2. 北京工业大学 机械工程与应用电子技术学院, 北京 100124
2. College of Mechanical Engineering and Applied Electronics Technology, Beijing University of Technology, Beijing 100124, China
时空斑图是在空间或时间上具有某种规律性的非均匀宏观结构, 在化学[1]、物理学[2]、神经网络[3]、生态学[4]等领域有着重要的应用.特别地, 在生态学中, 为了掌握物种空间分布的多样性, 国内外学者对物种的空间斑图进行了大量而深入的研究[5-10].文献[5]研究了带有自扩散项的系统中Allee效应对斑图形成的影响.文献[6]讨论了带有交叉扩散项的水生生物模型的斑图的形成.文献[7]研究了带有负交叉扩散项的捕食者-食饵模型中斑图的生成与选择, 并得到了不同类型的空间斑图. Sambath等人研究了以下带有Holling-Ⅱ型功能反应和双曲死亡率的捕食者-食饵模型[11]
$ \left\{ \begin{array}{l} \dfrac{{{\rm{d}}}U}{{{\rm{d}}}\tau}=rU\Big(1-\dfrac{U}{K}\Big)- \dfrac{a_{1}UV}{n+U}, \\ \dfrac{{{\rm{d}}}V}{{{\rm{d}}}\tau}=\dfrac{a_{2}UV}{n+U}- \dfrac{\delta V^{2}}{\epsilon+\eta V}.\\ \end{array} \right. $ | (1) |
其中
$ \left\{\!\! \begin{array}{l} \dfrac{\partial U}{\partial\tau}=D_{11}\Delta U+D_{12}\Delta V+rU\Big(1-\dfrac{U}{K}\Big) -\dfrac{a_{1}UV}{n+U}-\dfrac{qEU}{m_{1}E+m_{2}U}, \\ \dfrac{\partial V}{\partial\tau}=D_{21}\Delta U+D_{22}\Delta V+ \dfrac{a_{2}UV}{n+U}-\dfrac{\delta V^{2}}{\epsilon+\eta V}. \end{array} \right. $ | (2) |
其中
为了简化模型, 作如下的无量纲化:
$ u=\dfrac{U}{K}, v=\dfrac{a_{1}}{K}V, t=r\tau, h=\dfrac{qE}{rm_{2}K}, \rho=\dfrac{m_{1}E}{m_{2}K}, \beta=\dfrac{n}{K}, s=\dfrac{1}{r}, \delta=a_{2}\eta, \epsilon=\eta, \alpha=\dfrac{a_{2}}{r}, $ |
$ \gamma=\dfrac{K}{a_{1}}, d_{11}=\dfrac{D_{11}}{r}, d_{12}=\dfrac{D_{12}}{ra_{1}}, d_{21}=\dfrac{a_{1}D_{21}}{r}, d_{22}=\dfrac{D_{22}}{r}. $ |
这样, 系统(2)就转化为
$ \left\{\!\! \begin{array}{l} \dfrac{\partial u(X, t)}{\partial t}=d_{11}\Delta u+d_{12}\Delta v+u(1-u)-\dfrac{suv}{\beta+u}-\dfrac{hu}{\rho+u}, X\in\Omega, t>0, \\[2mm] \dfrac{\partial v(X, t)}{\partial t}=d_{21}\Delta u+d_{22}\Delta v+ \dfrac{\alpha uv}{\beta+u}-\dfrac{\alpha \gamma v^{2}}{1+\gamma v}, X\in\Omega, t>0, \\[2mm] \dfrac{\partial u}{\partial \nu}=\dfrac{\partial v}{\partial \nu}=0, X\in\partial \Omega, t>0, \\[2mm] u(X, 0)=u_{0}(X)\geq0, v(X, 0)=v_{0}(X)\geq0, X\in\Omega.\\ \end{array} \right. $ | (3) |
其中
对于上述系统(3), 文献[13]研究了在自扩散效应下时滞对Hopf分支产生的周期解的稳定性及其方向的影响, 文献[14]讨论了当
考虑系统(3)对应的常微分系统
$ \left\{\!\! \begin{array}{l} \dfrac{{{\rm{d}}}u}{{{\rm{d}}}t}=u(1-u)-\dfrac{suv}{\beta+u}- \dfrac{hu}{\rho+u}, \\ \dfrac{{{\rm{d}}}v}{{{\rm{d}}}t}=\dfrac{\alpha uv}{\beta+u}- \dfrac{\alpha \gamma v^{2}}{1+\gamma v}. \end{array} \right. $ | (4) |
