Identification of key driving factors for ecological environmental quality in Hainan Tropical Rainforest National Park using causal inference with double machine learning

The ecological environmental quality (EEQ) of national parks is directly related to regional ecological security and biodiversity conservation goals. This study employed the Remote Sensing Ecological Index (RSEI) to assess EEQ spatial distribution patterns and temporal trends in Hainan Tropical Rainforest National Park (HNTRNP) from 2019 to 2024. A Double Machine Learning (DML) model was used to identify key driving factors from land cover types, landscape pattern metrics, topographical characteristics, soil classifications, meteorological parameters, and ecosystem service functions. This approach precisely measured the influence of environmental factors on the distribution patterns and stability of EEQ. The results indicate that HNTRNP's EEQ remained at high levels, with an average RSEI of 0.75, demonstrating a stable and slightly increasing trend from 2019 to 2024 (Sen's slope ≈ 0.003); additionally, 97.58 % of the area maintained stable ecological conditions. Spatial distribution of EEQ was primarily driven by the interaction between soil conservation capacity (SCC) and water conservation capacity (WCC), while interactions between slope (Slp) and length-slope factor (LS), and between landscape heterogeneity (Het) and shape-edge effect (SE) also contributed significantly. RSEI stability was mainly affected by biodiversity (Bio), precipitation (Prec), and soil types (Soil), with LS and SCC positively enhancing ecological quality. The study demonstrates that multi-factor synergistic interactions and landscape structural complexity jointly regulate EEQ. Landscape structural features directly influence ecosystem dynamic transformations, whereas collaborative interactions among multiple elements significantly affect EEQ stability. The DML model provides innovative technical support for identifying key factors influencing national park EEQ and establishes a theoretical basis for science-based conservation and management strategies.