Turing pattern formation in a three species model with generalist predator and cross-diffusion (Q387147)

The paper deals with the Lotka-Volterra model \[ \begin{cases} \displaystyle\frac{du_1}{dt}=u_1(r_1-b_1u_1-\gamma_1u_3),\\ \displaystyle\frac{du_2}{dt}=u_2(r_2-b_2u_2-\gamma_2u_3),\\ \displaystyle\frac{du_3}{dt}=u_3(\beta_1u_1+\beta_2u_2-\delta),\end{cases}\leqno(S) \] where the variable \(u_i\) is the population density of prey species \(i,\,\,i=1,2\), and \(u_3\) is the population density of predator species. The authors also study a generalist predator system with cross-diffusion that represents the tendency of predators to avoid the group defense by a large number of preys (system (1.2) from the paper). They prove that the unique positive equilibrium is globally asymptotically stable for system \((S)\), and the positive equilibrium remains globally stable in the presence of self-diffusion without cross-diffusion. The authors show that the uniform stationary solution is unstable when the cross-difussion appears, and then they investigate the existence and nonexistence of non-constant positive steady states solutions of system (1.2).