X img





作者投稿 专家审稿 编辑办公 编委办公 主编办公

冰川冻土 ›› 2020, Vol. 42 ›› Issue (4): 1353-1362.doi: 10.7522/j.issn.1000-0240.2019.0089

• 冰冻圈与可持续发展 • 上一篇    下一篇


德勒格日玛null1,2(), 李一平2(), 孟雪峰2, 田颖3, 计燕霞2, 张莫日根2   

  1. 1.南京信息工程大学, 江苏 南京 210044
    2.内蒙古自治区气象台, 内蒙古 呼和浩特 010051
    3.内蒙古气象培训干部学院, 内蒙古 呼和浩特 010051
  • 收稿日期:2018-05-12 修回日期:2019-04-12 出版日期:2020-12-31 发布日期:2019-09-10
  • 通讯作者: 李一平 E-mail:gerima_dele@aliyun.com;617321974@qq.com
  • 作者简介:德勒格日玛(1981 - ), 女, 内蒙古鄂尔多斯人, 高级工程师, 从事短期天气气候预测、 灾害风险评估及研究工作. E-mail: gerima_dele@aliyun.com
  • 基金资助:

Study of the risk evaluation of snow disaster in pastoral areas of Xilingol League, Inner Mongolia

Delegerima1,2(), Yiping LI2(), Xuefeng MENG2, Ying TIAN3, Yanxia JI2, Morgen ZHANG2   

  1. 1.Nanjing University of Information Science & Technology,Nanjing 210044,China
    2.Inner Mongolia Meteorological Observatory,Hohhot 010051,China
    3.Inner Mongolia Meteorological Training Center,Hohhot 010051,China
  • Received:2018-05-12 Revised:2019-04-12 Online:2020-12-31 Published:2019-09-10
  • Contact: Yiping LI E-mail:gerima_dele@aliyun.com;617321974@qq.com


选取内蒙古锡林郭勒盟与积雪有关的雪灾致灾指标, 以气温、 风速为气象条件孕灾环境指标, 坡度、 植被盖度为下垫面孕灾环境指标, 人口密度、 牧民纯收入、 人均GDP、 牲畜超载率等数据为承灾体脆弱性指标, 基于BP方法、 层次分析法、 建立了内蒙古锡林郭勒盟白灾综合风险评价体系, 并对其进行了风险评价与区划。为提高灾害评估的准确率, 白灾的灾害等级是以月为尺度进行评定, 选取的气象指标多数都是以月为尺度的指标。研究表明: 白灾与积雪因子高度相关, 是气候灾害, 积雪、 低温、 大风等气象因子长期作用的结果。对白灾尝试用BP神经网络法进行风险评估, 评估的灾害等级和实际灾害等级十分吻合, 用训练好的神经网络对各个旗县(1980 - 2015年)的白灾进行了风险评估, 评估效果理想。因此, 可以通过数值预报产品、 气候预测产品获取相关评价因子, 采用BP方法形成白灾风险预评估产品, 进而应用于雪灾风险评估业务中, 为相关部门提供决策依据。

关键词: 白灾, 锡林郭勒盟, 内蒙古, BP神经网络, 层次分析法, 综合风险评价


In this study, a model has developed for risk assessment of snow disaster in the pastoral area of Xilingol League based on BP (back propagation artificial neural network), AHP (analytic hierarchy process) and fuzzy comprehensive evaluation method by using MATLAB, ArcGIS and SPSS. The indicator system of the snow disaster risk assessment is composed of disaster inducing indicators related with the snow depth; the disaster pregnant environment meteorological indicators are mainly constitute of temperature and wind speed indexes; disaster pregnant environment underlying surface indicators are formed by slope and vegetation coverage; vulnerability indicators of disaster bearing bodies are constituted by population density, herdsmen’s net income, GDP per capita, overload rate of livestock. The curve of the predicted hazard level and actual hazard level fits perfectly. Therefore, in this paper, using BP to do the risk assessment of the snow disaster by monthly from November to March of the following year from 1980 to 2015 for each county. Snow hazard is climatic disaster which is highly related with snow cover; It is the result of long-term effects of snow cover, low temperature and strong wind. Except the index that the rainfall during grass growth period, the other indexes are monthly scale, so it can enhance the accuracy of hazard assessment. The effect of risk assessment is comparably ideal, since it is possible to form product of the risk grade of white dzud based on BP method through gaining numerical prediction products and climate prediction products. Consequently, these products will be gradually applied to daily business and provided decision-making basis for party committee and government.

Key words: white dzud, Xilingol League, Inner Mongolia, BP neural network, analytic hierarchy process, comprehensive risk assessment


  • P426.63+4