Multiple Spatial Scales Chaos Characterization in Runoff Series by the 0-1 Test Algorithm

Abstract

The 0-1 test algorithm is a new nonparametric method for detecting the presence of chaos in the nonlinear science disciplines. This test applies directly to the time series and does not depend on phase-space reconstruction. This thesis take logistic map as an example, the numerical results demonstrate the advantages of the method. Then, the runoff time series (2002-2009) collected by the six hydrologic stations (Tangnaihai, Lanzhou, Toudaoguai, Longkou, Sanmenxia and Huayuankou) of Yellow River, are selected to detect the existence of chaotic behaviour in different spatial scales. The asymptotic growth rates of Kc are 0.8751, 0.8985, 0.9783, 0.9793, 0.9848 and 0.9976, respectively. Multiple spatial scales runoff data of Yellow River Basin shows chaotic characteristic. The minimum value of Kc is in the upper reaches of the Yellow River (Tangnaihai). The maximum value of Kc is in the lower reaches of the Yellow River (Huayuankou). The spatial runoff process changes greatly in temporal and spatial scales. And drainage area is an important factor in causing chaos variation. The conclusion illustrates the feasibility of this method and provides scientific data for runoff prediction.

Keywords

• chaos
• runoff
• the asymptotic growth rate
• spatial scales