考虑到生态学意义, 本文只研究系统(3)和系统(4)的正平衡点.通过求解, 易知当
$ A_{0}u^{3}+A_{1}u^{2}+A_{2}u+A_{3}=0, $ | (5) |
其中
$ A_{0}=\beta\gamma, A_{1}=\beta^{2}\gamma+\beta\gamma \rho+s-\beta\gamma, A_{2}=\beta^{2}\gamma\rho+\beta\gamma h+\rho s-\beta^{2}\gamma-\beta\gamma\rho, A_{3}=\beta^{2}\gamma(h-\rho). $ |
图 1中, 通过食饵零增长的等倾线和捕食者零增长的等倾线的交点来标记正平衡点
将系统(4)用如下形式来表示:
$ \dot{Y}=F(Y, \Lambda)=(P(u, v), Q(u, v))^{\rm T}, $ | (6) |
其中
$ J=\left( \begin{array}{ll} P_{u}(u_{\ast}, v_{\ast}) & P_{v}(u_{\ast}, v_{\ast})\\ Q_{u}(u_{\ast}, v_{\ast}) & Q_{v}(u_{\ast}, v_{\ast}) \end{array} \right) =\left( \begin{array}{ll} a_{11} & a_{12}\\ a_{21} & a_{22} \end{array} \right), $ | (7) |
其中
$ a_{11}=\dfrac{\partial P}{\partial u}\Big|_{E_{\ast}}=-u_{\ast}+ \dfrac{su_{\ast}^{2}}{\beta\gamma(\beta+u_{\ast})^{2}}+\dfrac{hu_{\ast}} {(\rho+u_{\ast})^{2}}, \quad a_{12}=\dfrac{\partial P}{\partial v}\Big|_{E_{\ast}}= -\dfrac{su_{\ast}}{\beta+u_{\ast}}, \\ a_{21}=\dfrac{\partial Q}{\partial u}\Big|_{E_{\ast}}= \dfrac{\alpha u_{\ast}}{\gamma(\beta+u_{\ast})^{2}}, \quad a_{22}= \dfrac{\partial Q}{\partial v}\Big|_{E_{\ast}}= -\dfrac{\alpha\beta u_{\ast}}{(\beta+u_{\ast})^{2}}. $ |
为了便于下面的讨论, 我们作如下假设.
定理1.1 对于系统(4), 有如下陈述成立.
(ⅰ)若
(ⅱ)若
$ h_{H}=\dfrac{\alpha\beta^{2}\gamma(\rho+u_{\ast})^{2}+ \beta\gamma(\beta+u_{\ast})^{2}(\rho+u_{\ast})^{2}-su_{\ast}(\rho+ u_{\ast})^{2}}{\beta\gamma(\beta+u_{\ast})^{2}}. $ |
证明 (ⅰ)矩阵
$ \lambda^{2}-{\rm tr}_{0}\lambda+{\rm det}J=0, $ | (8) |
其中,
$ {\rm tr}_{0}=-u_{\ast}+\dfrac{su_{\ast}^{2}}{\beta\gamma(\beta+u_{\ast})^{2}}+ \dfrac{hu_{\ast}}{(\rho+u_{\ast})^{2}}-\dfrac{\alpha\beta u_{\ast}}{(\beta+u_{\ast})^{2}}, $ |
$ {\rm det}J=\dfrac{\alpha\beta u_{\ast}^{2}}{(\beta+u_{\ast})^{2}}\Big(1-\dfrac{su_{\ast}}{\beta \gamma(\beta+u_{\ast})^{2}}-\dfrac{h}{(\rho+u_{\ast})^{2}}\Big)+ \dfrac{\alpha su_{\ast}^{2}}{(\beta+u_{\ast})^{3}\gamma}, $ |
tr
(ⅱ)通过以上分析, 若
$ \left\{\!\! \begin{array}{l} p^{2}-q^{2}-{\rm tr}_{0}\cdot p+{\rm det}J=0, \\ 2pq-{\rm tr}_{0}\cdot q=0. \end{array} \right. $ | (9) |
对上述方程组的第二个方程两边关于
$ \dfrac{{{\rm{d}}}p}{{\rm d}h}\Big|_{h=h_{H}}=\dfrac{u_{\ast}}{2(\rho+u_{\ast})^{2}}>0. $ |
因此, 陈述(ⅱ)成立.
