Preferential flow is a common flow pattern of infiltrated rain and irrigated water in unsaturated soils,and characterizing preferential flow is always a hot topic of soil hydrology researches.In this research,the wavelet packet entropy,including the Shannon entropy and logarithmic energy entropy,was used to characterizing the preferential flow heterogeneity under different infiltration conditions.The research results were compared with those characterized by matrix entropy and fractal characteristic parameter to show the feasibility of applying wavelet packet analysis to preferential soil water flow.The results show that:1) the wavelet packet entropy is capable of characterizing the heterogeneity of preferential flow patterns;2) the heterogeneity of preferential flow increases firstly and then decreases as the infiltration amount increases;3) the heterogeneity of preferential flow increases steadily as the experimental scale increases within this research.%土壤优先流是降雨和灌溉水在土壤中常见的运动形式,但如何准确描述土壤优先流的非均匀特征一直都是土壤水文学界的研究难点和热点.采用小波包熵(Shannon信息熵和log能量熵)对不同入渗水量和不同试验尺度条件下观测到的优先流运动进行了度量和分析,并将研究结果与采用基质熵和分形特征参数等度量的分析结果进行了比较.研究结果表明:①小波包熵可成功度量优先流的非均匀特征;②土壤优先流的非均匀特征随入渗水量的增大先增大后减少;③在研究尺度范围内,随着试验尺度的增大,土壤优先流运动的非均匀程度增加.
展开▼