摘要
为了开展危险性林木病害白蜡枯梢病的风险分析和检疫工作,研究对白蜡枯梢病菌在中国的潜在适生区进行了预测。利用Max Ent软件和DIVA-GIS数据库,选择19个环境因子变量和2.5 min的空间分辨率,并对预测结果的可靠性进行ROC曲线法分析评价,同时采用Jackknife刀切法分析了各环境变量对预测结果的影响。结果表明:该病菌在我国局部地区有一定适生风险;根据得到的潜在适生区预测图,白蜡枯梢病菌在中国的高度适生区有台湾中部,中度适生区有西藏西南边陲林区、湖北神农架和宜昌三峡林区、湖南武陵山和张家界林区、广东、广西、重庆、贵州、江西、福建的局部等地;各环境因子贡献率分析表明最冷季降水量对白蜡枯梢病菌的适生区分布影响最大,研究预测结果AUC值为0.932,表明预测结果可信度高。
For the risk analysis of the disease ash dieback, this study predicted potential distribution regions in China of the fungus Chalara fraxinea by using Max Ent software and DIVA-GIS database. 19 environmental factors and 2.5 min spatial resolution were selected, and the ROC curve and Jackknife analysis method were used to evaluate the reliability of prediction results and the influence of environmental variables on the prediction results, respectively. The results showed that C. fraxinea could distribute in some areas in China. According to the research figure, the middle of Taiwan was the highly suitable area, and moderately suitable areas included those forest regions located in southwest Tibet, Hubei Shennongjia and Yichang Three Gorges, Hunan Wuling Mountain and Zhangjiajie, as well as some parts of Guangdong, Guangxi, Chongqing, Guizhou, Jiangxi, Fujian etc. The precipitation in the coldest quarter was the most important environmental factor which could influence the potential distribution of C. fraxinea. The AUC value of the predication results was 0.932, which indicated that the results were highly reliable.
出处
《湖南农业科学》
2015年第10期83-87,共5页
Hunan Agricultural Sciences
基金
国家质检总局科技计划项目(2014JK015)
