Abstract: Morlet wavelet analysis was used to reveal the periodical regulations at different time scales of the temperature time series of Northeast China in the last 100 years. Based on the monthly and annual temperature data at Harbin, Shenyang, Changchun, and Dalian weather stations from 1905 to 2005, the periodic oscillation and the point of abrupt change at different time scales in the temperature time series were discovered, and the main periods for each serial were determined. The research results show that the temperature of Northeast China had an increasing trend of 0.165/10a in the past 100 years, and there were periodical variations of multiple time scales, which displayed different traits at different phases. There were obvious periodic oscillations of 2-3 a, 8-12 a, 20-25 a and 45 a periods for the seasonal and annual temperature variations.

Key words: Northeastern China, last 100 years, temperature time series, wavelet analysis