空间分层抽样在森林覆盖面积监测中的应用  

The application of spatial stratified sampling in remote sensing monitoring of forest cover

作　　者：张小伟[1,2] 佘光辉[2] 温小荣[2] 林国忠[2] 曹林[2] 罗扬 

机构地区：[1]浙江省森林资源监测中心,浙江杭州310020 [2]南京林业大学森林资源与环境学院,江苏南京210037 [3]贵州省林业科学院,贵州贵阳550000

出　　处：《南京林业大学学报（自然科学版）》2012年第3期81-84,共4页Journal of Nanjing Forestry University：Natural Sciences Edition

基　　金：国家自然科学基金项目(30571491);国家林业公益性行业科研专项项目(200804006);江苏省林业三项工程项目(lysx(2009)46)

摘　　要：为探讨空间分层抽样在森林资源调查中的应用,以对贵阳市花溪区的森林覆盖面积调查为例,利用融合后遥感数据SPOT5作为底图,布设100 m×100 m的抽样框,并从SPOT5影像上提取辅助因子进行先验抽样。采用Moran's I指数进行空间自相关分析,结果表明,调查目标具有显著空间自相关性,可以进行空间抽样。利用460个样本单元的观测值对整个研究区进行普通克里格空间插值,并以插值结果作为分层辅助因子进行空间分层,结果初步表明空间分层抽样与经典抽样和随机空间抽样相比具有明显优越性,可以降低成本,提高效率。In this paper,we selected the forest cover per hectare in Huaxi district,Guiyang city as an example to explore the application of spatial stratified sampling for forest resources.By using the information from SPOT5 image as foundation,a 100 m×100 m sampling frame was constructed,the information extracted from SPOT5 image were used to make priori sampling.The results showed that there is significant spatial autocorrelation of the forest cover per hectare from Moran＇s I.The result of ordinary Kriging interpolation is used as the auxiliary factor to stratify.Our research initially show that spatial stratified sampling is more efficient compared with classic sampling or spatial random sampling.

关 键 词：空间分层抽样 森林覆盖面积 Moran's Ⅰ指数 普通克里格 

分 类 号：S757[农业科学—森林经理学]