摘要
该文以农业土壤重金属Cd、Cu为研究对象,将研究区域722个样点,随机均分为2个均等数据集。分别利用IDW法、普通Kriging法对2个数据集进行制图模拟,基于地图直接对比方式,分析2种常用插值方法制图稳定性及其影响因素。通过5次抽样制图对比分析,结果表明:IDW法制图差异度主要集中在<20%以下,差异度较高区域在空间上分布相对分散。普通Kriging法在制图差异度<10%和10%～20%的区域所占比例低于IDW法,在差异度>30%所占比例远高于IDW法,空间上呈大片面状分布。IDW法插值结果稳定性相对普通Kriging法高,制图结果空间格局的重现率更高。制图结果稳定性受样本数据空间结构性影响,空间变异越大,插值结果差异值和差异度越大,且普通Kriging方法受样本点数据空间变异性影响更为明显。因此,在开展土壤重金属污染评价生产实践过程中,对于空间变异性较大的土壤属性,以及对空间变异性较大的局部区域,需要考虑增加采样点密度,以提高插值的精确度和保证插值结果反映真实土壤属性空间分布的规律。
722 agricultural topsoil samples were collected in the study area. Samples were randomly divided into two subgroups, each subgroup has 361 samples. The spatial distribution of containment Cd and Cu in soils were predicted using two subgroups respectively. Inverse distance weighted method(IDW) and ordinary Kriging method were applied in the mapping progress to predict the spatial distribution. The stability and influence factors were assessed and compared based on map direct comparison. The prediction from 5 samples were compared each other, showing that the difference level of result focused on below 20% by IDW method. The spatial distribution of higher difference level is relatively dispersed. The proportion of area in difference level<10% and 10%~20% by ordinary Kriging method was lower than that by IDW method. The area in difference level>30% by ordinary Kriging method was more and centralized than that by IDW method. The stability of IDW interpolation is higher than that of ordinary Kriging method. It is affected by the structural variability of samples, the larger of structural variability, the greater of difference level. The ordinary Kriging method is more influenced by the structural variability of the samples. Therefore, in the assessment process of the soil heavy metal pollution, for larger spatial variability of soil properties,as well as to the spatial variability of the larger local area, it is necessary to consider increasing the density of sampling points to improve the accuracy of interpolation and to ensure that the interpolation results reflect the spatial distribution of soil properties.
作者
张金兰
黄铁兰
黄秋鑫
ZHANG Jinlan;HUANG Tielan;HUANG Qiuxin(Department of Information Engineering in Surveying Mapping and Remote Sensing, Guangdong Polytechnic College of Industry and Commerce, Guangzhou 510510, China;CEPREI Environmental Assessment and Monitoring Center, The 5th Electronics Research Institute of the Ministry of Industry and Information Technology, Guangzhou 501610, China)
出处
《环境科学与技术》
CAS
CSCD
北大核心
2019年第3期206-213,共8页
Environmental Science & Technology
基金
广东省省级科技计划项目(2016A040403039)
关键词
地图直接对比
插值方法
稳定性
影响因素
direct map comparison
interpolation methods
stability
influence factor
