| 基于APCS-MLR模型和地理探测器的煤矸山周边土壤污染溯源解析和影响因素分析 |
| 摘要点击 1081 全文点击 197 投稿时间：2024-01-25 修订日期：2024-03-26 |
| 查看HTML全文
查看全文 查看/发表评论 下载PDF阅读器 |
| 中文关键词 土壤 重金属 绝对因子得分-多元线性回归（APCS-MLR）模型 地理探测器 源解析 影响因子 |
| 英文关键词 soil heavy metals absolute principal component scores-multiple linear regression（APCS-MLR）model GeoDetector source apportionment impact factors |
| 作者 | 单位 | E-mail | | 马杰 | 重庆市生态环境监测中心, 重庆 401147
中国环境监测总站, 北京 100012
有机污染物环境化学行为与生态毒理重庆市重点试验室, 重庆 401147 | pony312@qq.com | | 秦启荧 | 重庆市生态环境监测中心, 重庆 401147 | | | 王胜蓝 | 重庆市生态环境监测中心, 重庆 401147
中国环境监测总站, 北京 100012 | | | 李名升 | 中国环境监测总站, 北京 100012
国家环境保护环境监测质量控制重点实验室, 北京 100012 | | | 封雪 | 中国环境监测总站, 北京 100012
国家环境保护环境监测质量控制重点实验室, 北京 100012 | fengxue@cnemc.cn |
|
| 中文摘要 |
| 以重庆市煤矸山周边土壤为研究对象,运用绝对因子得分-多元线性回归（APCS-MLR）模型对土壤重金属开展源解析,考虑坡度、高程、土壤点位与煤矸山、居民点和主干道距离等5个环境因子,运用地理探测器开展土壤影响因素分析. 结果表明,研究区土壤ω（Cd）、 ω（Hg）、 ω（Pb）、 ω（Cr）、 ω（Cu）、 ω（Zn）和ω（Ni）均值分别为1.33、0.29、32.9、142、68.8、118和54.6 mg·kg-1,Cd是首要污染物. APCS-MLR模型源解析结果表明,研究区土壤受矿业源影响,贡献率为37.1%,污染因子以Cd、Hg和Pb为主；受农业和交通源影响,贡献率为36.2%,污染因子以Cu、Zn和Ni为主；受自然源影响,贡献率为26.7%,污染因子以Cr为主. 地理探测器分析结果表明,Cd、Hg和Pb在环境因子“与煤矸山距离”解释力最强,Cr、Cu、Zn和Ni在环境因子“与居民点距离”解释力最强,因子两两交互后,影响力均有提升,说明土壤重金属空间含量分布特征受多因子复合影响. APCS-MLR模型和地理探测器联用,可以在源解析和影响因素分析上相互验证,使解析结果更加全面、准确和可靠. |
| 英文摘要 |
| To analyze the source apportionment and influence factors of heavy metals in soils surrounding a coal gangue heap in Chongqing, the absolute principal component scores-multiple linear regression （APCS-MLR） model and GeoDetector were used. The results showed that Cd was the primary pollutant and the average values of Cd, Hg, Pb, Cr, Cu, Zn, and Ni were 1.33, 0.29, 32.9, 142, 68.8, 118, and 54.6 mg·kg-1, respectively. Using the APCS-MLR model analysis, mining sources, which were mainly affected by long-term accumulation of the coal gangue heap, had a contribution rate of 37.1% and the main heavy metal pollutants were Cd, Hg, and Pb. Agriculture and transportation sources, mainly affected by pesticide and fertilizer application and vehicle emissions, had a contribution rate of 36.2%, with the main heavy metal pollutants being Cu, Zn, and Ni. Natural sources, which were mainly affected by geotechnical weathering processes of their parent materials, had a contribution rate of 26.7% and the main heavy metal pollutant was Cr. Using GeoDetector analysis, the “distance from coal gangue heap” had the strongest explanatory power for the contents of Cd, Hg, and Pb, whereas the “distance from rural settlements” had the strongest explanatory power for the contents of Cr, Cu, Zn, and Ni. However, the interaction of each influencing factor was enhanced, which indicated that the spatial distribution characteristics of heavy metals were influenced by multiple factors. The combined application of the APCS-MLR model and GeoDetector can make the results of source apportionment and influence factors more comprehensive, accurate, and reliable. |
|
|
|