下面考虑系统(4)仅加入自扩散时平衡点的稳定性.首先在
$ \left\{\!\! \begin{array}{l} u(x, y, t)=u_{\ast}+U(x, y, t), |U(x, y, t)|\ll u_{\ast}, \\ v(x, y, t)=v_{\ast}+V(x, y, t), |V(x, y, t)|\ll v_{\ast}, \end{array} \right. $ | (10) |
其中,
$ \lambda^{2}+B_{1}\lambda+B_{2}=0, $ | (11) |
其中
由上面的分析可以得到, 若
注1 在系统(4)中仅加入自扩散项得到的偏微分系统, 系统的平衡点
为了得到Turing斑图的形成条件, 接下来我们考虑交叉扩散项对系统动力学的影响.
在介绍定理1.2之前, 作如下假设:
定理1.2 若
证明 记
$ \mid J-k^{2}D-\lambda I\mid=0, $ | (12) |
特征方程(12)的解为如下形式:
$ \lambda_{k}=\dfrac{{\rm tr}_{k}\pm\sqrt{{\rm tr}_{k}^{2}-4\delta_{k}}}{2}, $ | (13) |
其中
$ {\rm tr}_{k}=a_{11}+a_{22}-(d_{11}+d_{22})k^{2}, $ | (14) |
$ \delta_{k}=(d_{11}d_{22}-d_{12}d_{21})k^{4}-(a_{11}d_{22}+ a_{22}d_{11}-a_{12}d_{21}-a_{21}d_{12})k^{2}+{\rm det}J, $ | (15) |
注2 由上述定理可知, 在
图 2讨论了当
接下来, 将讨论Turing斑图的存在区域.选择
$ \dfrac{su_{\ast}}{\beta\gamma(\beta+u_{\ast})^{2}}+ \dfrac{h}{(\rho+u_{\ast})^{2}}-\dfrac{\alpha\beta}{(\beta+u_{\ast})^{2}} =1. $ | (16) |
当Im
$ \Big( \dfrac{su_{\ast}^{2}}{\beta\gamma(\beta+u_{\ast})^{2}}+ \dfrac{hu_{\ast}}{(\rho+u_{\ast})^{2}}-u_{\ast}\Big)d_{22}- \dfrac{\alpha\beta u_{\ast}}{(\beta+u_{\ast})^{2}}d_{11}-\dfrac{\alpha u_{\ast}}{(\beta+u_{\ast})^{2}\gamma}d_{12}+\dfrac{su_{\ast}} {\beta+u_{\ast}}d_{21} \notag\\ -2\sqrt{(d_{11}d_{22}-d_{12}d_{21}){\rm det}J}=0. $ | (17) |
通过Hopf分支曲线和Turing分支曲线, 我们得到了Hopf分支区域和Turing不稳定区域, 如图 3所示. Turing空间位于Turing分支曲线上方以及Hopf分支曲线下方, 当系统(3)中的参数位于此Turing空间内时会出现Turing失稳, 从而出现Turing斑图.
本节中, 主要利用多重尺度分析法推导系统(3)的振幅方程, 获得系统(3)的Turing斑图选择结果, 从而得到Turing斑图的类型.
