南京林业大学学报(自然科学版) ›› 2020, Vol. 44 ›› Issue (6): 161-168.doi: 10.3969/j.issn.1000-2006.201909052

• 研究论文 • 上一篇    下一篇

栎树猝死病在中国的入侵风险评估

林司曦(), 叶建仁*()   

  1. 南京林业大学,南方现代林业协同创新中心,江苏省有害生物入侵预防与控制重点实验室,南京林业大学林学院,江苏 南京 210037
  • 收稿日期:2019-09-30 修回日期:2020-09-08 出版日期:2020-11-30 发布日期:2020-12-07
  • 通讯作者: 叶建仁
  • 基金资助:
    国家自然科学基金项目(31800543);国家重点研发计划(2018YFD0600203);江苏高校优势学科建设工程资助项目(PAPD);江苏高校品牌专业建设工程资助项目(TAPP)

Invasion risk analysis of Phytophthora ramorum in China

LIN Sixi(), YE Jianren*()   

  1. Co-Innovation Center for the Sustainable Forestry in Southern China, Jiangsu Province Key Laboratory for Prevention and Management of Invasive Specie, College of Forestry, Nanjing Forestry University, Nanjing 210037, China
  • Received:2019-09-30 Revised:2020-09-08 Online:2020-11-30 Published:2020-12-07
  • Contact: YE Jianren

摘要:

【目的】栎树猝死病是一种为害林木和观赏植物的毁灭性病害,发病迅速,短期内即可造成寄主植物大量死亡,寄主范围非常广泛,主要为害阔叶树树种。对其在我国的适生区范围进行预测,并系统评估其入侵风险,有助于更好地制定针对性的防治及检疫措施。【方法】采用MaxEnt生态位模型,以栎树猝死病菌现有分布点的环境变量为基础运算预测模型,结合地理信息系统ArcGIS绘制其在中国的适生预测图;并以南京林业大学有害生物入侵预防与控制重点实验室建立的多指标综合评价体系为标准,从5个准则层下设18个指标层因子,对栎树猝死病菌在我国的入侵风险进行了定性和定量的分析。【结果】MaxEnt模型测试遗漏率与预测遗漏率基本吻合,ROC(receiver operating characteristic, ROC)曲线的AUC值(area under curver, AUC)为0.974,标准差为0.008,明显高于随机分布模型,说明该模型对栎树猝死病菌在我国的适生区预测结果可信度较高,可作为后续评估依据。栎树猝死病菌在中国的适生范围在101.9°~122.9° E,18.9°~38.0° N,主要位于我国秦岭淮河以南的南方地区,集中在长江中下游平原和武夷山脉、南岭以南的沿海地区,约占中国行政区划面积的19.6%。栎树猝死病菌在中国的入侵风险指标值R为2.64,属于极高风险的有害生物。【结论】模型预测结果的可信度较高,鉴于栎树猝死病菌在我国暂无分布记录,且在多指标评价体系中被归为极高风险等级,建议在进境检疫中对其可能寄主植物实施严格检疫及2年以上的隔离试种,防止其进入中国。

关键词: 栎树猝死病菌, 适生区分布, MaxEnt, 多指标综合评价体系, 入侵风险分析

Abstract:

【Objective】Sudden oak death caused by Phytophthora ramorum is known as the destructive disease which is usually found on trees and ornamental plants. It has very wide range of host species, mostly broad leaved trees, and kills them very quickly. This disease has not been found in China, but already being listed as the quarantine species by Chinese government based on its potential risk and prevention difficulty. It may become a big threat to our ecological environment security once it has been introduced someday. Thus, it is particularly important to predict the potential distribution and analyse the invasion risk of P. ramorum in China, which will help us to make a pertinent prevention and quarantine measures.【Method】The maximum entropy (MaxEnt) model based on environmental variables of the existing distribution areas was used to generate the suitable prediction map of P. ramorum with the geographic information system ArcGIS. The invasion risk analysis was performed using a multi-index comprehensive evaluation system, including 5 hierarchies (possibility of incoming, colonization, diffusion, the victim host economic importance and difficulty of risk management) and 18 indicators. 【Result】The test omission rate of MaxEnt model was in accordance with the predicted omission rate, which indicated the accuracy of this model and it could be used as the evaluation benchmark. The results showed that the potential distribution of P. ramorum was mainly located in the southern China. The geographical coordinates were ranged from 101.9 °-122.9 °E, 18.9 °-38.0 °N, including Jiangsu, Anhui, Shanghai, Henan, Hubei, Chongqing, Hunan, Guizhou, Jiangxi, Zhejiang, Fujian, Guangdong, Guangxi, Hainan, Hongkong, Macao, Taiwan and some areas of Shandong, Sichuan, Shanxi and Tibet, accounting for about 19.6% of administrative area of China. The invasion risk index value (R) of P. ramorum was 2.64 which suggested it should belong to an extreme high risk grade of harmful species. 【Conclusion】Due to no national record of P. ramorum and its high risk level in China, we believed that strict quarantine measures and more than 2 years isolated planting were mandatory to prevent the disease.

Key words: Phytophthora ramorum, potential distribution, maximum entropy model (MaxEnt), Integrated multi-index evaluation system, invasion risk analysis

中图分类号: