Revisiting the relationship between geopolitical risk and ecological footprint: A comprehensive analysis based on dual machine learning

Geopolitical conflicts and other risk events are subtly reshaping the global political and economic landscape, gradually disrupting the balance between economic development and ecological sustainability. Understanding the pathways through which geopolitical risks affect the ecological footprint is crucial for achieving ecological sustainability goals. This study employed dual machine learning models for high-precision analysis to deeply explore the intrinsic patterns of how geopolitical risks impact the ecological footprint. Income heterogeneity was also considered. On the one hand, this research constructed a multi-window kernel density estimation (KDE) model to analyze the spatial and temporal characteristics of the ecological footprint. By incorporating dual machine learning models (DML), it innovatively discovered a positive effect of geopolitical risks on the ecological footprint. On the other hand, this study investigated the moderating mechanisms of energy transition through two paths, finding that under the dual-pathway regulation of energy transition, the positive effect of geopolitical risks on the ecological footprint gradually weakens and may even turn negative. Furthermore, compared to high-income countries, geopolitical risks have a stronger impact on the ecological footprint in lower-income countries. The policy recommendations proposed in this study offer a new perspective for decision-makers and are of significant importance for ecological sustainability.