令
$ \left\{\!\! \begin{array}{l} \dfrac{\partial u}{\partial t}= d_{11}\Delta u+d_{12}\Delta v+a_{11}u+a_{12}v+ \dfrac{1}{2}P_{20}u^{2}+P_{11}uv+\dfrac{1}{2}P_{02}v^{2}+ \dfrac{1}{6}P_{30}u^{3}\\[2mm] \qquad +\dfrac{1}{2}P_{21}u^{2}v+\dfrac{1}{2}P_{12}uv^{2}+ \dfrac{1}{6}P_{03}v^{3}+o(\rho^{3}), \\[2mm] \dfrac{\partial v}{\partial t}= d_{21}\Delta u+d_{22}\Delta v+a_{21}u+a_{22}v+\dfrac{1}{2}Q_{20} u^{2}+Q_{11}uv+\dfrac{1}{2}Q_{02}v^{2}+\dfrac{1}{6}Q_{30}u^{3}\\[2mm] \qquad +\dfrac{1}{2}Q_{21}u^{2}v+\dfrac{1}{2}Q_{12}uv^{2}+ \dfrac{1}{6}Q_{03}v^{3}+o(\rho^{3}), \end{array} \right. $ | (18) |
系统
$ \dfrac{\partial U}{\partial t}=LU+N, $ | (19) |
其中
$ N=\left( \begin{array}{c}\frac{1}{2}P_{20}u^{2}+P_{11}uv+ \frac{1}{2}P_{02}v^{2}+ \frac{1}{6}P_{30}u^{3}+\frac{1}{2}P_{21}u^{2}v+\frac{1}{2} P_{12}uv^{2}+\frac{1}{6}P_{03}v^{3}\\ \frac{1}{2}Q_{20}u^{2}+Q_{11}uv+\frac{1}{2}Q_{02}v^{2}+ \frac{1}{6}Q_{30}u^{3}+\frac{1}{2}Q_{21}u^{2}v+\frac{1} {2}Q_{12}uv^{2}+\frac{1}{6}Q_{03}v^{3} \end{array} \right), $ |
$ L=L_{T}+(h-h_{T})M=\left( \begin{array}{cc} a^{T}_{11}+d_{11}\Delta & a^{T}_{12}+d_{12}\Delta\\[2mm] a^{T}_{21}+d_{21}\Delta & a^{T}_{22}+d_{22}\Delta \end{array} \right)+ (h-h_{T})M, $ |
$ M=\left( \begin{array}{cc} m_{11} & m_{12}\\ m_{21} & m_{22} \end{array} \right) =\left.\left( \begin{array}{cc} \dfrac{\partial a_{11}}{\partial h} & \dfrac{\partial a_{12}} {\partial h}\\[2mm] \dfrac{\partial a_{21}}{\partial h} & \dfrac{\partial a_{22}}{\partial h} \end{array} \right)\right|_{h=h_{T}}. $ |
这里,
$ L_{T}\left( \begin{array}{l} f\\ 1 \end{array} \right) =0, \quad L^{\ast}_{T}\left( \begin{array}{l} 1\\ g \end{array} \right) =0, $ |
其中
在计算中, 我们只分析控制参数在临界值附近的行为, 因此将控制参数
$ h-h_{T}=\varepsilon h_{1}+\varepsilon^{2}h_{2}+\varepsilon^{3}h_{3}+o(\varepsilon^{3}), $ | (20) |
其中
$ U=\left( \begin{array}{l} u\\ v \end{array} \right) =\varepsilon\left( \begin{array}{l} u_{1}\\ v_{1} \end{array} \right) +\varepsilon^{2}\left( \begin{array}{l} u_{2}\\ v_{2} \end{array} \right) +\varepsilon^{3}\left( \begin{array}{l} u_{3}\\ v_{3} \end{array} \right) +o(\varepsilon^{3}), $ |
