Journal of Arid Land


Baseflow, which represents the drainage of groundwater aquifers, is an essential component of runoff in hydrological basins. In the source region of the Yangtze River, the change of baseflow typically reflects the interactions between groundwater system and climatic factors in cold and arid areas. With modified Kalinen separation method, annual baseflow between 1957 and 2009 in this region was estimated and calculated. In comparison with the inner-annual variations of total streamflow, baseflow showed a weaker fluctuation. Before the 1980s, it was in a steady state; and after then, it demonstrated dramatic variations and large amplitudes. Based on the calculation results of baseflow, the real Morlet wavelet method was applied to reveal the periodical characteristics of baseflow as well as the precipitation and air temperature in the study area. It was found that annual baseflow has a 43-year trend as well as a 21-year period and a 7-year period. The 21-year period is most significant, with its wavelet coefficient having the largest fluctuation and amplitude. Summation of wavelet coefficients on these periods exhibits a similar change pattern with respect to that of annual baseflow. The summation curve takes a “W” shape, which means that the baseflow follows a four-stage sequence of descending–ascending–descending–ascending. As analyzed, the relationship among baseflow, precipitation and temperature is implied in the correlation between their normalized wavelet coefficients at different temporal scales. By the significant positive linear correlations both between precipitation and baseflow (correlation coefficient is 0.98) and between temperature and baseflow (correlation coefficient is 0.90) for the 43-year wavelet coefficients, it is suggested that the long-term increasing trends of precipitation and air temperature will lead to an increasing trend of baseflow. For wavelet coefficients of 21-year and 7-year periods, the positive linear correlation between precipitation and baseflow is significant. However, the correlation between air temperature and baseflow is not so evident, especially for the 21-year period. As a conclusion, correlation analysis with normalized wavelet coefficients showed that the change of annual baseflow was contributed mostly by the change of precipitation and secondly by the change of temperature.


baseflow; climate; Kalinen separation method; Morlet wavelet; Yangtze River

First Page


Last Page