欢迎访问《中国农学通报》,

中国农学通报 ›› 2020, Vol. 36 ›› Issue (22): 92-99.doi: 10.11924/j.issn.1000-6850.casb20190800540

• 资源·环境·生态·土壤·气象 • 上一篇    下一篇

基于主成分和聚类分析的柳州市森林火灾特征研究

蓝柳茹(), 刘蕾, 吴镇疆   

  1. 柳州市气象局,广西柳州 545001
  • 收稿日期:2019-08-14 修回日期:2019-11-07 出版日期:2020-08-05 发布日期:2020-07-22
  • 作者简介:蓝柳茹,女,1989年出生,广西忻城人,工程师,硕士,主要从事短期天气预报与气候变化研究。通信地址:545001 广西柳州市庆丰路5号 柳州市气象局,Tel:0772-2302211,E-mail: liuru_lan@126.com
  • 基金资助:
    广西区气象局气象科研计划项目“柳州市森林火险气象等级本地化研究与预报平台开发”(桂气科2019M15);2018年柳州气象局气象科研计划项目“柳州市森林火险气象等级本土化研究与预报平台开发”

Characteristics of Forest Fire in Liuzhou: Based on Principal Component Analysis and Clustering Analysis

Lan Liuru(), Liu Lei, Wu Zhenjiang   

  1. Liuzhou Meteorological Bureau, Liuzhou Guangxi 545001
  • Received:2019-08-14 Revised:2019-11-07 Online:2020-08-05 Published:2020-07-22

摘要:

为给柳州市预防森林火灾提供参考依据,根据柳州市2010—2017年的森林火灾资料,采用常规统计方法对柳州市森林火灾的时空动态和分布规律进行研究,利用主成分分析和聚类分析对柳州市森林火灾形势进行评价。结果表明:(1)柳州市森林火灾的发生年际波动显著,2017年林火发生次数最高,其次是2011年和2014年。林火主要集中在秋末、冬季和春初;(2)从空间分布看,森林火灾主要发生在中南部的鹿寨县、柳江区、柳城县,以鹿寨县最为集中,火场总面积及受灾森林面积也最大。(3)在火灾形势方面,按照严重程度将森林火灾分为8类,其中鹿寨县的火灾形势较为严峻,级别波动明显,其次是柳城县的火灾形势呈波动恶化态势。研究结果为建立柳州市森林火险气象等级预报提供了借鉴。

关键词: 森林火灾, 时空分布特征, 年际变化, 主成分分析, 聚类分析, 柳州市

Abstract:

The paper aims to provide references for prevention and forecast of forest fire in Liuzhou. Based on the forest fire data of Liuzhou from 2010 to 2017, we studied the spatio-temporal dynamic and distribution law of forest fire in Liuzhou by using conventional statistical methods, and evaluated the situation of forest fire by the methods of principal component analysis and cluster analysis. The results showed that: (1) the occurrence of forest fires in Liuzhou had an obvious interannual fluctuation, with the highest number of forest fire in 2017, followed by 2011 and 2014; forest fire mainly concentrated in late autumn, winter and early spring; (2) in terms of spatial distribution, forest fires mainly occurred in Luzhai, Liujiang and Liucheng in south central Liuzhou, and were the most in Luzhai, causing the largest total fire burned area and affected forest area; (3) in terms of fire situation, forest fires were divided into 8 categories according to the severity, among them, the fire situation in Luzhai was severe with obvious grade fluctuation, followed by the fire situation in Liucheng, which showed a worsening trend with fluctuation. The results provide a reference for establishing the meteorological level forecast of forest fire risk in Liuzhou.

Key words: forest fire, spatio-temporal distribution characteristic, interannual variation, principal component analysis, cluster analysis, Liuzhou

中图分类号: