基于MaxEnt模型分析新疆特色林果区春尺蠖发生风险

Potential Risk Zone Analysis of Apocheima cinerarius Erschoff disaster in Xinjiang Featured Forest Fruits

【目的】分析并预测新疆特色林果种植区域春尺蠖潜在风险发生范围,为新疆特色林果春尺蠖灾害的有效预防和重点防治提供依据。【方法】以新疆特色林果春尺蠖实际分布数据及环境因子为依据,采用MaxEnt模型结合GIS空间分析技术预测新疆特色林果春尺蠖潜在风险区,用聚类分析法划分风险等级,使用百分比贡献率分析其主要环境因子及生态位参数。【结果】(1)ROC评价显示新疆特色林果春尺蠖潜在风险区的训练数据集和测试数据集的AUC值分别为0.979和0.970,模拟效果优秀。(2)根据MaxEnt模型预测结果将新疆特色林果春尺蠖潜在分布区划分为风险区和无风险区,其中风险区总面积约745.38×104hm2。(3)将新疆特色林果春尺蠖潜在风险区划分为高风险区115.84×104hm2、中风险区201.28×104hm2、低风险区428.26×104hm2。(4)影响新疆特色林果春尺蠖灾害风险性的主要环境因子是最冷月最高温度、最冷季平均温度、最干季降水量、最湿季平均温度。最冷月最高温度>-14.25℃,最冷季平均温度>-4.01℃,最干季降水量在1.23~8.29mm,最湿季平均温度在21.63~24.82℃为新疆特色林果春尺蠖适宜生长的生态位参数。【结论】基于MaxEnt模型的新疆特色林果春尺蠖潜在风险区预测结果与春尺蠖实际分布区完全相符,明确了春尺蠖的地理分布特征:高风险区主要分布于喀什地区和和田地区,中风险区主要分布于阿克苏地区、巴音郭楞蒙古自治州、和田地区、喀什地区和克孜勒苏柯尔克孜自治州,低风险区布于阿克苏地区、巴音郭楞蒙古自治州、和田地区、喀什地区和克孜勒苏柯尔克孜自治州,在吐鲁番地区和乌鲁木齐市有少量分布。该研究对于制定检疫和防治政策具有较高的参考价值。

【Objective】 To analyze and predict the potential risk areas of Apocheima cinerarius Erschoff in Xinjiang fruit forests in the hope of providing important reference value for effective prevention and key prevention of Apocheima cinerarius Erschoff disaster in Xinjiang fruit forests.【Method】 Based on the actual distribution data and environmental factors data of Apocheima cinerarius Erschoff disaster in Xinjiang fruit forests, the potential risk areas of Apocheima cinerarius Erschoff in Xinjiang fruit forests were predicted by MaxEnt model and GIS spatial analysis technology, and the main environmental factors and niche parameters were analyzed by percentage contribution rate.【Results】 The results showed that:(1) The ROC evaluation (Receiver Operating Characteristic, ROC) showed that the AUC values of the training data set and the test data set of MaxEnt model predicting the potential risk area of Ectropis sinensis were 0.979 and 0.970 respectively, which indicated that the simulation effect was excellent.(2) According to the prediction results of MaxEnt model, the potential distribution area of Apocheima cinerarius Erschoff in Xinjiang fruit forests was divided into risk area and risk-free area, and the total area of risk area was about 74,538,800 hm 2.(3)Based on cluster analysis, the potential risk areas of Apocheima cinerarius Erschoff in Xinjiang fruit forests were divided into high-risk areas: 1.158 million , medium-risk areas:2.218 million hm 2and low-risk areas:4.226 million hm 2.(4) The main environmental factors affecting the risk of Apocheima cinerarius Erschoff disaster in fruit forests in Xinjiang were the highest temperature in cold month, the average temperature in the coldest season, the precipitation in the driest season and the average temperature in the wettest season. The response curves showed that the highest temperature in the cold month > 14.25 C, the average temperature in the coldest quarter > 4.01 C, the pre-cipitation in the driest quarter was between 1.23-8.29 mm, and the average temperature in the wettest quarter was between 21.63-24.82 C, which was the niche parameter suitable for the growth of Apocheima cinerarius Erschoff in Xinjiang fruit forests and fruits.【Conclusion】 Based on MaxEnt model, the prediction results of potential risk areas of Apocheima cinerarius Erschoff in Xinjiang characteristic forests are completely consistent with the actual distribution areas of Apocheima cinerarius Erschoff. The geographical distribution characteristics of Apocheima cinerarius Erschoff are clear: the high-risk areas are mainly located in Kashgar and Hetian areas, and the middle-risk areas are mainly located in Aksu area, Bayingol Mongolian Autonomous Prefecture and Hetian area. The low-risk areas are located in Aksu area, Bayingol Mongolian Autonomous Prefecture, Hotan area, Kashgar area and Kizilesu Kirgiz Autonomous Prefecture, with a small distribution in Turpan area and Urumqi city. This study has a high reference value for formulating quarantine and prevention policies.