Identifying the spatio-temporal dynamics of regional ecological risk based on Google Earth Engine: A case study from Loess Plateau, China

• DOI: 10.1016/j.scitotenv.2023.162346
• Publication Date: 2023-02-22T01:55:02Z
• Authors: Wencang Shen; Jianjun Zhang; Ke Wang; Zhengfeng Zhang
• Keywords: 13. Climate action; 11. Sustainability; 15. Life on land; 6. Clean water

Located in the middle reaches of the Yellow River Basin, the Loess Plateau is one of the typical eco-fragile regions in China, which has complex and diverse ecological problems and is in urgent need of comprehensive ecological governance and restoration. Ecological risk assessment as a method to quantify complex ecological risks can provide decision-makers with quantification and visualization of risk scenarios, which can serve the ecological protection and restoration of the Loess Plateau. Pre-processed and acquired through GEE, the multi-source and long-term remote sensing data are introduced to facilitate the efficiency of risk monitoring. There are few studies about GEE-based spatio-temporal ecological risk monitoring in the Loess Plateau. Therefore, a regional ecological risk assessment system from the perspective of risk sources, risk receptors, and exposure responses has been established to visualize and quantify the spatial heterogeneity of regional multi-perspective ecological problems at a 1 km-grid scale from 2001 to 2020. Meanwhile, multi-aspects spatial statistics and trend analysis are taken to identify the spatial and temporal characteristics. The results demonstrated that the comprehensive ecological risk of the Loess Plateau presented a decreasing trend from 2001 to 2020, which is mainly due to the improvement of vegetation conditions. Spatially, the low-risk pixels are primarily in the east and west, while the high-risk pixels mainly locate in the north central and northwest. From 2001 to 2020, the high-risk pixels in the central region decrease, and the high-risk pixels are concentrated in the northwest. In general, the ecological risks is significantly spatially heterogeneous, varying according to latitude and longitude, land cover, topography. Identifying grid-based dominant risks and trends provides decision-makers with accurate risk control and helps propose appropriate countermeasures.