【摘要】 降水是中国西北干旱区水资源的重要组成部分,利用合理方法有效认识降水的区域变化规律对指导农业发展尤为重要。基于新疆16个国际交换站1961—2012年降水距平时间序列,利用集合经验模态分解(EEMD)方法,分析了新疆降水变化趋势的多尺度特征,并对其空间差异进行了初步探讨。近50多年来,新疆降水量整体上呈现出非线性的显著增多趋势,且其变化存在明显的年际尺度(3 a和6 a准周期)和年代际尺度(10 a和31 a准周期);各周期分量方差贡献率显示年际变化在新疆降水变化中占据主导地位,重构的降水年际变化趋势能精细刻画原始降水序列在研究时期内的波动状况;降水年代际变化揭示了新疆降水在1987年前后气候模态有了显著转换,由原来降水以负相位为主的气候模态转向正相位显著的气候模态;此外,降水变化趋势和转折时间均具有明显的区域差异。EEMD方法有助于加深人们对新疆降水多尺度变化特征的认识,是一种适用于非线性、非平稳信号分析的有效方法。更多还原

【Abstract】 Precipitation is the most key factor for water resources in the arid region of northwestChina. Therefore, it is especially important for guiding agricultural development to understand theregional characteristics of precipitation with effective and reasonable methods. Based onprecipitation anomaly time series from 16 international exchange stations in Xinjiang during theperiod from 1961 to 2012, the multi-scale characteristics of precipitation variability were analyzedusing ensemble empirical mode decomposition(EEMD)method. Regional differences in variationtrend and change-point were also preliminarily discussed. In the last more than 50 years, theoverall precipitation in Xinjiang has exhibited a significant non-linear upward trend, and itschanges have obviously presented an inter-annual scale(quasi-3 and quasi-6-year)and inter-decadal scale(quasi-10 and quasi-31-year); The variance contribution rates of each componentsproved that the inter-annual change had a strong influence on the overall precipitation change inXinjiang, and the reconstructed inter-annual variation trend could describe the fluctuation state oforiginal precipitation anomaly during the study period. The reconstructed inter-decadal variabilityrevealed that the climate mode in Xinjiang had a significant transformation before and after 1987,namely the precipitation anomaly shifting from a negative phase to a positive one. Furthermore,there were obviously regional differences in the non-linear changes and change-points ofprecipitation. At the same time, the results also suggested that the EEMD method can effectivelyreveal variations in long-term precipitation at different time scales and can be used for the complexdiagnosis of non-linear and non-stationary signal changes.更多还原