关键词: 气候变化, 尺度效应, 小波分析, 青海省达日县

Abstract: Wavelet analysis, known as "Mathematical Microscope", has become a new mathematical approach in recent 20 years, being applied widely in many fields. Especially, with the significant advantage of characteristic of variable scale, it is suitable and effective enough to demonstrate the multi-scale traits of climate change, such as temporal series of temperature and precipitation. Accordingly, in this paper, Mexican Hat Wavelet is used to analyze the multi-scale characteristics of the climate change in Dari County, based on the temperature and precipitation of instrument data for 588 months during 1956~2004. Consequently, the periods and turning points of temperature and precipitation series in different time-scale are revealed, and the primary periods determined are 3 years or 4 years in a small scale and 16 years in a large scale. Meanwhile, the wavelet coefficients patterns exhibit a probable increasing trend in temperature and humidity in the following years. In the past 50 years, the temperature climbed to peak values for 3 times with the maximum at the beginning of the 21st century, and the precipitation is 2 times with the maximum in the early years of the 1980s. In general, the tendency of temperature and precipitation in Dari County is approximately consistent with that on the Tibetan Plateau and the source regions of the Yangtze and Yellow rivers except for some local lags in several periods. Changes in underdecadal scales are nested into complicated background in larger scales. The turning points varied along with the time-scale. According to the above analysis, we can conclude that there is a significant advantage of wavelet analysis in revealing the multi-scale characteristics, the primary period, and forecasting the trend of climate change in the following years.

Key words: climate change, scale effect, wavelet analysis, Dari County in Qinghai Province