Abstract:

【Objective】 To better understand the potential spread and risk of willow phytoplasma in China.【Methods】 In combination with ArcGIS software, the maximum entropy model (MaxEnt) was used to predict the extent of willow phytoplasma in China's adaptive zones, and the model precision was validated by subject working characteristics (ROC), combined with jackknife-test screening for dominant environmental variables. At the same time, a quantitative analysis system of pest risk was set up to calculate the risk value of willow phytoplasma and evaluate it.【Results】The mean area (AUC) under the ROC curve of the MaxEnt model for willow phytoplasma was 0.9795, indicating that the model predicted reliable results. Willow phytoplasma is mainly concentrated in the northwestern, central, eastern, southwestern, northern, and northeastern regions of China in the temperate zones of 22°36'~49°10'N、73°40'~135°05'°E. Among them, the highly suitable areas are mainly in the northwest, central and eastern China, while the moderately suitable areas are mainly in the northeast and central China. The low-conforming areas are mainly in the northeast and southwest, while the non-conforming areas are mainly in southern China, Hong Kong, Macao and Taiwan. Analysis of the significance of environmental variables obtained by the MaxEnt model formalization training gain cutter method showed that the coldest season mean temperature (Bio11) and driest monthly precipitation (Bio14) had the greatest impact on the distribution of willow phycoplasmosis in China, contributing 72.1% and 17.4%, respectively. Isothermia (Bio3), driest quarterly precipitation (Bio17), coldest quarterly precipitation (Bio19), driest quarterly mean temperature (Bio9) had an impact, contributing 3.7%, 3.7%, 1.9%, 1.2%, in order of magnitude. This means that cooler, drier areas are conducive to the development of willow mycoplasmosis. The risk analysis resulted in the creation of a comprehensive multi-indicator evaluation system with 5 guideline levels and 14 indicator levels, and the quantitative analysis of each indicator level showed that willow phytoplasma had a risk value (R value) of 1.997 in our country, which is considered to be a moderately hazardous forest pest.【Conclusion】 The risk of willow phytoplasma is high, so it is necessary to establish a surveillance system and take effective control measures against the disease.

Key words: willow; phytoplasma; disease; fitness analysis; risk assessment

摘要：

【目的】基于MaxEnt模型的柳树植原体病害在我国的适生区预测及定量风险分析。【方法】利用最大熵模型(MaxEnt),结合ArcGIS软件,预测柳树植原体病害在中国的适生区范围,并通过受试者工作特征(receiver operating characteristic,ROC)验证模型精度,并结合刀切法(jackknife test)筛选主导环境变量。同时建立有害生物风险量化分析体系,计算柳树植原体病的风险性危害值并对其进行评价。【结果】柳树植原体病MaxEnt 模型预测结果的ROC曲线下面积(AUC)平均值为0.979 5,模型预测结果可靠。柳树植原体病害在我国的适生区在22°36'~49°10'N、73°40'~135°05'°E的中温带、暖温带、高原气候区,主要集中在我国的西北、华中、华东、西南、华北和东北地区。其中高度适生区主要集中在西北、华中、华东地区,中度适生区主要集中在东北地区和华中地区中部;低度适生区主要集中在东北和西南地区,非适生区主要集中在华南和港澳台地区。最冷季平均温度(Bio11)、最干月降水量(Bio14)对柳树植原体病在中国的分布状况影响最大,贡献率分别为72.1%和17.4%;等温性(Bio3)、最干季度降水量(Bio17)、最冷季度降水量(Bio19)、最干季度平均温度(Bio9)有一定影响,贡献率依次为3.7%、3.7%、1.9%、1.2%。低温、干旱的区域有利于柳树植原体病的发生。建立5 个准则层、14个指标层的多指标综合评价体系,并对各指标层定量分析,柳树植原体病在我国的风险性危害值(R值)为1.997,属于中度危险林业有害生物。【结论】柳树植原体病风险性较高,需要尽快建立监测体系并针对该病害采取有效控制措施。

关键词: 柳树, 植原体, 病害, 适生分析, 风险评估