新疆玛纳斯河径流过程的非线性特征

结合玛纳斯河肯斯瓦特水文站1954-2007年的逐月径流及流域内气候资料,利用小波多尺度分析、混沌理论与周期性叠加趋势模型研究了玛纳斯河径流的非线性特征及影响因素。结论如下:①玛纳斯河径流变化的周期分别在15、18和22 a,并在18 a处最为显著;②玛纳斯河径流变化呈现复杂的非线性特征,其关联维为3.317 8,是非整数,这意味着其径流变化具有分形和混沌特性;③根据周期性叠加趋势模型,2007年后的7 a内玛纳斯河径流仍将保持上升趋势;④玛纳斯河径流与NAO关系密切,在夏季表现为负相关且相关性最强,其次在冬季为较强的正相关,在全年关联性最差,并且两者存在多尺度的复杂相关性;⑤NAO通过气流循环影响该流域的气候变化进而作用于径流。

Nonlinear Characteristics of Runoff Processes of the Manas River in Xinjiang

Based on the monthly stream discharge data of Kenswat Station and the climate data of ten meteorological stations from 1954 to 2007 in the Manas River Basin, the paper studied the nonlinear characteristics of the Manas River runoff and affecting factors by using methods of wavelet transform, chaotic theory and the periodic trend superposition model. The results show that: 1) The primary periods of the annual runoff are respectively 15,18 and 22 years, but are most significant at just 18 years; 2) the annual runoff change of the Manas River presents a complex nonlinearity, and the correlative dimension is 3.3178 which is non-integral, indicating that the variation process of the annual runoff has fractal and chaotic characteristic; 3) based on the periodic trend superposition model,the annual runoff will keep an increasing trend in the coming seven years after 2007; 4) the NAO has the best positive relationships with the runoff in winter, while in summer the relationships are negative and in annual the worst; and 5) the NAO has strong impacted on the climate variations which are caused by the atmospheric circulation in the Manas River Basin, then the runoff of the Manas River is affected.