文献类型：期刊文献

中文题名：基于气象因子的马尾松毛虫发生率空间格局研究

英文题名：Spatial Pattern of Dendrolimus punctatus Incidence Rate Based on Meteorological Factors

作者：王庆[1] 毕猛[1] 杜婷[1] 廖怀建[1] 石雷[1]

第一作者：王庆

机构：[1]中国林业科学研究院资源昆虫研究所

年份：2016

卷号：29

期号：2

起止页码：256-260

中文期刊名：林业科学研究

外文期刊名：Forest Research

收录：CSTPCD;;Scopus;北大核心:【北大核心2014】；CSCD:【CSCD2015_2016】；

基金：国家林业局林业公益性行业科研专项"重大森林病虫害防控技术的关键理论基础"(201204501);云南省技术创新人才培养计划(2012HB054)

语种：中文

中文关键词：马尾松毛虫;气象因子;发生率;偏最小二乘;空间格局

外文关键词：Dendrolimus punctatus Walker; meteorological factor; insect incidence rate; partial least squares; spa tial pattern

分类号：S791.248

摘要：[目的]为预测未来我国马尾松毛虫的潜在变化趋势,以2002—2012年全国范围内马尾松毛虫的地级逐年平均发生率作为预测指标,[方法]运用偏最小二乘回归方法,获得马尾松毛虫平均发生率与相关气象因子的回归方程,并结合地理空间数据与未来气象数据,得到马尾松毛虫平均发生率空间格局模型。[结果]表明:以筛选后的12个气象因子建立的马尾松毛虫平均发生率空间格局模型精度达到86.98%,具有较强的可靠性。据此预测2020s,2050s,2080s的马尾松毛虫平均发生率空间格局,并与2002—2012年的空间格局相比,结果显示:华东及华中地区虫害中度和重度发生面积均明显增加,有扩散的趋势;华东地区的轻度发生面积总体为缩减;而华南部分地区虫害轻度发生面积扩增。[结论]以偏最小二乘回归方法所得的空间格局模型具有实际预测意义,可以预测我国未来马尾松毛虫平均发生率的变化趋势。
[Objective]Taking the average incidence rate based on China＇s nationwide data from 2002- 2012 as indicator to predict the potential trend of Dendrolimus punctatus incidence rate. [Method]By means of partial least squares regression,the regression equation about average incidence rate and the related meteorological factors was obtained. Combined with the geographic spatial data and future meteorological data,the spatial pattern model of the average incidence rate of D. punctatus was established. [Result]The spatial pattern model of D. punctatus＇ average incidence rate built by 12 selected meteorological factors has the prediction accuracy of 86. 98%. Based on this model,the spatial pattern models for 2020 s,2050s,and 2080 s were established. It was predicted that compared with 2002- 2012,the area of moderate and severe insect pests in East and Central China would significantly increase,and there would be a trend of spreading. The mild incidence area would decrease in East China,while the mild incidence area would has a trend of amplification in parts of Southern China. [Conclusion]The spatial pattern model obtained by partial least squares regression method can be used to predict the potential changes of the average incidence rate of D. punctatus in China.