摘要
为了掌握梨木虱在云南的潜在分布区,为梨木虱在云南省的发生、早期预警和防控提供参考资料,利用梨木虱的分布点数据和环境因子数据,通过MaxEnt生态位模型预测了梨木虱在云南省的适生范围区。结果表明:梨木虱在云南省适生范围分为4级,高适生区以黑色表示;中适生区以深灰色表示;低适生区以浅灰色表示;非适生区以白色表示。利用ROC曲线分析法对预测结果进行验证表明:其训练数据和测试数据分别为0.995和0.997,远大于随机分布模型的AUC值0.5,说明预测结果可靠。
In order to grasp the potential distribution of the Cacopsylla chinensis in Yunnan Province, to provide reference for insect pest situation, early warning, the prevention and control of C. chinensis. Ecological niche modelling technique, Maximum Entropy （MaxEnt）, were used to predict potential distribution of Cacopsylla cbinensis in Yunnan Province based on associations between known occurrence records and a set of environmental variables. The results showed that： the suitable areas for C. chinensis were divided into 4 grades by different colors： black was for the highly suitable areas; the moderate suitable areas was dark grey; light grey was on half of the low suitable areas and white represented the unsuitable areas. The results were proved by ROC analytical method and the AUC value of MaxEnt were 0.995 arid 0.997, which was greater than 0.5 of the stochastic AUC value and it meant that the prediction was highly dependable.
出处
《中国农学通报》
CSCD
2013年第25期97-101,共5页
Chinese Agricultural Science Bulletin
基金
国家林业局"948"项目"仁果类果树综合栽培管理技术引进"(2010-04-07)