$ N=\varepsilon^{2}N_{1}+\varepsilon^{3}N_{2}+o(\varepsilon^{3}), $ |
$ N_{1}=\left(\!\! \begin{array}{c} \frac{1}{2}P_{20}u_{1}^{2}+P_{11}u_{1}v_{1}+\frac{1}{2} P_{02}v_{1}^{2}\\ \frac{1}{2}Q_{20}u_{1}^{2}+Q_{11}u_{1}v_{1}+ \frac{1}{2}Q_{02}v_{1}^{2} \end{array} \!\!\right)\Bigg|_{h=h_{T}}, $ |
$ N_{2}\!=\!\left(\!\!\!\! \begin{array}{c} P_{20}u_{1}u_{2}\!+\!P_{11}(u_{2}v_{1}\!+\!u_{1}v_{2})\!+\!P_{02}v_{1} v_{2}\!+\! \frac{1}{6}P_{30}u_{1}^{3}\!+\!\frac{1}{2}P_{21}u_{1}^{2}v_{1} + \frac{1}{2}P_{12}u_{1}v_{1}^{2}\!+\!\frac{1}{6}P_{03}v_{1}^{3}\\ Q_{20}u_{1}u_{2}\!+\!Q_{11}(u_{2}v_{1}\!+\!u_{1}v_{2})\!+\!Q_{02}v_{1} v_{2}\!+\! \frac{1}{6}Q_{30}u_{1}^{3}\!+\!\frac{1}{2}Q_{21}u_{1}^{2}v_{1}\!+\! \frac{1}{2}Q_{12}u_{1}v_{1}^{2} \!+\!\frac{1}{6}Q_{03}v_{1}^{3} \end{array} \!\!\!\!\right)\Bigg|_{h=h_{T}}. $ |
$ \left( \begin{array}{l} u_{1}\\ v_{1} \end{array} \right)= \left( \begin{array}{l} f\\ 1 \end{array} \right) \Bigg(\sum\limits_{j=1}^{3}(W_{j}\exp({\rm i}\overrightarrow{{ k}_{j}}\cdot\overrightarrow{ r}))\Bigg)+c.c., $ |
$ \left( \begin{array}{l} u_{2}\\ v_{2} \end{array} \right)= \left( \begin{array}{l} U_{0}\\ V_{0} \end{array} \right)+ \sum\limits_{j=1}^{3} \left( \begin{array}{l} U_{j}\\ V_{j} \end{array} \right) \exp({\rm i}\overrightarrow{{ k}_{j}}\cdot\overrightarrow{ r})+ \sum\limits_{j=1}^{3} \left( \begin{array}{l} U_{jj}\\ V_{jj} \end{array} \right) \exp(2{\rm i}\overrightarrow{{ k}_{j}}\cdot\overrightarrow{ r}) \\ + \left( \begin{array}{l} U_{12}\\ V_{12} \end{array} \right) \exp({\rm i}(\overrightarrow{{ k}_{1}}-\overrightarrow{{ k}_{2}})\cdot \overrightarrow{ r})+ \left( \begin{array}{l} U_{23}\\ V_{23} \end{array} \right) \exp({\rm i}(\overrightarrow{{ k}_{2}}-\overrightarrow{{ k}_{3}}) \cdot\overrightarrow{ r})\\ + \left( \begin{array}{l} U_{31}\\ V_{31} \end{array} \right) \exp({\rm i}(\overrightarrow{{ k}_{3}}-\overrightarrow{{ k}_{1}})\cdot\overrightarrow{ r})+c.c., $ |
其中
