Dynamic Analysis and Optimization of Agro - ecosystems Based on the LV Model

With the advancement of agriculture, deforestation for farmland has disrupted ecological balance and species patterns. This study constructs multiple models for analysis: the Basic Farmland Ecosystem Model (BFW) established based on the Lotka-Volterra competition model, and the Marginal Species Regression model (MSR) that combines improved cellular automata with the LV model. Results demonstrate that the return of marginal species enhances system stability. The impacts of reducing pesticide application and introducing bats are also investigated. An Eco-Sustainability Indicator (ESI) is developed to optimize organic farming models, identifying the optimal parameter combination. Specifically, a 0.5 increase in the fertilizer factor and a 20% increase in secondary-consumer biomass yield the maximum ESI of 0.7098. This research provides theoretical foundations and model support for the sustainable development of agro-ecosystems, integrating ecological dynamics and human intervention strategies through interdisciplinary modeling approaches.