摘要
定量土壤重金属含量空间分布预测结果的不确定性对于控制土壤空间数据的应用风险具有重要意义。采用序贯指示模拟方法对张家港土壤表层（0～15cm）Cu含量的空间不确定性进行了定量评价。结果表明，张家港东部的集约农业区及中北部的冶金和电子类企业聚集区土壤表层Cu含量较高，其空间预测的波动性也较大，而该市南部土壤Cu含量较低，空间表达的波动性也较小；序贯指示模拟的模拟实现图像没有平滑效应，当给定一个Cu含量阈值时，序贯指示模拟可以定量任-空间位置Cu含量大于该阈值的概率，同时序贯指示模拟也可以评价描绘Cu含量大于该阈值的区域的空间不确定性。
The mapping of heavy metal distribution is usually associated with certain uncertainty that needs to be considered in decision-making for heavy metal pollution management. The uncertainty in prediction of heavy metal contents at unsampled locations is particularly important because such uncertainty can be propagated into subsequent modelling or decision-making process and has fundamental impacts on ultimate results. Based on the Cu content data of a total of 547 topsoil （0 - 15 cm） samples in Zhangjiagang City, the sequential indicator simulation （SIS） were applied to perform a conditional stochastic simulation in this study to quantitatively assess the uncertainty of mapping total Cu content. Results show that prediction of the spatial pattern of soil Cu fluctuated the most in the eastern and central northern parts of the study area, where intensive agriculture and metallurgic and electronic enterprises dominate and the total Cu contents were high, and the least in the southern part, where the total Cu contents were low. The pattern simulated with SIS did not show any smoothing effect. With a given threshold of Cu content, SIS can quantitatively assess both local uncertainty and spatial uncertainty of Cu content that is greater the threshold.
出处
《土壤学报》
CAS
CSCD
北大核心
2007年第6期974-981,共8页
Acta Pedologica Sinica
基金
江苏省自然科学基金项目(BK2006581)
中国科学院知识创新工程领域前沿项目(ISSASIP0604)
国家重点基础研究发展规划项目(2002CB410810)资助
关键词
土壤制图
空间不确定性
随机模拟
Soil mapping
Spatial uncertainty
Stochastic simulation
