文献类型：期刊文献

英文题名：A hybrid spatial framework for risk mapping and driving factor diagnosis of soil combined contamination in Guangzhou

作者：Chen, Lian[1];Wang, Zhenjiang[1];Sun, Lingyun[1,2];Wu, Zhipeng[2];Lin, Sen[1];Wang, Dan[1];Wu, Jianan[1];Li, Zhiyi[1];Li, Gaocong[3]

机构：[1]Guangdong Acad Agr Sci, Sericultural & Agri Food Res Inst, Guangzhou 510610, Guangdong, Peoples R China;[2]Hainan Univ, Sanya Inst Breeding & Multiplicat, Sch Breeding & Multiplicat, Sanya 572025, Peoples R China;[3]Guangdong Ocean Univ, Coll Elect & Informat Engn, 1 Haida Rd, Zhanjiang 524088, Guangdong, Peoples R China

年份：2026

卷号：283

外文期刊名：JOURNAL OF GEOCHEMICAL EXPLORATION

收录：SCI-EXPANDED(收录号：WOS:001676626500001)、、EI(收录号：20260419946409)、Scopus(收录号：2-s2.0-105027925349)、WOS

基金：This research was supported Youth S&T Talent Support Programme of Guangdong Provincial Association for Science and Technology (SKXRC2025474); the National Natural Science Foundation of China (42007379); Special fund for the introduction of scientific and technological talents of Guangdong Academy of Agricultural Sciences (R2021YJ-YB3007); Youth Mentoring Program of Guangdong Academy of Agricultural Sciences (R2020QD-044).

语种：英文

外文关键词：Combined pollution; Risk assessment; Multi-criteria decision analysis; Random forest; Spatial interaction; Urban soil

外文摘要：The spatial characterization of risk patterns and their driving factors in urban soils co-contaminated with potentially toxic elements (PTEs) and polycyclic aromatic hydrocarbons (PAHs) remains a critical research challenge. To address this, we developed a novel integrated framework, MGRB (MCDA-GIS-RF-BLMI), that synergistically combines Multi-Criteria Decision Analysis (MCDA), Geographic Information Systems (GIS), Random Forest (RF), and Bivariate Local Moran's I (BLMI). This framework enables comprehensive assessment of combined contamination risks, generation of detailed risk zoning maps, identification of key influencing factors, and analysis of spatial interactions between risk levels and environmental drivers. Guangzhou, a major political, economic, and cultural center in southern China, was applied to the MGRB framework to demonstrate its feasibility and innovation. Results derived from MCDA and GIS revealed a distinct spatial risk gradient, with risk zones distributed as follows: low-risk (55.0%, 4092.4 km(2)) > moderate-risk (23.1%, 1713.9 km(2)) > considerable-risk (12.7%, 942.1 km(2)) > no-risk (7.7%, 572.9 km(2)) > high-risk (1.5%, 113.0 km(2)). Notably, approximately 37% of Guangzhou's territory exhibited moderate-to-high risk levels, with significant spatial clustering in western and central districts as well as scattered high-risk patches in northern, eastern, and southern areas. RF analysis identified road density, population density, and GDP as the dominant drivers, displaying most frequently and strongest associations with PTE-PAH risk. BLMI results further indicated that high-high clusters, where both contamination risk levels and driving factors are elevated, were predominantly concentrated in western and southern regions, which should be prioritized for targeted pollution control and mitigation strategies. This study develops an innovative methodology for the comprehensive spatial assessment of combined PTE-PAH risks and their driving factors in Guangzhou. This framework serves both to guide local risk management and to provide a transferable model for other rapidly urbanizing regions with analogous challenges.