福建省马尾松毛虫害空间自相关分析  

Analysis of the spatial autocorrelation of Dendrolimus punctatus Walker in Fujian

作　　者：许章华[1,2] 刘健[1,2,3] 余坤勇[1,2] 李聪慧[4] 龚从宏[1,2] 赖日文[1,2] 唐梦雅[1,2] 谢婉君[1,2] 李增禄[1,3] 

机构地区：[1]福建农林大学3S技术应用研究所,福州350002 [2]福建农林大学林学院,福州350002 [3]三明学院资源与化工学院,福建三明365000 [4]北京林业大学自然保护区学院,北京100083

出　　处：《安全与环境学报》2013年第6期167-171,共5页Journal of Safety and Environment

基　　金：国家自然科学基金项目(30871965);福建省科技计划重点项目(2011N0031);福建省水利厅科研基金项目

摘　　要：马尾松毛虫是危害最严重的森林害虫,掌握该虫害的空间分布规律有助于实现其有效监测与预防。以福建省67个县域单元为例,对马尾松毛虫害全局和局部空间自相关性进行分析。结果表明:1)福建省马尾松毛虫害Moran's I的值为0.613 5,Z值为13.464 2,达到极显著水平,表明虫害存在极为显著的全局正空间自相关性,集聚效应明显;2)从局部空间自相关结果看,空间差异小的县域单元数量远大于差异大的数量,"高-高"、"低-高"及"高-低"相关均位于森林资源丰富的闽中、闽北各县,呈"3横2纵"状分布格局,"低-低"正相关均位于沿海一线,呈带状分布格局,沙县、永安、邵武是马尾松毛虫害分布的"冷点",建瓯则是虫害分布的"热点";3)在大部分县(市),马尾松毛虫害空间自相关性不明显,呈随机型分布。研究表明,空间自相关分析可在马尾松毛虫害分布特征识别与规律研究方面得到较好的应用。This paper is aimed to explore the spatial distribution reg ularity of Dendrolimus punctatus Walker relying on the technology of geographic information system （GIS） by taking the 67 counties of Fu jian as the case study sample. In this paper, we have conducted the investigation of the global spatial autocorrelation and the local spatialautocorrelation analysis of D. punctatus with the method of GIS. As is known, D. punctatus Walker is one of the pests that lead to the serious damage of forest. Therefore, it is necessary to grasp and con trol the spatial distribution regularity of the said pest, which can help to monitor and prevent the pest damage effectively. The socalled spatial autocorrelation analysis is by nature a quantitative statistical method, which is fit for the analysis of D. punctatus Walker. The results of our analysis indicate that, （1） The Moran＇ s I value of D. punctatus in Fujian proves to be 0.613 5, while Z = 13.4642. The results imply that there exists a more significant positive global spatial autocorrelation of this pest distribution in the province, with an obvi ous cluster effect. （2） The local spatial autocorrelation results we have gained demonstrate that, the number of county with smaller spa tial difference is much more than those of larger spatial difference. The correlations of ＂High High＂, ＂Low High＂ and ＂High Low＂ all belong to the areas with abundant forest resources in the central and northern Fujian, which turns to be of the ＂3 horizontal and 2 ver tical＂ distribution pattern. The positive correlation of ＂Low Low＂ turns to appear in the costal line area, such as in Shaxian, Yongan and Shaowu, the ＂cold spots＂ of the distribution of D. punctatus, while the ＂hot spots＂ lie in Jianou. （3） The correlation of D. punc tatus in the most counties of the province does not prove to be obvi ous, and the pest damage turns to show a random distribution. Thus, the above facts imply that the method of spatial autocorrelation analy sis can be ef

关 键 词：森林经理学 马尾松毛虫害 全局空间自相关 局部空间自相关 Moran's I Moran散点图 LISA 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]