欢迎访问《中国农学通报》,

中国农学通报 ›› 2019, Vol. 35 ›› Issue (9): 115-121.doi: 10.11924/j.issn.1000-6850.casb17110072

所属专题: 农业气象

• 资源 环境 生态 土壤 气象 • 上一篇    下一篇

气象干旱综合指数MCI在四川省的适用性分析及修订

王春学, 张顺谦, 陈文秀, 孙蕊   

  1. 四川省气候中心/高原与盆地暴雨旱涝灾害四川省重点实验室
  • 收稿日期:2017-11-20 修回日期:2019-03-13 接受日期:2018-03-13 出版日期:2019-03-26 发布日期:2019-03-26
  • 通讯作者: 张顺谦
  • 基金资助:
    四川省应用基础研究重点项目“四川盆地季节性干旱监测评估关键技术及机理研究”(2017JY0294)。

Applicability and Revision of MCI in Sichuan Province

  • Received:2017-11-20 Revised:2019-03-13 Accepted:2018-03-13 Online:2019-03-26 Published:2019-03-26

摘要: 使用1991-2013年四川省156站逐日气温、降水资料和干旱灾情资料,利用灰色关联方法建立了实际干旱灾害等级划分标准,进而分析了MCI指数在四川省的适用性,最后利用投影寻踪方法对MCI指数进行修订。结果表明,选择农作物受灾面积、社会受灾人口和直接经济损失三项指标建立的灰色关联度可以很好的表征干旱灾害的综合损失情况。适用性分析表明,基于MCI的干旱等级评估准确率普遍偏低,其中实际发生特旱21次,而MCI评估结果达115次。MCI指数对旱情等级评估偏重,这可能与四川省灌溉和供水能力逐渐提高有关。以灾害灰色关联度为参考序列,利用投影寻踪方法揭示MCI各个分量对实际旱灾的影响,找到最优投影方向,建立了MCI指数的修订系数。修订后的MCInew较MCI对旱灾等级的评估能力明显提高,其中对特旱评估的准确率从16.5%提高到62.5%,个例分析也表明MCInew与实际旱情的发生发展过程更加吻合。

Abstract: Based on the daily meteorological data and the drought disaster data of 156 stations in Sichuan Province during 1991-2013, the classification standard of the actual drought disasters was established with the method of grey correlation. The applicability of MCI in Sichuan Province was analyzed, and which was revised by using the method of projection pursuit. The results showed that the grey correlation degree of selecting three indexes of crop disaster area, social disaster population and direct economic loss could well describe the comprehensive loss of drought disaster. The applicability analysis showed that the accuracy rate of drought grade assessment based on MCI was generally low, of which the special drought occurred 21 times, and the MCI evaluation result reached 115. The inaccurate assessment of drought levels might be related to the gradual improvement of irrigation and water capacity in Sichuan province. Based on the reference sequence of the disaster grey correlation degree, the influence of each component of MCI on the actual drought was revealed by the projection pursuit method, and the optimal projection direction was found, and the revised coefficient of MCI indexes were established. Compared with MCI, the revised MCInew had significantly improved the evaluation ability of drought grade, and the accuracy of drought assessment increased from 16.5% to 62.5%. A case study also indicated that MCInew was more consistent with the actual drought occurrence and development.