中国农业气象 ›› 2018, Vol. 39 ›› Issue (01): 1-8.doi: 10.3969/j.issn.1000-6362.2018.01.001

• 论文 •    下一篇

下限温度对北京樱花盛花始期模拟效果的影响

叶彩华,周孝煌,姜会飞,姜少杰,张玉莹,张子源,李超   

  1. 1.北京市气象服务中心,北京 100089;2.中国农业大学资源与环境学院,北京 100193;3.北京市气候中心,北京 100089
  • 收稿日期:2017-08-07 出版日期:2018-01-20 发布日期:2018-01-19
  • 作者简介:叶彩华(1968-),女,研究员,从事农业气候研究和应用服务。E-mail:ych681012@126.com
  • 基金资助:
    国家重点研发计划课题“主要粮食作物气象灾害发生规律及指标研究”(2017YFD0300401);北京市财政专项“2017年气象为农—都市生态休闲农业气象科技支撑服务”

Impacts of Different Base Temperature on Simulation Effect of Beginning Flowering Date of Prunus discoidea in Beijing

YE Cai-hua, ZHOU Xiao-huang, JIANG Hui-fei, JIANG Shao-jie, ZHANG Yu-ying, ZHANG Zi-yuan, LI Chao   

  1. 1.Beijing Meteorological Service Center, Beijing 100089, China; 2.College of Resource and Environment Sciences, China Agriculture University, Beijing 100193; 3. Beijing Regional Climate Center, Beijing 100089
  • Received:2017-08-07 Online:2018-01-20 Published:2018-01-19

摘要: 利用实际物候观测资料,通过人为设定不同生长发育下限温度值,比较不同指标对应发育阶段所需积温的稳定性和物候模拟准确率,是鉴定植物生长发育下限温度指标的常用统计方法。利用1998-2014年北京市海淀区玉渊潭公园樱花盛花始期观测资料,在设定下限温度0.0~5.0℃范围内,以0.1℃为步长,分别统计历年1月1日-樱花盛花始期的阶段活动积温和有效积温,对比分析不同温度指标对两种模型阶段积温稳定性和发育期模拟准确率的影响。结果表明:(1)阶段积温随设定下限温度指标而变化,且阶段活动积温值及其年际变化值均大于有效积温;(2)两种积温的标准差以活动积温的标准差较大,表明采用有效积温测算樱花盛花始期的误差小于活动积温测算结果;阶段有效积温在设定下限温度0.0~2.9℃范围内变异系数较小,3.0℃及其以上时活动积温的变异系数小于有效积温,活动积温离散程度更小;(3)不同设定下限温度值的始花期预测结果,均以2~3℃的准确率较高。对应设定下限温度0.0~3.2℃范围,有效积温模拟始花期基本准确率(误差小于3d)和准确率(误差2d以内)均大于活动积温模拟结果,但设定下限温度>3.8℃时结果相反。设定下限温度处于3.3~3.8℃时,有效积温模拟误差的基本准确率大于活动积温,误差小于2d准确率则低于活动积温。综上所述,无论活动积温还是有效积温,均以设定下限温度2.0~3.5℃的计算结果标准差和变异系数最小,准确率较高,在未进行生理测定的条件下,可将其作为樱花盛花始期的下限温度指标。

关键词: 积温, 生长假设, 基点温度, 变异系数, 模拟准确率

Abstract: It is a commonly used statistical method to determine the base temperature index (Tb) of plant growth and development by comparing stability and prediction accuracy of stage growing degree days (GDD) calculated by different artificial given base temperature index using actual observed phenomena data.In this paper, active growing degree days (GDDA) and effective growing degree days (GDDE) from Jan. 1st to the beginning flowing date of Prunus discoidea were calculated by different given Tb in the range of 0.0-5.0 ℃ with step of 0.1℃ using the observed data during 1998-2014 at Yuyuantan Park, Haidian District, Beijing. Effects of different given Tb on stability of two kinds GDD and simulated phenomena accuracy were also compared and analyzed. The results showed that: (1)stage GDD changes as Tb changes. GDDA and its inter-annual variability are both bigger than that of GDDE.(2)Both standard deviation (SD) and coefficient of variation (CV) of calculated GDDA or GDDE are smallest at the Tb range of 2.0-3.5℃, which could be the real Tb of Prunus discoidea determined by statistics under the condition without physiological experiment. SD of GDDA is bigger than that of GDDE, suggested that error of the beginning flowing date estimated by GDDE is smaller than that by GDDA. As an index of relative discrete degree, CV of GDDE calculation is smaller than that of GDDA in the Tb range of 0.0-2.9℃ but is bigger if Tb ≥3.0℃.(3)Comparing estimated beginning flowing dates of different given Tb, the accuracy is highest in the range of 2-3℃. In the range of 0.0-3.2℃, both basic accuracy (error<3days) and accuracy (error<2days) of the simulated date by GDDE is better than that by GDDA, but is opposite when Tb>3.8℃.In the range of 3.3-3.8℃, basic accuracy estimated by GDDE is better than that of GDDA, but accuracy is opposite.

Key words: Growing degree-days, Growth and developmental assumption, Base temperature, Coefficient of variation, Simulation accuracy