$ \left( \begin{array}{l} U_{0}\\ V_{0} \end{array} \right)= \left( \begin{array}{l} u^{0}\\ v^{0} \end{array} \right) (|W_{1}|^{2}+|W_{2}|^{2}+|W_{3}|^{2}), \quad U_{j}=fV_{j}, \\ \left( \begin{array}{l} U_{jj}\\ V_{jj} \end{array} \right)= \left( \begin{array}{l} u^{1}\\ v^{1} \end{array} \right) W_{j}^{2}, \quad \left( \begin{array}{l} U_{jk}\\ V_{jk} \end{array} \right)= \left( \begin{array}{l} u^{\star}\\ v^{\star} \end{array} \right) W_{j}\overline{W}_{k}, \\ u^{0}=\dfrac{2(l_{1}a^{T}_{22}-l_{2}a^{T}_{12})}{a^{T}_{11}a^{T} _{22}-a^{T}_{12}a^{T}_{21}}, \quad v^{0}=\dfrac{2(l_{2}a^{T}_{11}-l_{1}a^{T}_{21})}{a^{T}_{11} a^{T}_{22}-a^{T}_{12}a^{T}_{21}}, \\ u^{1}=\dfrac{l_{1}(a_{22}^{T}-4d_{22}k_{T}^{\ast2})-l_{2}(a_{12}^{T} -4d_{12}k_{T}^{\ast2})}{(a_{11}^{T}-4d_{11}k_{T}^{\ast2})(a_{22}^{T} -4d_{22}k_{T}^{\ast2})-(a_{12}^{T}-4d_{12}k_{T}^{\ast2})(a_{21}^{T}- 4d_{21}k_{T}^{\ast2})}, \\ v^{1}=\dfrac{l_{2}(a_{11}^{T}-4d_{11}k_{T}^{\ast2})-l_{1}(a_{21}^ {T}-4d_{21}k_{T}^{\ast2})}{(a_{11}^{T}-4d_{11}k_{T}^{\ast2}) (a_{22}^{T}-4d_{22}k_{T}^{\ast2})-(a_{12}^{T}-4d_{12}k_{T}^{\ast2}) (a_{21}^{T}-4d_{21}k_{T}^{\ast2})}, \\ u^{\star}=\dfrac{2l_{1}(a_{22}^{T}-3d_{22}k_{T}^{\ast2})-2l_{2} (a_{12}^{T}-3d_{12}k_{T}^{\ast2})}{(a_{11}^{T}-3d_{11}k_{T}^ {\ast2})(a_{22}^{T}-3d_{22}k_{T}^{\ast2})-(a_{12}^{T}-3d_{12} k_{T}^{\ast2})(a_{21}^{T}-3d_{21}k_{T}^{\ast2})}, \\ v^{\star}=\dfrac{2l_{2}(a_{11}^{T}-3d_{11}k_{T}^{\ast2})-2l_{1} (a_{21}^{T}-3d_{21}k_{T}^{\ast2})}{(a_{11}^{T}-3d_{11}k_{T}^{\ast2}) (a_{22}^{T}-3d_{22}k_{T}^{\ast2})-(a_{12}^{T}-3d_{12}k_{T}^{\ast2}) (a_{21}^{T}-3d_{21}k_{T}^{\ast2})}, \\ l_{1}=-\dfrac{f^{2}}{2}\cdot P_{20}-\dfrac{1}{2}P_{02}-f\cdot P_{11}, \quad l_{2}=-\dfrac{f^{2}}{2}\cdot Q_{20}-\dfrac{1}{2}Q_{02}-f\cdot Q_{11}, $ |
而
利用中心流形理论推导得到如下的振幅方程[15]
$ \left\{ \begin{array}{l} \tau_{0}\dfrac{\partial A_{1}}{\partial t}=\mu A_{1}+l\overline{A_{2}} \overline{A_{3}}-[g_{1}|A_{1}|^{2}+g_{2}(|A_{2}|^{2}+|A_{3}| ^{2})]A_{1}, \\[2mm] \tau_{0}\dfrac{\partial A_{2}}{\partial t}=\mu A_{2}+l \overline{A_{1}}\overline{A_{3}}-[g_{1}|A_{2}|^{2}+g_{2}(|A_{1}| ^{2}+|A_{3}|^{2})]A_{2}, \\[2mm] \tau_{0}\dfrac{\partial A_{3}}{\partial t}=\mu A_{3}+l\overline{A_{1}}\overline{A_{2}}-[g_{1}|A_{3}|^{2}+g_{2}(|A_{1}|^{2}+|A_{2}|^{2})]A_{3}, \\ \end{array} \right. $ | (21) |
