热带地理 ›› 2016, Vol. 36 ›› Issue (5): 869-877.doi: 10.13284/j.cnki.rddl.002829

• 论文 • 上一篇    下一篇

湖南月蒸发量的时空变化特征及其与月平均气温的相关性

陈阿娇1,贺新光1,曹思沁2,章新平1   

  1. (1.湖南师范大学 资源与环境科学学院,长沙 410081;2.湖南省气象局,长沙 410007)
  • 收稿日期:2015-11-17 出版日期:2016-09-05 发布日期:2016-09-05
  • 通讯作者: 贺新光(1973―),男,湖南人,教授,博士,主要研究方向为水文气象、地下水数值模拟预测,(E-mail)xghe@hunnu.edu.cn
  • 作者简介:陈阿娇(1991―),女,四川人,硕士研究生,主要从事水文气象研究,(E-mail)ajchen0807@163.com
  • 基金资助:
    国家自然科学基金项目(41472238);湖南省教育厅资助科研项目(14A097)

Spatiotemporal Variability of Monthly Evaporation and Correlation between Monthly Evaporation and Monthly Mean Temperature in Hunan Province

CHEN Ajiao1,HE Xinguang1,CAO Siqin2,ZHANG Xinping1   

  1. (1.School of Resources and Environmental Science,Hunan Normal University,Changsha 410081,China;2.Hunan Meteorological Administration,Changsha 410007,China)
  • Received:2015-11-17 Online:2016-09-05 Published:2016-09-05

摘要: 运用经验正交函数(EOF)和交叉小波变换法,分析了湖南省1981―2013 年月蒸发量的时空分布特征及其与月平均气温之间的非线性交互关系。月蒸发量的EOF 分解表明:前3 个空间模态的累计贡献率达77.9%。其中EOF1(63.4%)揭示了湖南省月蒸发量最主要的时空分布特征呈东西反向型,体现出月蒸发量与高程之间存在一定的负相关性;EOF2(8.5%)具有正负位相频繁交替的时间系数,反映了更为复杂的时空变化特征;EOF3(6.0%)的值自西北向东南呈“+、-、+”分布。然后,分别选用洪江、桑植和新晃站作为3 个空间模态的代表站点,并计算出各站点的月蒸发量与月平均气温之间的交叉小波谱和小波一致性,结果表明:两者之间的相关性在8~16个月的时间尺度下最强,且其相位角均为0°,说明气温对蒸发量的影响在该时间尺度下最为显著且不存在时滞效应,但在<8 个月和>16 个月的时间尺度下,强相关性的出现频率和分布以及相位角的方向沿时间轴均呈现出较大的差异性。

关键词: 月蒸发量, 月平均气温, 经验正交函数, 交叉小波变换, 相关性

Abstract: The spatiotemporal distribution of monthly evaporation and the nonlinear interaction between monthly evaporation and monthly mean temperature in the period 1981-2013 in Hunan Province are analyzed by applying the empirical orthogonal function (EOF) and cross wavelet transform.The EOF decomposition of the monthly evaporation shows that the first three EOF modes can account for 77.9% of total variability.The EOF1 (63.4%) reveals that the main spatiotemporal distribution of monthly evaporation in Hunan province is the type of East-West contrary, which reflects that there exists a certain negative correlation between the monthly evaporation and elevation.The EOF2 (8.5%) is of a frequent positive and negative alternation in phase of the time coefficient, which reflects a more complex spatiotemporal change of the monthly evaporation.The value of EOF3(6.0%) shows a distribution as ‘+、-、+’.Then, the cross wavelet and wavelet coherence between monthly evaporation and monthly mean temperature are computed by respectively selecting Hongjiang, Sangzhi and Xinhuang as the representative station of the first three EOF modes. The results show that at the time scale of 8-16 months, the most significant correlation between monthly evaporation and monthly mean temperature occurs and all their phase angles are zero, which illustrates that the temperature strongly effects on the evaporation and there exists no lag effect at these time scales. But at the time scales of less than 8 months or more than 16 months, the occurrence rate and distribution of strong correlation and phase angles are obviously different along time axis.

Key words: monthly evaporation, monthly mean temperature, empirical orthogonal function, cross wavelet, transform, correlation