Existence of non-coexisting steady-state solutions for a class of reaction-diffusion competition systems

Technical Summary: Existence of Non-Coexisting Steady-State Solutions for a Class of Reaction-Diffusion Competition Systems

Problem Statement
This research investigates the existence and properties of non-coexisting steady-state solutions for a class of reaction-diffusion competition systems based on the Shigesada-Kawasaki-Teramoto (SKT) model. The study focuses on how cross-diffusion terms influence the system's behavior under two distinct boundary conditions: Neumann (zero-flux) and mixed boundary conditions. Specifically, the authors analyze the system where two competing species, denoted by population densities $u$ and $v$, interact through self-diffusion and cross-diffusion. The primary objective is to determine exact parameter ranges under which the system admits only constant steady-state solutions where one species drives the other to extinction (non-coexistence), rather than allowing for coexistence or non-constant spatial patterns.

Methodology
The authors employ a rigorous analytical approach grounded in the theory of elliptic partial differential equations. The methodology proceeds through the following steps:

1. Transformation and Positivity: The system is first transformed from its divergence form into a non-divergence form elliptic system to facilitate analysis. Using the strong maximum principle and Hopf's boundary lemma, the authors establish the strict positivity of non-negative classical solutions (i.e., if a species is present, its density is strictly positive throughout the domain).
2. A Priori Estimates: By applying a specific lemma regarding the extrema of solutions to elliptic equations with Neumann boundary conditions, the authors derive upper and lower bound estimates for the population densities. These estimates relate the maximum and minimum values of one species to the extrema of the other.
3. Auxiliary Function Construction: A critical component of the analysis involves constructing an auxiliary function, $\phi(s)$, which encapsulates the relationship between the parameters and the solution bounds. The authors perform a detailed analysis of the monotonicity and bounds of this function under specific parameter constraints.
4. Contradiction Arguments: The core of the proofs relies on contradiction. By assuming the existence of a non-trivial coexisting solution (where both $u \not\equiv 0$ and $v \not\equiv 0$), the authors utilize the derived bounds and the properties of the auxiliary function to demonstrate that the assumed conditions lead to mathematical contradictions (e.g., a variable being simultaneously greater than and less than a specific threshold).

Key Contributions and Results

• Neumann Boundary Conditions:

  • With Cross-Diffusion ($a_{12} \neq 0$): The study establishes that if the intrinsic growth rate of species $u$ is greater than 1 ($a_1 > 1$) and that of species $v$ is less than 1 ($a_2 < 1$), and if the cross-diffusion and self-diffusion coefficients satisfy $2a_{22} - a_2 a_{12} \geq 0$, the system admits only non-coexisting constant solutions. Specifically, if $u$ is present, $v$ must be zero ($u \equiv 1, v \equiv 0$), and vice versa. The analysis reveals that the influence of cross-diffusion on non-coexistence depends heavily on the intensity of interspecific competition.
  • Without Cross-Diffusion ($a_{12} = 0$): In the absence of cross-diffusion, the system reduces to a self-diffusion model. The authors prove that under the same growth rate conditions ($a_1 > 1, a_2 < 1$), the system also admits only non-coexisting constant solutions. Notably, the results suggest that when only self-diffusion is present, the specific competitive strength parameters do not alter the existence of these non-coexisting solutions in the same way cross-diffusion does.

• Mixed Boundary Conditions:

  • The paper extends these findings to a mixed boundary condition scenario where $u = 0$ and $\frac{\partial v}{\partial \nu} = 0$ on the boundary. Under the condition $a_1 > 1$ and $a_2 < 1$, the authors prove that the system possesses a unique non-coexisting constant solution $(u, v) = (0, 1)$. This indicates that under mixed boundaries, the species with the weaker competitive ability ($u$) is driven to extinction, while the stronger competitor ($v$) persists at its carrying capacity.

• Numerical Validation:

  • The theoretical results are supported by numerical simulations. The simulations illustrate that while the steady-state solution $(1, 0)$ is mathematically proven to exist under specific conditions, it may be unstable in dynamic simulations, leading to the extinction of species $u$ and the persistence of species $v$ in the observed time evolution.

Significance and Claims
The paper claims to extend the existing literature on the SKT model by providing exact parameter ranges that guarantee the existence of non-coexisting steady-state solutions under both Neumann and mixed boundary conditions. A significant finding is the clarification of the role of cross-diffusion: the authors demonstrate that the influence of cross-diffusion on the existence of non-coexisting solutions is contingent upon the intensity of interspecific competition.

Furthermore, the work distinguishes between scenarios with and without cross-diffusion, showing that the conditions for non-coexistence are more restrictive or dependent on specific parameter inequalities when cross-diffusion is present. The study contributes to the broader understanding of how spatial segregation mechanisms (driven by diffusion) and interspecific competition interact to determine the long-term survival or extinction of competing species in bounded domains. The results are presented as theoretical proofs of existence for constant solutions, offering a mathematical basis for understanding competitive exclusion in ecological models.