Spatiotemporal dynamics and suitability of greenhouse agriculture in China: A GEE-based analysis for sustainable land use planning in Henan Province

Greenhouse cultivation, integral to modern agricultural facilities, is pivotal for sustainable agricultural development and land resource management. However, long-term monitoring of their spatiotemporal dynamics and structural stability remains limited in seasonal climate regions. This study employed the Google Earth Engine (GEE) cloud computing platform and the Random Forest algorithm to analyze spatiotemporal dynamics and assess land suitability of greenhouses in Henan Province from 2017 to 2023. Key findings include: (1) The classification model achieved an overall accuracy exceeding 90 % and Kappa coefficients above 0.85. (2) Greenhouse area increased from 235.08 km² in 2017 to a peak of 496.93 km² in 2020, then declined to 389.21 km² in 2021. Subsequently, the area stabilized with minor fluctuations, reaching 403.50 km² in 2023. (3) Greenhouses were concentrated in Zhoukou City, Shangqiu City, and Zhongmu County of Zhengzhou City. Over 85 % were located in areas below 300 m elevation and on slopes less than 8°, primarily expanding along county roads and near residential areas. (4) Land suitability assessment using Principal Component Analysis indicated that 69.42 % of the provincial area is suitable for greenhouse development. This study provides a framework integrating spatiotemporal dynamics and structural stability for large-scale greenhouse monitoring and suitability assessment, offering information for sustainable agricultural development and land resource utilization, particularly in regions with seasonal climates.