Research on grassland ecological evolution based on cellular automata

Grassland ecology is a vital subfield within ecology, focusing on the structure, function, and dynamic changes of grassland ecosystems. And the interactions between these ecosystems and the environment required with valid computer simulations. This study models the grassland ecosystem as a simplified system consisting of representative species-wolves, sheep, and grass, considering factors such as predator-prey interactions, population growth, and resource competition. Using cellular automata (CA) method, the model successfully simulates species population dynamics and state changes. The CA method not only captures species interactions but also reflects local changes within the ecosystem, offering valuable insights into the internal mechanisms of grassland ecosystems. However, despite offering a novel perspective for simulation, the CA method has limitations in maintaining ecological balance, as the simulation results show poor stability in species population numbers. To address this limitation, and considering the feasibility and simplicity of human intervention in grassland ecosystems, this study integrates a genetic algorithm (GA) and optimizes the predator-prey parameters in the Lotka-Volterra model. By adjusting the mortality rates of wolves and sheep, the optimal parameter combination was found to be a sheep death rate of 0.216961 and a wolf death rate of 0.135227. Incorporating the genetic algorithm improves the species balance in the cellular automata model, demonstrating better performance in controlling species numbers, stabilizing fluctuations, and reaching equilibrium. Compared to traditional CA models, the CA-GA model effectively reduces species population fluctuations, ensures relative stability between wolves and sheep populations, and reaches ecological equilibrium faster in fewer iterations. This provides a quantitative basis for human intervention in species mortality rates. These results not only confirm the effectiveness of the CA-GA model in addressing species interaction issues in complex ecosystems but also highlight its potential for ecological research and ecosystem management.