高级检索
    周玉婷, 葛雪贞, 邹娅, 郭思维, 王涛, 陶静, 宗世祥. 基于Maxent模型的长林小蠹的全球及中国适生区预测[J]. 北京林业大学学报, 2022, 44(11): 90-99. DOI: 10.12171/j.1000-1522.20210345
    引用本文: 周玉婷, 葛雪贞, 邹娅, 郭思维, 王涛, 陶静, 宗世祥. 基于Maxent模型的长林小蠹的全球及中国适生区预测[J]. 北京林业大学学报, 2022, 44(11): 90-99. DOI: 10.12171/j.1000-1522.20210345
    Zhou Yuting, Ge Xuezhen, Zou Ya, Guo Siwei, Wang Tao, Tao Jing, Zong Shixiang. Prediction of the potential geographical distribution of Hylurgus ligniperda at the global scale and in China using the Maxent model[J]. Journal of Beijing Forestry University, 2022, 44(11): 90-99. DOI: 10.12171/j.1000-1522.20210345
    Citation: Zhou Yuting, Ge Xuezhen, Zou Ya, Guo Siwei, Wang Tao, Tao Jing, Zong Shixiang. Prediction of the potential geographical distribution of Hylurgus ligniperda at the global scale and in China using the Maxent model[J]. Journal of Beijing Forestry University, 2022, 44(11): 90-99. DOI: 10.12171/j.1000-1522.20210345

    基于Maxent模型的长林小蠹的全球及中国适生区预测

    Prediction of the potential geographical distribution of Hylurgus ligniperda at the global scale and in China using the Maxent model

    • 摘要:
        目的  长林小蠹是一种重要的林木检疫性害虫,对智利、新西兰等多国松树造成严重危害,并随林木进口入侵我国山东,对我国林业安全造成威胁。根据长林小蠹的已知分布,对其全球和中国范围内的适生区范围进行准确预测,明确该虫在我国潜在的分布区范围,为相关部门提供理论参考,有助于采取科学高效的监测和防治措施,降低潜在的生态和经济损失。
        方法  利用Maxent物种分布模型,通过筛选生物气候变量和优化模型参数构建拟合优度和复杂度综合表现最佳的模型,进而应用于长林小蠹在全球和中国范围内的适生区的预测。利用ArcGIS软件对预测结果进行可视化处理和面积统计,并利用模型的结果对影响长林小蠹定殖的主要环境因素进行分析。
        结果  相较默认参数下的模型,参数优化后的模型精度明显提高,平均AUC值达到0.964 6。同时,模型预测结果显示:长林小蠹在全球的适生区主要位于欧洲、北美洲东部和西部沿海、亚洲东部、南美洲东南部、非洲最南端以及大洋洲的东南沿海。其中高度适生区、中度适生区和低度适生区的面积分别占比1.21%、1.92%和3.95%。中国范围内该虫的适生区主要位于中部和南部,其南界至中国台湾,北界至辽宁大连。其中高度适生区、中度适生区和低度适生区的面积分别占比0.28%、5.00%和13.43%。在所有生物环境变量中,贡献率最为显著的为最干季降水量、最冷月份的最低温度、温度季节性、最热月份的最高温度、年降水量。贡献率分别为37.9%、25.4%、12.7%、7.0%和6.6%。
        结论  已有长林小蠹发生的区域外,美国东南部和西北部、阿根廷东部、巴西的南部和中国大部的适生区尚未有长林小蠹的捕获记录,被入侵风险较高。长林小蠹在中国的适生区主要以中度和低度适生区为主,几乎囊括了中国中部和南部所有的省份,其中,山东省和江苏省的沿海地带及其周边地带、中国西部局地和中国中南部局地的适生性相对较高。由此我们建议在国内开展长林小蠹的普查及防控工作,严防其扩散蔓延;同时加强入境口岸的各项检疫措施。

       

      Abstract:
        Objective  Hylurgus ligniperda is an important quarantine pest of pine species, which has caused severe damages on pines in Chile, New Zealand and many other countries. The pest was recently confirmed to have invaded and established populations in Shandong Province of eastern China through imported wood materials, posing threats to the forestry security of China. Thus, accurate prediction of the climate-suitable regions of H. ligniperda is urgent for relevant institutions to take efficient quarantine and control measures, thus reducing the potential ecological and economic losses.
        Method  Our research used Maxent model to predict the climate-suitable regions of H. ligniperda both at the global scale and in China, following the steps of bioclimatic variable selecting and model parameter optimizing. ArcGIS software was used to visualize the predictive results and make area statistics. The model results were used to analyze the contributing bioclimatic variables affecting the colonization of the pest.
        Result  Compared with the Maxent model with the default parameters, the accuracy of the model with optimized parameters was significantly improved, with the average AUC value of 0.964 6. Meanwhile, the predictive results showed that the climate-suitable regions were mainly distributed in Europe, the east and west coasts of North America, the east of Asia, the southeast of South America, the southernmost tip of Africa and the southeast coast of Oceania, with the very suitable regions, suitable regions and marginal regions accounting for 1.21%, 1.92% and 3.95%, respectively. Within China, the climate-suitable regions of the pest were mainly located in the middle and south zones, which was bounded to Taiwan Province in the South and Dalian of Liaoning Province in the north. The areas of very suitable regions, suitable regions and marginal regions accounted for 0.28%, 5.00% and 13.43%, respectively. The bioclimatic variables having the most significant contribution are the precipitation of the driest quarter (37.9%), the minimum temperature of the coldest month (25.4%), the temperature seasonality (12.7%), the maximum temperature of the warmest month (7%) and the annual precipitation (6.6%).
        Conclusion  Except for the areas where H. ligniperda already existed, the pest has not occurred in the climate-suitable regions in the southeast and northwest of the United States, the east of Argentina, the southern Brazil and most of China, indicating the high risk of H. ligniperda invasion in these regions. The climate-suitable areas of H. ligniperda in China are mainly represented by suitable regions and marginal regions, including almost all provinces in central and southern China, among them the coastal areas and surrounding regions of Shandong Province and Jiangsu Province, Western China, central and southern China have relatively higher suitability. Therefore, it is suggested that the survey and prevention measures should be implemented to control the spread of H. ligniperda, while the quarantine and control on wooden material should be strengthened at entry ports.

       

    /

    返回文章
    返回