其中,
$ \begin{align*} \tau_{0}=&\dfrac{f+g}{h_{T}[fm_{11}+m_{12}+g(fm_{21}+m_{22})]}, \mu=\dfrac{h-h_{T}}{h_{T}}, \\ l=&\dfrac{L}{h_{T}[fm_{11}+m_{12}+g(fm_{21}+m_{22})]}, g_{1}=\dfrac{G_{1}}{h_{T}[fm_{11}+m_{12}+g(fm_{21}+m_{22})]}, \\ g_{2}=&\dfrac{G_{2}}{h_{T}[fm_{11}+m_{12}+g(fm_{21}+m_{22})]}, \end{align*} $ |
其中
$ L=-2l_{1}-2gl_{2}|_{h=h_{T}}, $ |
$ G_{1}=-\Big\{(f\cdot P_{20}+P_{11})(u^{0}+u^{1})+(f\cdot P_{11}+P_{02})(v^{0}+v^{1})+\dfrac{f^{3}}{2}\cdot P_{30}+ \dfrac{1}{2}P_{03}\\ +\dfrac{3f^{2}}{2}\cdot P_{21}+\dfrac{3f}{2}\cdot P_{12}+g\Big[(f\cdot Q_{20}+Q_{11})(u^{0}+u^{1})+(f\cdot Q_{11}+Q_{02})(v^{0}+v^{1})\\ +\dfrac{f^{3}}{2}\cdot Q_{30}+\dfrac{1}{2}Q_{03}+\dfrac{3f^{2}}{2}\cdot Q_{21}+\dfrac{3f}{2}\cdot Q_{12}\Big]\Big\}\Big|_{h=h_{T}}, $ |
$ G_{2}=-\{(f\cdot P_{20}+P_{11})(u^{0}+u^{\star})+(f\cdot P_{11}+ P_{02})(v^{0}+v^{\star})+f^{3}\cdot P_{30}+P_{03}\\ +3f^{2}\cdot P_{21}+3f\cdot P_{12}+g[(f\cdot Q_{20}+Q_{11})(u^{0}+ u^{\star})+(f\cdot Q_{11}+Q_{02})(v^{0}+v^{\star})\\ +f^{3}\cdot Q_{30}+Q_{03}+3f^{2}\cdot Q_{21}+3f\cdot Q_{12}]\}| _{h=h_{T}}. $ |
方程组
$ \left\{ \begin{array}{l} \tau_{0}\dfrac{\partial \phi}{\partial t}=-l\dfrac{\rho_{1}^{2} \rho_{2}^{2}+\rho_{1}^{2}\rho_{3}^{2}+\rho_{2}^{2}\rho_{3}^{2}} {\rho_{1}\rho_{2}\rho_{3}}\sin\phi, \\ \tau_{0}\dfrac{\partial \rho_{1}}{\partial t}=\mu\rho_{1} +l\rho_{2}\rho_{3}\cos\phi-g_{1}\rho_{1}^{3}-g_{2}(\rho_{2}^{2} +\rho_{3}^{2})\rho_{1}, \\ \tau_{0}\dfrac{\partial \rho_{2}}{\partial t}=\mu\rho_{2}+ l\rho_{1}\rho_{3}\cos\phi-g_{1}\rho_{2}^{3}-g_{2}(\rho_{1}^{2} +\rho_{3}^{2})\rho_{2}, \\ \tau_{0}\dfrac{\partial \rho_{3}}{\partial t}=\mu\rho_{3}+l\rho_{1}\rho_{2}\cos\phi-g_{1}\rho_{3}^{3}-g_{2}(\rho_{1}^{2}+\rho_{2}^{2})\rho_{3}, \\ \end{array} \right. $ | (22) |
其中
通过上述讨论, 我们知道
定理2.1 若系统(3)的参数值在Turing空间中发生变化, 则
(ⅰ)当
(ⅱ)当
(ⅲ)当
本节中, 我们通过Matlab软件对关于系统(3)的空间斑图进行了数值模拟.所有的数值模拟都在离散格子
首先, 设定参数值
图 5展示了当
图 6和图 7分别展示了当
图 8展示了当
基于收获策略对生态系统的动力学行为具有重要的影响, 本文从理论和数值两方面研究了一类具有非线性收获效应的捕食者-食饵模型的空间斑图生成与选择问题, 所得结果表明:
(1) 理论结果表明自扩散不会导致系统(3)产生Turing斑图现象.即系统(3)的Turing斑图的产生是由交叉扩散项引起的.
