基于小波消噪和符号动力学的参考作物腾发量变化的复杂性研究

Complexity of Reference Evapotranspiration Based on Wavelet De-noising and Symbolic Dynamics

提出了基于小波消噪的符号动力学途径研究参考作物腾发量(ET0)变化的复杂度,以辽宁省的锦州、北镇、义县、黑山4个站为例,对各气象因素(1957～2006年)及ET0序列进行小波消噪,运用Lempel-Ziv算法计算其复杂度。结果表明:整个锦州市由于噪声影响ET0的平均复杂度为9.25,较真实序列平均复杂度大;义县ET0变化比其他3个站复杂;从时间上分析,锦州4个站中有3个站(锦州、义县、黑山)近20年呈现下降趋势。从符号动力学的角度研究ET0系统的复杂性,为ET0研究提供了新的工具。

Abstract： Complexity reference evapotranspiration （ETo） was studied based on wavelet de-noising and symbolic dynamics. With Jinzhou, Beizhen, Yixian and Heishan four stations in Liaoning as application examples, wavelet de-noising was carried out for meteorological factors and ETo sequence attained from these four stations （1957-2006） and then the complexity could be calculated using Lempel-Ziv algorithm. The average complexity of ETo in Jinzhou as a whole was 9.25 because of noise influence, greater than its real average complexity; Yixian ETo variation was more complex than thai： of other three stations. In terms of temporal trend analysis, there was a decrease in complexity in three stations, namely, Jinzhou, Yixian and Heishan. In a word, the complexity research of ETo system from the perspective of symbolic dynamics provided a new tool for ETo research.