Using annual mean air temperature,precipitation,and evaporation data extracted from 58 meteorological stations in the Liaohe River Basin and the runoff data from Tieling hydrological station from 1956to 2009,we explored relationships between the mentioned climate factors and runoff at multiple time scales, and their periodic fluctuation characters based on wavelet analysis,wavelets cross correlation and partial correlation analysis.The results show that:(1)the main periods of runoff are around 10 years or 36 years; (2)calibrated by partial correlation analysis,the fluctuation cycles of temperature and runoff have positive correlation at all the studied time scales(from 1 to 54years);(3)the correlation between precipitation and runoff is significantly negative;(4)there is no significant correlation between evaporation and runoff.Therefore, temperature and precipitation are considered as the main driving forces on runoff change in the Liaohe River Basin. This study has improved methodological framework on runoff change researches based on wavelet analysis,and pointed out the shortcoming of previous studies by introducing the partial correlation method.These results also provide important information for water resource management in the Liaohe River Basin.