欢迎访问《中国农学通报》,

中国农学通报 ›› 2015, Vol. 31 ›› Issue (34): 257-263.doi: 10.11924/j.issn.1000-6850.casb14100055

所属专题: 农业气象

• 资源 环境 生态 土壤 气象 • 上一篇    下一篇

西安市降雨特征的多尺度分析

沈姣姣,巨晓璇,高红燕,马 磊   

  1. 陕西省气象服务中心,陕西省气象服务中心,陕西省气象服务中心,陕西省气象服务中心
  • 收稿日期:2014-10-18 修回日期:2015-11-23 接受日期:2015-03-23 出版日期:2015-12-17 发布日期:2015-12-17
  • 通讯作者: 沈姣姣
  • 基金资助:
    2013 年陕西省气象服务中心业务服务类课题(2013-05)

Analysis on the precipitation characteristics in multiple scales in Xi

  • Received:2014-10-18 Revised:2015-11-23 Accepted:2015-03-23 Online:2015-12-17 Published:2015-12-17

摘要: 为对西安市不同等级降雨特征及未来演变趋势进行研究,采用线性回归、Mann-Kendall 检验、R/S 分析和ARIMA模型等方法分析1951—2013 年降雨量(日)资料。结果表明:(1)小雨、中雨、大雨及以上和总降雨量均表现出减少趋势,倾向率分别为-7.3、-2.6、-3.0、-13.6 mm/10 a,多在20 世纪60—70 年代出现减少的突变(除中雨外)。(2)小雨和总降雨日表现出减少趋势,倾向率分别为-3.3、-3.6 d/10 a,均在20 世纪80 年代后期出现减少的突变,90 年代后期后减少趋势达到显著性水平。大雨及以上等级降雨日表现出阶段性增加和减少的特点。(3)R/S 分析法和Hurst 指数试验(包括20 年试验和累加试验)分析均表明,除中雨外,未来西安市小雨、大雨及以上和总降雨量(日)将持续减少。未来西安市不同等级降雨量(日)继续减少的持续性强度为小雨量(日)>总降雨量(日)>大雨量(日),Hurst 指数序列图可用于突变点或显著变化点的检测。ARIMA模型在短期内预测精度仍较高[降雨量(日)的预报相对误差低于20%],可以用于降雨量(日)的气候预测。

关键词: 挥发油, 挥发油, 朱砂叶螨, 气质联用(GC-MS), 半数致死浓度, 川芎

Abstract: In order to study the precipitation characteristics and future trend in Xi’an, the data of different grades of precipitation and the days of precipitation in Xi’an from 1951 to 2013 were analyzed with linear regression, Mann- Kendall detection, R/S analysis and ARIMA model. The conclusions are as follows: (1) different grades of precipitation, including light rain,moderate rain,heavy rain and the above, and the total rain all showed a decreasing trend from 1951 to 2013, with a decreasing rate of - 7.3, - 2.6, - 3.0 and -13.6 mm/10 a respectively; except for moderate rain, different grades of precipitation and the total precipitation reduced suddenly in the late nineteen sixties to seventies; (2) the days of light rain and total rain showed a decreasing trend, with a decreasing rate of -3.3 and -3.6 d/10 a respectively, and reduced suddenly in the late nineteen eighties, especially reduced at 0.05 significant level after the late nineteen nineties; the days of heavy rain and the above appeared periodical increase and decrease; (3) in the future, the precipitation and the days of light rain, heavy rain and the above, and the total rain will continue to decrease; the persistent strength will perform as light rain>the total rain>heavy rain and the above, but the moderate rain will increase. Hurst Index experiment can be used to detect the moment of mutation or significant change. The relative prediction error of future annual precipitation and precipitation days by ARIMA model is less than 20%, so it can be used for the short-term precipitation prediction.

中图分类号: