文章摘要
高秉博,郝朝展,李发东,胡茂桂,李晓岚,郜允兵,潘瑜春.面向土壤环境质量等级划分的统计推断与加密采样优化方法研究综述[J].农业环境科学学报,2021,40(4):712-722.
面向土壤环境质量等级划分的统计推断与加密采样优化方法研究综述
Statistical inference and the additional sampling optimization method for soil environmental quality grade classification: A review
投稿时间:2020-10-19  
DOI:10.11654/jaes.2020-1202
中文关键词: 环境质量  等级划分  加密采样  统计推断
英文关键词: environmental quality  grade classification  additional sampling  statistical inference
基金项目:国家重点研发计划项目(2016YFD0800301);国家自然科学基金项目(41601425)
作者单位
高秉博 中国农业大学土地科学与技术学院, 北京 100083
农业农村部农业灾害遥感重点实验室, 北京 100083 
郝朝展 中国农业大学土地科学与技术学院, 北京 100083 
李发东 中国科学院地理科学与资源研究所, 北京 100011 
胡茂桂 中国科学院地理科学与资源研究所, 北京 100011 
李晓岚 国家农业信息化工程技术研究中心, 北京 100097 
郜允兵 国家农业信息化工程技术研究中心, 北京 100097 
潘瑜春 国家农业信息化工程技术研究中心, 北京 100097 
摘要点击次数: 1823
全文下载次数: 1939
中文摘要:
      环境质量等级划分是土壤环境调查和监测的一项重要目标。不同于总体估计和空间插值,等级划分只需要准确估计污染物浓度与等级划分阈值之间的大小关系,同时由于土壤环境质量等级具有空间连续性,加密采样时应该将样点布设在等级过渡地带,准确获取等级边界。但是初步调查的样点一般较为稀疏,导致等级估计结果具有较大的不确定性,难以反映真实的等级过渡情况,而且具体应用对两类等级划分错误也有不同的控制要求,构成了土壤环境质量等级划分及加密采样面临的特殊问题。本文分析梳理了国内外土壤环境质量等级划分统计推断与加密采样布局优化方法的研究进展,归纳总结了以等级划分为目的的统计推断方法与加密采样布局优化方法体系,并探讨了下一步研究的方向,为我国目前开展的农用地土壤环境类别划分、工矿企业土壤污染修复区确定等重大需求提供方法选择依据。
英文摘要:
      Environmental quality grade classification is one crucial goal of soil surveying and monitoring. Unlike global mean estimation and spatial interpolation, environmental quality grade classification only focuses on estimating the relationship between pollutant concentration and grading threshold. Due to spatial autocorrelation, soil environmental quality grades are usually spatially continuous. Thus, to improve the classification precision of environmental quality grades, more sampling points should be laid in the transition area for environmental quality grades. However, because the primary survey sampling points are usually sparse and produce large errors and uncertainty in the estimated result, it is difficult to reflect the transition area of true grades. Additionally, applications have different requirements for the two types of errors in grade classification. The particularity of environmental quality grade classification needs special inference and sampling optimization design methods. This study analyzed the inference and additional sampling optimization design method for environmental quality grade classification locally and globally. The method framework is also summarized, and future research is discussed. This work's outcome can provide a method selection basis for several agricultural soil environmental management applications, such as soil environmental quality grade classifications of agricultural land and definition of polluted areas for soil environmental remediation.
HTML    查看全文   查看/发表评论  下载PDF阅读器