中国媒介生物学及控制杂志 ›› 2014, Vol. 25 ›› Issue (5): 393-398.DOI: 10.11853/j.issn.1003.4692.2014.05.003

• 论著 • 上一篇    下一篇

基于最大熵模型的中华按蚊潜在分布预测

马爱民1, 王劲峰2, 王多全3, 任周鹏2, 4   

  1. 1 中国矿业大学(北京),地球科学与测绘工程学院,北京 100083;
    2 中国科学院地理科学与资源研究所,资源与环境信息系统国家重点实验室;
    3 中国疾病预防控制中心寄生虫病预防控制所;
    4 中国科学院大学
  • 收稿日期:2014-07-16 出版日期:2014-10-20 发布日期:2014-10-20
  • 通讯作者: 王劲峰,Email: wangjf@lreis.ac.cn
  • 作者简介:马爱民,男,硕士研究生,主要从事地理信息系统空间分析研究,Email: maam@lreis.ac.cn
  • 基金资助:
    国家“973”重大科学研究计划(2012CB955503)

Prediction of potential distribution of Anopheles sinensisin China based on MaxEnt

MA Ai-min1, WANG Jin-feng2, WANG Duo-quan3, REN Zhou-peng2,4   

  1. 1 College of Geoscience and Surveying Engineering, China University of Mining and Technology, Beijing 100083, China;
    2 State Key Laboratory of Resources and Environmental Information System, Institute of Geographic Sciences and NaturalResources Research, Chinese Academy of Sciences;
    3 The Chinese Center for Disease Control and Preventionof Parasitic Disease Prevention and Control;
    4 University of Chinese Academy of Sciences
  • Received:2014-07-16 Online:2014-10-20 Published:2014-10-20
  • Contact: WANG Jin-feng, Email: wangjf@lreis.ac.cn
  • Supported by:
    Supported by the National “973” Major Scientific Research Projects (No. 2012CB955503)

摘要: 目的 构建中华按蚊潜在分布和生物气候因子关系模型,研究影响中华按蚊潜在分布的主导气候因子,预测中华按蚊在中国的潜在分布,为在中国消除疟疾提供媒介分布的数据支持。方法 根据2005-2010年媒介按蚊监测数据,应用最大熵模型(MaxEnt),采用“10%训练存在逻辑阈值”定义最小适生环境阈值,划分中华按蚊的适生区和非适生区。利用地理信息系统(GIS)估算暴露于中华按蚊分布区内的人群数量。结果 采用受试者工作特征曲线(ROC曲线)分析方法来检验模型精度,得到的AUC值2005-2010年分别为0.814、0.791、0.783、0.801、0.774和0.802,显示预测精度较好。模型结果表明,年总降雨量、年平均气压、最湿季降雨量、最冷月最低气温对中华按蚊的分布有着重要影响。结论 2005-2010年,中华按蚊在中国的潜在分布和暴露人口均呈缩减趋势。掌握并了解中华按蚊在中国的潜在分布,对于中国消除疟疾和预防相关传染病具有重要意义。

关键词: 中华按蚊, 最大熵模型, 地理分布, 暴露人口

Abstract: Objective To build a model for the relationship between the potential distribution of Anopheles sinensis and relevant bio-climatic factors, and to identify the main climatic influencing factors as well as predict the potential distribution of An. sinensis in China, thereby providing supporting data of vector distribution for the nationwide elimination of malaria. Methods  A MaxEnt model was built to predict the potential distribution of An. sinensis using monitoring data from 2005 to 2010. The potential distribution areas of An. sinensis were divided into suitable and unsuitable areas, where “10 percentile training presence logistic threshold” was used to define the minimum threshold of suitable environment. The size of human population exposed to the distribution area of An. sinensis was evaluated using geographic information system. Results In the MaxEnt model, a receiver operating characteristic (ROC) curve was used to test the precision. The values of area under the ROC curve for 2005 to 2010 were estimated to be 0.814, 0.791, 0.783, 0.801, 0.774, and 0.802, respectively, indicative of good prediction precision. The modeling data showed that total annual precipitation, mean annual air pressure, precipitation of the wettest quarter, and minimum temperature of the coldest month strongly influenced the distribution of An. sinensis. Conclusion In China, the suitable area of An. sinensis and the exposed human population both showed a decreasing trend from 2005 to 2010. It is of great significance to the nationwide elimination of malaria and the prevention of related infectious diseases by grasping and understanding the potential distribution of An. sinensis in China.

Key words: Anopheles sinensis, MaxEnt model, Potential distribution, Exposed population

中图分类号: