长江中下游地区20世纪60年代以来气温变化的标度特征

Scaling behavior of atmospheric temperature change since the 1960s in the middle and lower reaches of the Yangtze River

标度特征广泛存在于气候系统中,是气候系统内部各组分在不同时空尺度上相互作用的结果.本文基于长江中下游59个气象站点近43年来的日均温资料,分析气温变化的主要特点,采用多重分形消除趋势波动分析(MF-DFA)方法,研究其日均温序列、日均温增幅及增向序列的标度特征,并探讨其气温变化的多重分形特征.其中,增幅序列主要反映气候系统的非线性机制,增向序列则反映气候系统的线性机制.结果表明,长江中下游地区目前正处于近40多年来气温最高的阶段.该地区的日均温序列、线性过程以及非线性过程均具有多尺度的标度行为,在不同尺度上具有不同的标度特征,并且3个序列都在10 d左右存在一个交叉点.此外,长江中下游日均温变化具有自相似的多重分形特征,小波动的影响在日均温变化中占优势,这也是日均温序列具有长程相关性的重要原因之一.

Scale-invariant structure broadly exists in the climate system, which is the output of the components in the system interacting over a wide range of time and space scales. Based on daily temperature data of 59 meteorological stations over the Middle and Lower Reaches of the Yangtze River from 1960 to 2002, this article analyzes the characteristics of the temperature change. Then the paper investigates the scaling behaviors of atmospheric temperature, its magnitude and sign series, and explores the multifractal properties of temperature change by applying Multifractal Detrended Fluctuation Analysis （MF-DFA）. Sign series carry information regarding the linear process while magnitude series are mainly related to the non-linear properties. The results reveal that the temperature of the Middle and Lower Reaches of the Yangtze River is now the highest ever since 1960s. There exists multi-scale characteristics in the temperature series, the linear process and the non-linear process, and all the processes have crossovers of 10 days, which reflect different physical mechanisms. Moreover, the temperature series in the Middle and Lower Reaches of the Yangtze River are a kind of self-similar multifraetality, and the small fluctuations prevail in the change of daily temperature, which is one of the major reasons for the long-term correlations in the temperature series.