Land subsidence and its relation with groundwater aquifers in Beijing Plain of China

Beibei Chen; Huili Gong; Yun Chen; Xiaojuan Li; Chaofan Zhou; Kunchao Lei; Lin Zhu; Li Duan; Xiaoxiao Zhao

doi:10.1016/j.scitotenv.2020.139111

Land subsidence and its relation with groundwater aquifers in Beijing Plain of China

Sci Total Environ. 2020 Sep 15:735:139111. doi: 10.1016/j.scitotenv.2020.139111. Epub 2020 May 4.

Authors

Beibei Chen¹, Huili Gong², Yun Chen³, Xiaojuan Li¹, Chaofan Zhou¹, Kunchao Lei⁴, Lin Zhu¹, Li Duan¹, Xiaoxiao Zhao¹

Affiliations

¹ Key Laboratory of the Ministry of Education Land Subsidence Mechanism and Prevention, Capital Normal University, Beijing 100048, China; College of Geospatial Information Science and Technology, Capital Normal University, Beijing 100048, China; College of Resources Environment and Tourism, Capital Normal University, Beijing 100048, China.
² Key Laboratory of the Ministry of Education Land Subsidence Mechanism and Prevention, Capital Normal University, Beijing 100048, China; College of Geospatial Information Science and Technology, Capital Normal University, Beijing 100048, China; College of Resources Environment and Tourism, Capital Normal University, Beijing 100048, China. Electronic address: applechenbeibei@126.com.
³ Commonwealth Scientific and Industrial Research Organisation Land and Water, Canberra, ACT, Australia.
⁴ Beijing Institute of Hydrogeology and Engineering Geology, Beijing 100195, China.

PMID: 32464408
DOI: 10.1016/j.scitotenv.2020.139111

Abstract

Beijing is a major metropolis with significant land subsidence because of long-term overexploitation of groundwater. While the South-to-North Water Diversion Project (SWDP) has provided new water sources Beijing; it has changed the pattern of land subsidence evolution in Beijing since 2015. Here we address how land subsidence evolution before and after SWDP, and we quantify also the impact of groundwater level changes in different aquifers on land subsidence at spatial scale. Subsidence evolution before and after SWDP were compared by adopting Persistent Scatterer Inteferomotry (PSI) with Radarsat-2 and Sentinel-1 data. Spatial correlation between Interferometric Synthetic Aperture Radar (InSAR) derived subsidence and groundwater levels in four aquifers was investigated using the Random Forest (RF) machine learning algorithm and Geographical Detectors (GD) technique. Extensometer deformation data and corresponding variation in groundwater level observations at three monitoring stations were used for validations. The study reveals that: firstly, both InSAR-derived subsidence area and maximum annual deformation rate decreased from 79.2% and 141 mm/yr before SWDP, to 60.1% and 135 mm/yr after SWDP. A reduction of time series deformation at four subsidence centers started about two years after the commence of SWDP in 2015. Secondly, the variation of groundwater level in the second confined aquifer has the strongest spatial correlation with subsidence in all the aquifers, but its impact on this aquifer has decreased after SWDP. These findings have an important scientific significance for the rational allocation of water resources and management strategy for mitigating hazards associated with subsidence against the background of SWDP.

Keywords: Geographical Detectors (GD) method; Land subsidence; Random Forest (RF) algorithm; Spatial correlation.