Time‐Series Surface Deformation and Driving Mechanism Analysis in the Beijing Plain, China
Journal of Sensors · 2026-01 · 2026 (1)
DOI: 10.1155/js/9184485 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Sensors, online erschienen am 10.11.2026. Verfasst von Feng Li und 3 weiteren Autor:innen. Die Arbeit nennt 53 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Science Foundation of Hebei Province.
Abstract (Originalsprache)
Land subsidence poses a significant challenge to sustainable urban development in rapidly urbanizing regions. This study aims to investigate the spatiotemporal characteristics of surface deformation and quantify the nonlinear relationships between deformation and its driving factors in the Beijing Plain, China. Sentinel‐1 SAR images acquired from 2023 to 2024 were processed using the small baseline subset interferometric synthetic aperture radar (SBAS‐InSAR) technique to derive time‐series surface deformation. An interpretable machine learning framework integrating Optuna‐optimized extreme gradient boosting (XGBoost) and SHAP (SHapley Additive exPlanations) was developed to identify dominant controlling factors and reveal their nonlinear response mechanisms. Multiple driving factors, including groundwater‐level variation, nighttime light (NTL) intensity, road density (RoadDen), geological elevation, soil compressibility, and lithological characteristics, were incorporated into the analysis at regional, urban‐core, and suburban subsidence scales. The results reveal pronounced spatial heterogeneity in surface deformation across the Beijing Plain, characterized by widespread stability or uplift in central urban areas and persistent localized subsidence in peripheral plains. Groundwater‐level variation was identified as the dominant controlling factor across different spatial scales, followed by anthropogenic activities and hydrogeological conditions. SHAP analysis revealed significant nonlinear threshold effects between driving factors and surface deformation. Groundwater recovery promoted surface rebound, whereas groundwater declines intensified subsidence. Urbanization‐related factors, including NTL intensity and RoadDen, exhibited stage‐dependent effects on deformation responses. This study demonstrates that the integration of SBAS‐InSAR and interpretable machine learning provides an effective framework for understanding complex land deformation mechanisms. The findings improve the understanding of groundwater‐driven subsidence evolution in the Beijing Plain and provide scientific support for refined groundwater management, subsidence risk assessment, and sustainable urban planning.
Förderung
- Natural Science Foundation of Hebei Province
Zitieren
APA
Li, F., Yan, F., Qin, Z., & Zhao, Y. (2026). Time‐Series Surface Deformation and Driving Mechanism Analysis in the Beijing Plain, China. Journal of Sensors, 2026(1). https://doi.org/10.1155/js/9184485
Vancouver
Li F, Yan F, Qin Z, Zhao Y. Time‐Series Surface Deformation and Driving Mechanism Analysis in the Beijing Plain, China. Journal of Sensors. 2026;2026(1). doi:10.1155/js/9184485
BibTeX
@article{Li2026,
author = {Li, Feng and Yan, Fangfang and Qin, Zihan and Zhao, Yunbo},
title = {Time‐Series Surface Deformation and Driving Mechanism Analysis in the Beijing Plain, China},
journal = {Journal of Sensors},
year = {2026},
volume = {2026},
number = {1},
doi = {10.1155/js/9184485}
}
EUVIMED