中国农业气象 ›› 2019, Vol. 40 ›› Issue (07): 422-434.doi: 10.3969/j.issn.1000-6362.2019.07.002

• 论文 • 上一篇    下一篇

近60年中国不同气候区极端温度事件的时空变化特征

张大任,郑静,范军亮,方智超,姬清元,袁叶子,刘文斐   

  1. 西北农林科技大学水利与建筑工程学院,杨凌 712100
  • 收稿日期:2018-12-03 出版日期:2019-07-20 发布日期:2019-07-08
  • 作者简介:张大任(1997-),本科,研究方向为气候变化下农业水文过程。E-mail:172876790@qq.com
  • 基金资助:
    国家自然科学基金(51879226;51509208)

Spatiotemporal Variations of Extreme Temperature Indices in Different Climatic Zones of China over the Past 60 Years

ZHANG Da-ren, ZHENG Jing, FAN Jun-liang, FANG Zhi-chao, JI Qing-yuan, YUAN Ye-zi, LIU Wen-fei   

  1. College of Water Resources and Architectural Engineering, Northwest A&F University, Yangling 712100, China
  • Received:2018-12-03 Online:2019-07-20 Published:2019-07-08

摘要: 利用1956-2015年中国200个气象站点的逐日最高和最低气温资料,采用国际ETCCDI推荐的10个极端温度指标,结合Mann-Kendall非参数检验、Sen’s slope斜率估计和Pettitt突变点检验等方法,研究了中国近60a极端温度事件的变化趋势及其在中国四大气候区的空间分布特征。结果表明:(1)整体上看,暖夜日数(TN90p)、暖昼日数(TX90p)分别以2.12和1.00d10a-1(P<0.01)的速度显著增加,而冷夜日数(TN10p)、冷昼日数(TX10p)以1.44和0.70d10a-1(P<0.01)的速度显著下降;基于阈值的极端温度事件指标中,霜冻日数(FD0)显著下降(2.84d10a-1,P<0.01),夏日日数(SU25)和热夜日数(TR20)均显著上升(分别为1.77和1.44d10a-1,P<0.01);基于持续期的指标中,暖日持续日数(WSDI)和作物生长期(GSL)表现为显著上升趋势,而冷日持续日数(CSDI)在全国范围内变化不显著。(2)冷指标的变暖幅度大于暖指标的变暖幅度;与最低气温相关联的夜指数(如TN10p、TN90p和FD0)的变暖幅度明显大于与最高气温相关联的昼指数(如TX10p、TX90p和WSDI)的变暖幅度。(3)四个气候区极端温度事件发生突变的时间段主要集中在20世纪80年代和90年代,80年代中期以后绝大部分指标的变化幅度相对以前更为显著。(4)在空间上,极端温度指标变化幅度在各区域间差异较大;高原山地气候区冷指数减小幅度最大,部分暖指数(如SU25、TR20和WSDI)在(亚)热带季风性气候区增长幅度最大。

关键词: 气候变化, 极端温度, 趋势分析, 突变检验

Abstract: Based on daily maximum and minimum temperature data during 1956-2015 obtained from 200 weather stations, ten extreme temperature indices recommended by ETCCDI were used to study the temporal trends of extreme temperature indices and their spatial distribution in four climatic zones of China, with the help of Mann-Kendall test, Sen’s slope estimator and Pettitt test. The results showed that, (1) warm nights (TN90p) and warm days (TX90p) tended to increase significantly in China over the past 60 years, with a rate of 2.12 and 1.00d10y-1 (P<0.01), but cold nights (TN10p) and cold days (TX10p) tended to decrease significantly, with a rate of 1.44 and 0.70d10y-1 (P<0.01), respectively. For threshold indices, the change rate of frost days (FD0) was -2.84d10y-1 (P<0.01), but the change rates of summer days (SU25) and tropical nights (TR20) were 1.77 and 1.44d10y-1 (P<0.01), respectively. For duration indices, warm spell duration index (WSPI) and growing season length (GSL) significantly increased, but cold spell duration index (CSDI) showed no significant trend during the period of 1956-2015. (2) The increasing rates of cold extremes were greater than those of warm extremes. Further, faster increases were observed for nightly indices related to the minimum temperature (e.g. TN10p, TN90p and FD0), compared with the daytime indices related to the maximum temperature (e.g. TX10p, TX90p and WSDI). (3) The abrupt change years of temperature extremes mainly occurred in the 1980s and 1990s. The change rates of most temrepature extreme indices were more significant after the mid-1980s than those before the mid-1980s. (4) The change rates of extreme temperature indices varied greatly among different climatic zones, with the greatest decreases in cold extreme indices in the mountain plateau zone and the greatest increases in warm extreme indices such as SU25, TR20 and WSDI in the subtropical monsoon zone.

Key words: Climate change, Extreme tempreature, Trend analysis, Abrupt change detection