Socioeconomic determinants of suburban cropland vitality: an interpretable machine learning analysis of four national central cities in China

The social functions of suburban cropland are increasingly prominent, yet remain severely undervalued in land management due to a long-standing lack of effective quantification methods. Traditional research primarily relies on static socioeconomic statistics, limiting the ability to capture real-time human-cropland interactions and fine-scale spatial heterogeneity. To address this, the concept of "cropland vitality" is introduced to characterise these social functions, and an RF-RFE-SHAP explainable machine learning framework is developed by utilising multi-source data. Focusing on four national central cities in China (Beijing, Xi’an, Guangzhou and Shanghai), we systematically evaluate spatial gradient patterns and non-linear predictive associations between socioeconomic features and population-based cropland vitality across the urban–rural interface. Results demonstrate that: (1) Spatially, high-vitality cropland proportions in the inner, middle, and outer suburban zones average 74.47%, 47.60%, and 29.35%, respectively, exhibiting a dual spatial configuration of distance-decay gradients and patchy aggregation; (2) The optimised model, identifying eight core socioeconomic features, achieved robust predictive performance (average test-set R 2 of 0.74). Specifically, daily life services (DLS), transportation (TS), and enterprises (EN) generally showed positive SHAP contributions, whereas governmental organization & social group (GOSG) exhibited inverted U-shaped patterns in Xi’an and Beijing, a broadly similar pattern in Guangzhou, and a weaker response in Shanghai; (3) City-specific predictive patterns reveal spatially differentiated associations between socioeconomic features and cropland vitality in contrasting local development contexts. This study advances the quantitative assessment of cropland multifunctionality and informs differentiated spatial governance. Its findings highlight the importance of context-specific management for promoting urban–rural integration and sustainable regional development.

Authors

Institutions

Publication Details

Journal
Frontiers in Sustainable Food Systems
Published
2026-11-18
DOI
https://doi.org/10.3389/fsufs.2026.1972826
Primary Topic
Land Use and Ecosystem Services
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Socioeconomic determinants of suburban cropland vitality: an interpretable machine learning analysis of four national central cities in China

Chaoxiang Wen, Jiayu Cai, Hongkai Geng, Chunqiang Li et al.
6 links
Frontiers in Sustainable Food Systems
Land Use and Ecosystem Services
article

Socioeconomic determinants of suburban cropland vitality: an interpretable machine learning analysis of four national central cities in China

Chaoxiang Wen, Jiayu Cai, Hongkai Geng, Chunqiang Li, Shiying Li, Yicheng Zheng, Xuewei Zhang, Yuan Chen, Tao Lin, Xiangzhong Guo
article en

Abstract

The social functions of suburban cropland are increasingly prominent, yet remain severely undervalued in land management due to a long-standing lack of effective quantification methods. Traditional research primarily relies on static socioeconomic statistics, limiting the ability to capture real-time human-cropland interactions and fine-scale spatial heterogeneity. To address this, the concept of "cropland vitality" is introduced to characterise these social functions, and an RF-RFE-SHAP explainable machine learning framework is developed by utilising multi-source data. Focusing on four national central cities in China (Beijing, Xi’an, Guangzhou and Shanghai), we systematically evaluate spatial gradient patterns and non-linear predictive associations between socioeconomic features and population-based cropland vitality across the urban–rural interface. Results demonstrate that: (1) Spatially, high-vitality cropland proportions in the inner, middle, and outer suburban zones average 74.47%, 47.60%, and 29.35%, respectively, exhibiting a dual spatial configuration of distance-decay gradients and patchy aggregation; (2) The optimised model, identifying eight core socioeconomic features, achieved robust predictive performance (average test-set R 2 of 0.74). Specifically, daily life services (DLS), transportation (TS), and enterprises (EN) generally showed positive SHAP contributions, whereas governmental organization & social group (GOSG) exhibited inverted U-shaped patterns in Xi’an and Beijing, a broadly similar pattern in Guangzhou, and a weaker response in Shanghai; (3) City-specific predictive patterns reveal spatially differentiated associations between socioeconomic features and cropland vitality in contrasting local development contexts. This study advances the quantitative assessment of cropland multifunctionality and informs differentiated spatial governance. Its findings highlight the importance of context-specific management for promoting urban–rural integration and sustainable regional development.

Frontiers in Sustainable Food SystemsVol. 10
Xiamen University (CN), Chinese Academy of Sciences (CN), Fujian Business University (CN), Institute of Urban Environment (CN), Fuzhou University (CN), Nanjing University (CN)
Sustainable cities and communities, Zero hunger
Land Use and Ecosystem Services
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon