湘江中下游肾综合征出血热传播风险预测和环境危险因素分析
作者简介:肖 洪(1964-),女,湖南湘潭县人,博士,副教授,主要研究方向为空间流行病学。E-mail:xiaohong.hnnu@gmail.com
收稿日期: 2012-05-05
要求修回日期: 2012-07-22
网络出版日期: 2013-01-20
基金资助
湖南省重点学科建设项目(2008001)、湖南省自然科学基金(11JJ3119)、湖南省高校创新平台开放基金(11K037)、湖南师范大学地图学与地理信息系统校级重点学科(2011001)资助
世界野生动物基金会
Environmental Factors Contributing to the Spread of Hemorrhagic Fever with Renal Syndrome and Potential Risk Areas Prediction in Midstream and Downstream of the Xiangjiang River
Received date: 2012-05-05
Request revised date: 2012-07-22
Online published: 2013-01-20
Copyright
肖洪 , 林晓玲 , 高立冬 , 代翔宇 , 贺新光 , 陈碧云 , 张锡兴 , 赵暕 , 田怀玉 . 湘江中下游肾综合征出血热传播风险预测和环境危险因素分析[J]. 地理科学, 2013 , 33(1) : 123 -128 . DOI: 10.13249/j.cnki.sgs.2013.01.123
HFRS (human hemorrhagic fever with renal syndrome) occurs at very high frequency in the midstream and downstream of the Xiangjiang River. It is of great significance for understanding the regulation of the transmission of HFRS in this area due to its major impact on local economy development and public health. The risk factors and potential risk of HFRS in the midstream and downstream of theXiangjiang River were explored by combining the ecologic niche modeling, geographic information systems (GIS) and remote sensing technique in this article. The average area under the Receiver Operating Characteristic curve was 0.785. The 64 cases in 2010 were predicted correctly except for one case in a low risk area. These showed that the model possess good prediction performance. The modeling results indicated that the north of Wangcheng County and Changsha County, the border of Hengdong County , Zhuzhou County and Xiangtan Countiy, and the south of Hengdong County were predicted as the potential risk area. The following factors were found to be closely relative to the highest risk of HFRS transmission: mean annual temperature of 18℃, annual precipitation of about 1 500 mm, at urban and constructive land, and normalized difference vegetation index between 0.3 and 0.4 in May and July and with low elevation. The transmission of HFRS in the midstream and downstream of the Xiangjiang River was analyzed by using the ecologic niche model combining the geographical environment, climatic condition and human activity data. Our results are more accurate than those from other models and these findings in the study. Therefore, they are applicable for targeting control and prevention efforts.
Fig.1 Location of the study area图1 研究区地理位置 |
Table 1 Environmental variables of ecologic niche models表1 生态位模型环境变量 |
变量名称 | 数据来源 | 数据类型 |
---|---|---|
土地利用 | MODIS遥感影象提取与分类 | 分类型 |
CTI | 美国地质调查局 | 连续型 |
MHT | 世界野生动物基金会 | 分类型 |
人迹指数 | 国际地球科学信息中心 | 连续型 |
生态系统 | 世界环保组织 | 分类型 |
温度 | 中国气象科学数据服务共享网 | 连续型 |
降水量 | 中国气象科学数据服务共享网 | 连续型 |
NDVI | TM遥感影响提取 | 连续型 |
高程 | 国际科学数据服务平台 | 连续型 |
坡度 | 高程数据中提取 | 连续型 |
1.3 最大熵值法生态位模型 通过构建生态位模型[10,11,22],根据已知病例分布数据与相关环境变量(如用地类型和气象因子),探索疾病传播对地理景观和气象因子的潜在需求,对区域内的疾病传播风险进行模拟和预测[23]。本研究使用最大熵值法生态位模型探索已知病例分布区环境特征与研究区的非随机关系,根据最大熵模型,选择最优分布,对疫情进行预测[23,24]。假设研究区X是由有限数量的网格单元所组成,π为疫病在研究区内的分布,为π的分布赋予每个单元x的值,总和为1,在满足已知各种环境变量fj(j=1,2,…,n)的条件下(式2和3),根据π的近似期望分布的熵(式1)选择熵最大的分布作为疫病的最优分布。 |
Fig. 2 The distribution of human hemorrhagic fever with renal syndrome (HFRS) cases in 2005-2009 and the predicted HFRS risk in 2010 by Maximum Entropy Model in midstream and downstream of the Xiangjiang River图2 湘江中下游HFRS发病情况与生态位模型预测结果 |
Fig. 3 Response curves for the variables related to presence of HFRS(Red lines are mean values for the 10 Maxent runs and blue bars represent ±1 SD)图3 湘江中下游环境对肾综合征出血热传播的响应响应曲线 |
The authors have declared that no competing interests exist.
[1] |
|
[2] |
|
[3] |
|
[4] |
|
[5] |
|
[6] |
|
[7] |
|
[8] |
|
[9] |
|
[10] |
|
[11] |
|
[12] |
|
[13] |
|
[14] |
|
[15] |
|
[16] |
|
[17] |
|
[18] |
|
[19] |
|
[20] |
|
[21] |
|
[22] |
|
[23] |
|
[24] |
|
[25] |
|
[26] |
|
[27] |
|
[28] |
|
[29] |
|
[30] |
|
[31] |
|
[32] |
|
[33] |
|
[34] |
|
/
〈 | 〉 |