Comparison and spatial heterogeneity of eco-efficiency in two major river basins of China: nonlinear mechanism analysis based on spatial machine learning
- 1usiness School, Henan University of Science and Technology, Luoyang, Henan, China
Abstract
Achieving a synergy between economic expansion and environmental conservation across major watersheds is vital for China’s sustainable trajectory. This study proposes a unified cross-basin framework, using a three-stage super-efficiency Slack-Based Measure (SE-SBM) and Global Malmquist-Luenberger (GML) index to assess the static and dynamic eco-efficiency (EE) of the Yellow River Basin (YRB) and the Yangtze River Economic Belt (YREB). For the analysis of spatiotemporal evolution, kernel density estimation (KDE) and the Dagum Gini coefficient (DGC) are implemented. A Geographically Weighted Random Forest (GWRF) model is introduced to characterize nonlinear associations and spatial heterogeneity among the factors associated with EE. Results reveal that: (1) The adjusted EE of the YRB declined, but it remained superior overall compared to the slightly rising YREB. In terms of space, the YRB shows a stepwise decline, while the YREB presents a concave shape with a lower center. (2) The technical progress (TC) is the primary contributor to GML-based EE growth in the two major river basins. The GWRF model uncovers significant nonlinear threshold effects (e.g., urbanization and environmental regulations) and U-shaped evolutionary characteristics (e.g., industrial structure and financial development). Local feature-importance patterns further indicate distinct spatial structures: a regulation-oriented constraint pattern in the YRB and a market-oriented radiation pattern in the YREB. This multi-method integration provides a novel theoretical perspective and a solid basis for decision-making in formulating cross-basin collaborative policies.
Related
- Unraveling nonlinear ecosystem water use efficiency dynamics and environmental drivers across Iran using remote sensing and XGBoost-SHAP
- A Quasi-Michaelis-Menten Mechanism Explains the Nonlinear Relationship between Transition Metals and PM2.5 Oxidative Potential Measured by the Dithiothreitol (DTT) Assay.
- GECKO-A v1.0: exploring VOC oxidation trajectories through comparison with the master chemical mechanism
- Quantifying nonlinear thresholds of grey-green space morphology on urban carbon metabolism across local climate zones: a machine learning study in Changchun city
- Comparison of machine learning and spatial interpolation methods for hourly rainfall data imputation in the Oliwski Stream catchment
BibTeX
@article{liu2026comparison,
title = {{Comparison and spatial heterogeneity of eco-efficiency in two major river basins of China: nonlinear mechanism analysis based on spatial machine learning}},
author = {Panpan Liu and Dandan Dong and Fei Zhao and Jing Meng and Guoqu Deng},
journal = {Frontiers in Environmental Science},
year = {2026},
month = oct,
doi = {10.3389/fenvs.2026.1937303},
url = {https://doi.org/10.3389/fenvs.2026.1937303},
note = {CC BY},
}
RIS
TY - JOUR
TI - Comparison and spatial heterogeneity of eco-efficiency in two major river basins of China: nonlinear mechanism analysis based on spatial machine learning
AU - Panpan Liu
AU - Dandan Dong
AU - Fei Zhao
AU - Jing Meng
AU - Guoqu Deng
PY - 2026
DA - 2026/10/23
JO - Frontiers in Environmental Science
DO - 10.3389/fenvs.2026.1937303
UR - https://doi.org/10.3389/fenvs.2026.1937303
AB - Achieving a synergy between economic expansion and environmental conservation across major watersheds is vital for China’s sustainable trajectory. This study proposes a unified cross-basin framework, using a three-stage super-efficiency Slack-Based Measure (SE-SBM) and Global Malmquist-Luenberger (GML) index to assess the static and dynamic eco-efficiency (EE) of the Yellow River Basin (YRB) and the Yangtze River Economic Belt (YREB). For the analysis of spatiotemporal evolution, kernel density estimation (KDE) and the Dagum Gini coefficient (DGC) are implemented. A Geographically Weighted Random Forest (GWRF) model is introduced to characterize nonlinear associations and spatial heterogeneity among the factors associated with EE. Results reveal that: (1) The adjusted EE of the YRB declined, but it remained superior overall compared to the slightly rising YREB. In terms of space, the YRB shows a stepwise decline, while the YREB presents a concave shape with a lower center. (2) The technical progress (TC) is the primary contributor to GML-based EE growth in the two major river basins. The GWRF model uncovers significant nonlinear threshold effects (e.g., urbanization and environmental regulations) and U-shaped evolutionary characteristics (e.g., industrial structure and financial development). Local feature-importance patterns further indicate distinct spatial structures: a regulation-oriented constraint pattern in the YRB and a market-oriented radiation pattern in the YREB. This multi-method integration provides a novel theoretical perspective and a solid basis for decision-making in formulating cross-basin collaborative policies.
SN - 2296-665X
ER -