|
|
|
| Suitability analysis of automatic baseflow separation methods in typical watersheds of water-wind erosion crisscross region on the Loess Plateau, China |
| Lei Yongnan1,2,Zhang Xiaoping1,3,Zhang Jianjun1,2, Liu Erjia1,2, Zhang Qingyin1,3, Chen Ni1,3 |
| 1.The State Key Laboratory of Soil Erosion and Dryland Farming in Loess Plateau, Institute of Soil and Water Conservation, Chinese Academy of Sciences and Ministry of Water Resources, 712100, Yangling, Shaanxi; 2. Graduate University of Chinese Academy of Sciences, 100049, Beijing; 3. Institute of Soil and Water Conservation, Northwest A&F University, 712100, Yangling, Shaanxi: China |
|
|
|
|
Abstract The water-wind erosion crisscross region on the Loess Plateau is the main source of the coarse sediments of the Yellow River, where eco-environment is very fragile. The baseflow is of great significance to maintain sustainable development of the river ecosystem in this region. In order to test the applicability of automated baseflow separation methods on the water-wind erosion crisscross region on the Loess Plateau, several commonly and widely used methods including minimum smoothing method, HYSEP and recursive digital filter were tested by the runoff data of Kuyehe Watershed. The results show that there were significant difference of baseflow indices of eight methods with the maximum 0.6515 and minimum 0.3305. The statistical characteristics of annual baseflow indices that were used Chapman-Maxwell filtering method and Eckhardt filtering method to estimate are the best with the mean values 0.3868 and 0.3305, respectively, and the standard deviation 0.0428 and 0.0461, respectively. The baseflow discharge curves are smooth, which accords with realistic base flow condition. The verification results show that Nash-Sutcliffe efficiency coefficients were 0.691 and 0.718, respectively, and the average relative errors were 9.83% and 5.05%, respectively. Therefore Chapman-Maxwell filtering method and Eckhardt filtering method separate most objective and stable base flow, which could be used as optimal base flow separation method in this study area. The mean baseflow estimated by Chapman-Maxwell filtering method and Eckhardt filtering method showed a significant negative trend in the near 50 years and the reduction of baseflow account for 41% of the total runoff.
|
|
Received: 25 June 2011
|
|
|
|
|
|
|
