$. Here $d = 2$, and $\tau = 0.82>\tau_{0\lambda }\approx 0.785$

Figure 5.  The periodic orbits induced by Hopf bifurcation near the non-homogeneous steady state of Eq. (2) for case that $m(x)$ is sine function: (a) $m(x) = \sin(x)+2$; (b) $m(x) = \sin(1.5x)+1.788$; (c) $m(x) = \sin(2x)+2$. Here $d = 2$, and $\tau = 0.82>\tau_{0\lambda }\approx 0.785$

Figure 6.  The periodic orbits induced by Hopf bifurcation near the non-homogeneous steady state of Eq. (2) for case that $m(x)$ is monotone linear function: (a) $m(x) = 1+x/\pi$; (b) $m(x) = 3-x/\pi$. Here $d = 2$, and $\tau = 0.82>\tau_{0\lambda }\approx 0.785$

• ## Article Metrics  DownLoad:  Full-Size Img  PowerPoint