On the effectiveness of recession analysis methods for capturing the characteristic storage-discharge relation: An intercomparison study

Abstract. Storage-discharge (S-Q) relations are widely used to derive watershed properties and predict streamflow responses. These relations are often obtained using different recession analysis methods, which vary in recession period identification criteria and Q vs. −dQ/dt fitting scheme. Although previous studies have indicated that different recession analysis methods can result in significantly different S-Q relations, several challenges remain regarding the evaluation of relative effectiveness of these methods in obtaining the characteristic S-Q relation. Here we demonstrated these challenges and presented a new control setup based experimental approach to compare four recession analysis methods. Results indicated that irregular binning and event-based methods show superior performance at obtaining the characteristic S-Q relation and reconstructing streamflow, while lower envelope method performs the worst. Notably, accuracy of the methods is influenced by the extent of scatter in the ln(−dQ/dt) vs. ln(Q) plot. In addition, the derived S-Q relation was very sensitive to the criteria used for identifying recession periods. These results raise a warning sign against indiscriminate application of recession analysis methods and derived S-Q relations for watershed characterizations or hydrologic simulations. Thorough evaluation of representativeness of the derived S-Q relation should be performed before it is used for hydrologic analysis.