• 关键词：模糊c-均值算法  连续分类  空间预测  土壤污染  重金属

• 基金项目：国家自然科学基金项目(No.40571065);中国科学院知识创新方向性项目(No.KZCX3-SW-427);国家自然科学基金项目(No.40235054)

• 檀满枝
• 1. 中国科学院南京土壤研究所土壤与农业可持续发展国家重点实验室, 南京 210008; 2. 中国科学院研究生院, 北京 100039

• 陈杰
• 中国科学院南京土壤研究所土壤与农业可持续发展国家重点实验室, 南京 210008

• 郑海龙
• 香港中文大学 地理与资源管理系

• 张学雷
• 中国科学院南京土壤研究所土壤与农业可持续发展国家重点实验室, 南京 210008

• 摘要：基于93个土壤表层样品重金属Cu、Zn、Cr、Cd、Pb和Hg全量浓度分析数据,应用模糊分类方法结合地统计学手段,对南京市边缘带梅山钢铁厂附近总面积约14km²样区内的土壤重金属污染状况进行了空间预测.结果显示,样区北部及东北角区域的土壤遭受重金属元素Cu、Zn、Cr、Cd的轻微污染,样区西侧沿长江呈带状分布的土壤由于其重金属环境容量较低,存在潜在污染风险,样区其它区域的土壤未发现由上述重金属元素导致的污染现象.样区东北角土壤Pb、Hg污染严重,并与Cu、Zn、Cr、Cd污染在空间上重叠,是土壤重金属污染控制与土壤修复的重点区域.样区的东部和东北部区域土壤受Pb、Hg中度污染,在梅山集团的北部土壤轻微污染.研究表明,与直接对重金属浓度分析数据实施插值分析的常规克里格方法相比,模糊c-均值聚类结合地统计学方法获得的空间预测结果蕴含信息更加丰富,而模糊集理论及其相关算法在土壤学诸多领域中也具有极大的应用潜力.

• Abstract：Based on the total concentrations of Cu, Zn, Cr, Cd, Pb and Hg determined in 93 topsoil (0～20cm) samples collected from the study area about 14 km² around the Meishan Steel Industry located in peri-urban Nanjing, prediction of soil pollution induced by the measured heavy metals was achieved by the use of fuzzy c-mean algorithm (FCM). The result revealed that, the polluted soils induced by Cu, Zn, Cr and Cd concentrated in the northern part and northeastern part of the study area, while the soils free of pollution were mainly distributed in the eastern and northern part. And the soils with potential risk of pollution derived from the mentioned heavy metals were located in the long and narrow Yangtze River in the western part of the study region. The heavily contaminated soils by Pb and Hg were concentrated in the northeastern corner of the study area, where soil pollution derived by Cu, Zn, Cr and Cd was also checked out. Moderately polluted soils induced by Pb and Hg were distributed in the eastern and northeastern part, while slight soil pollution was checked out in northern part of the study area. It was demonstrated that FCM approach combined with geostatistical techniques could produce a desirable prediction results which has more information compared with conventional direct kriging methods. And prediction of soil pollution which was achieved by FCM methods in this paper indicated that fuzzy set theory had a great potential in the fields of soil science.

• Key words：fuzzy c-means  continuous classification  spatial prediction  soil pollution  heavy metal

• 
  摘要点击次数： 2602 全文下载次数： 3988