湖北农业科学 ›› 2020, Vol. 59 ›› Issue (21): 177-183.doi: 10.14088/j.cnki.issn0439-8114.2020.21.039

• 信息工程 • 上一篇    下一篇

基于决策树算法的气象因子对油茶产量影响研究

黄超1,3, 廖玉芳2,3, 蒋元华1,3, 彭嘉栋1,3   

  1. 1.湖南省气候中心,长沙 410008;
    2.湖南省气象科学研究所,长沙 410008;
    3.气象防灾减灾湖南省重点实验室,长沙 410008
  • 收稿日期:2020-04-24 出版日期:2020-11-10 发布日期:2020-12-21
  • 通讯作者: 廖玉芳(1962-),女,湖南常德人,正研级高级工程师,主要从事气候和气候变化研究,(电子信箱)lyf_13975681873@163.com。
  • 作者简介:黄 超(1991-),男,湖南常德人,工程师,硕士,主要从事智能算法在气候变化上的应用研究,(电话)18874031231(电子信箱)376906821@qq.com;
  • 基金资助:
    湖南省科技重大专项(2018NK1030)

Research on the effect of meteorological factors on Camellia oleifera yield based on decision tree algorithm

HUANG Chao1,3, LIAO Yu-fang2,3, JIANG Yuan-hua1,3, PENG Jia-dong1,3   

  1. 1. Hunan Climate Center,Changsha 410008, China;
    2. Institute of Meteorological Sciences of Hunan Province, Changsha 410008,China;
    3. Hunan Key Laboratory of Meteorological Disaster Prevention and Reduction,Changsha 410008,China
  • Received:2020-04-24 Online:2020-11-10 Published:2020-12-21

摘要: 采用分类与回归树(CART)和卡方自动交叉检验(CHAID)两种决策树算法,基于不同物候期气象指标对2010—2016年湖南省24个油茶(Camellia oleifera Abel.)测产点的油茶产量进行分析。结果表明,两种决策树算法对于历史产量数据模拟的平均相对误差分别达8.80%、14.30%,趋势准确率分别为97.40%、92.20%;开花期的0 ℃以上积温和平均最高气温对油茶产量影响最大,在果实第一次膨大期、油脂转化和积累高峰期,气温日较差、平均最低气温和高温日数最重要。对提升油茶产业效益极具现实意义。

关键词: 油茶(Camellia oleifera Abel.), 气象因子, 决策树算法

Abstract: In this study, two kinds of decision tree algorithms, CART(Classification and regression tree) and CHAID(Chi-Square automatic interaction detection) were adopted to simulate the yield of Camellia oleifera of Hunan province based on meteorological factors of different phenological phase over from 2010 to 2016. The results showed that the average relative errors of CART and CHAID algorithms were 8.80% and 14.30% respectively, and the trend accuracy were 97.40% and 92.20% respectively. The accumulated temperature above 0 ℃ and the average maximum temperature at flowering stage had the greatest influence on Camellia oleifera yield. In the first fruit expansion period and the peak of oil transformation and accumulation, the diurnal temperature range, the mean minimum temperature and high temperature days were the most important. It is of great significance for improving the efficiency of Camellia oleifera industry chain.

Key words: Camellia oleifera Abel., meteorological factors, decision tree algorithm

中图分类号: