AI-ASSISTED PREDICTION OF CLIMATE CHANGE IMPACTS ON PLANT GROWTH AND BIODIVERSITY

Abstract

Climate change poses significant challenges to ecosystems by altering plant growth patterns and biodiversity. Accurate prediction of these impacts is essential for ecological conservation and sustainable land management. This study employs an AI-assisted predictive modeling approach to assess the effects of climatic variables including temperature anomalies, precipitation changes, and atmospheric CO₂ concentrations on plant biomass and species diversity across diverse ecological regions. A simulated dataset of 300 samples representing tropical forests, temperate forests, grasslands, wetlands, and agricultural lands was analyzed using Random Forest Regression, complemented by Support Vector Regression and Gradient Boosting for comparison. Model validation demonstrated strong predictive performance (R² ≈ 0.72), while partial dependence analyses revealed nonlinear interactions and threshold effects between climate variables and plant growth. The study further identifies high-risk regions for biodiversity loss, providing actionable insights for conservation planning. Findings underscore the utility of AI-based approaches in forecasting ecosystem responses under climate change scenarios.