基于小波变换的流量变化特征分析——以长江大通站为例

Analysis of Flow Change Characteristics Based on Wavelet Transform——Take Datong Yangtze River Station as an example

本文以大通站1900-1999年的年平均径流量资料为基础,运用Morlet复小波变换和Mann-Kendall检验法来对大通站的年径流变化特征及趋势变化做初步探讨,以期更加深入了解其流量的变化特征。总体上,大通站径流量1900~2000年的演化过程中存在着6-11年小尺度、18-26年中尺度、29-35年和48-61年大尺度的4类尺度的周期变化规律,除6-11年尺度外,其余三个尺度在整个时间序列上表现得非常稳定,丰枯交替过程比较有规律性。该结果表明,用小波变换分析年平均流量资料序列,研究流量在时域-频域中的周期分布及变化时具有比较明显的优势,能很好揭示水文序列周期变化的特征。

Based on the data of annual mean runoff from 1900 to 1999 in Datong Station,this paper uses Morlet complex wavelet transform and Mann-Kendall test to discuss the changes in annual runoff characteristics and trends of Datong Station in order to understand its flow more deeply. In general,there are four types of cyclical changes: 29-35-year and 48-61-year large-scale,18-26-year mesoscale,and the 6-11 year small-scale during the period from 1900 to 2000. Except for the scale of 6-11 years,the other three scales are very stable in the entire time series,and the abundance and alternation processes are more regular. The results show that using the wavelet transform to analyze the annual average flow data series and study the flow distribution in the time domain and frequency domain has obvious advantages in the distribution and change,which can well reveal the characteristics of periodic changes in hydrological sequences.