(2) 理论与数值结果表明非线性收获效应影响了Turing斑图的生成.即在其他参数固定的情况下, 收获能力参数
(3) 理论和数值结果表明非线性收获效应影响了Turing斑图的选择.即随着收获能力参数
[1] | ZADORIN A S, RONDELEZ Y, GINES G, et al. Synthesis and materialization of a reaction-diffusion French flag pattern[J]. Nature Chemistry, 2017, 9(10): 990-996. DOI:10.1038/nchem.2770 |
[2] | 李新政, 白占国, 李燕, 等. 双层非线性耦合反应扩散系统中复杂Turing斑图[J]. 物理学报, 2013, 62(22): 45-51. |
[3] | ZHAO H Y, HUANG X X, ZHANG X B. Turing instability and pattern formation of neural networks with reaction-diffusion terms[J]. Nonlinear Dynamics, 2014, 76(1): 115-124. DOI:10.1007/s11071-013-1114-2 |
[4] | SONG Y L, YANG R, SUN G Q. Pattern dynamics in a Gierer-Meinhardt model with a saturating term[J]. Applied Mathematical Modelling, 2017, 46: 476-491. DOI:10.1016/j.apm.2017.01.081 |
[5] | SUN G Q. Mathematical modeling of population dynamics with Allee effect[J]. Nonlinear Dynamics, 2016, 85(1): 1-12. DOI:10.1007/s11071-016-2671-y |
[6] | WANG X L, WANG W D, ZHANG G H. Vegetation pattern formation of a water-biomass model[J]. Commun Nonlinear Sci Numer Simulat, 2017, 42: 571-584. DOI:10.1016/j.cnsns.2016.06.008 |
[7] | 张道祥, 赵李鲜, 孙光讯, 等. 一类带负交叉扩散项二维系统的空间Turing斑图[J]. 吉林大学学报(理学版), 2017, 55(3): 537-546. |
[8] | SUN G Q, WANG C H, WU Z Y. Pattern dynamics of aGierer-Meinhardt model with spatial effects[J]. Nonlinear Dynamics, 2017, 88(2): 1385-1396. DOI:10.1007/s11071-016-3317-9 |
[9] | GUIN L N. Existence of spatial patterns in a predator-preymodel with self-and cross-diffusion[J]. Applied Mathematics andComputation, 2014, 226: 320-335. DOI:10.1016/j.amc.2013.10.005 |
[10] | 张道祥, 赵李鲜, 胡伟. 一类三种群食物链模型中交错扩散引起的Turing不稳定[J]. 山东大学学报(理学版), 2017, 52(1): 88-97. DOI:10.6040/j.issn.1671-9352.0.2016.286 |
[11] | SAMBATH M, BALACHANRAN K, SUVINTHRA M. Stability and Hopfbifurcation of a diffusive predator-prey model with hyperbolicmortality[J]. Complexity, 2016, 21: 34-43. DOI:10.1002/cplx.21708 |
[12] | XIAO M, CAO J D. Hopf bifurcation and non-hyperbolic equilibrium ina ratio-dependent predator-prey model with linear harvesting rate:Analysis and computation[J]. Mathematical and Computer Modelling, 2009, 50: 360-379. DOI:10.1016/j.mcm.2009.04.018 |
[13] | ZHANG F R, LI Y. Stability and Hopf bifurcation of adelayed-diffusive predator-prey model with hyperbolic mortality andnonlinear prey harvesting[J]. Nonlinear Dynamics, 2017, 88(2): 1397-1412. DOI:10.1007/s11071-016-3318-8 |
[14] | LI Y. Dynamics of a delayed diffusive predator-prey modelwith hyperbolic mortality[J]. Nonlinear Dynamics, 2016, 85(4): 2425-2436. DOI:10.1007/s11071-016-2835-9 |
[15] | 欧阳颀. 非线性科学与斑图动力学导论[M]. 北京: 北京大学出版社, 2010. |
[16] | KUZNETSOV Y A. Elements of applied bifurcation theory. 2nd ed. New York: Springer, 1995. |