Precipitation Complexity Measurement Using Multifractal Spectra Empirical Mode Decomposition Detrended Fluctuation Analysis

The stability of current methods of complexity measurement are generally Inefficient. In this study, multifractal spectra (MFS) analysis, which depends on empirical mode decomposition detrended fluctuation analysis (EMD–DFA), was used to measure the complexity of the monthly precipitation series from 1964 to 2013 (50 years) of 11 districts in Harbin, Heilongjiang Province, China. By comparing the anti-noise capability of MFS–EMD–DFA with that of conventional complexity measurement approaches, such as sample entropy, Lempel–Ziv complexity, and approx mate entropy, it was established that MFS–EMD–DFA has greater robustness in anti-noise jamming, and thus it could be applied more widely. The precipitation series complexity strength map of the 11 regions was drawn using a geographical information system. This study analyzed the correlation between precipitation and some meteorological factors and then ranked their strengths. The results showed that many meteorological factors have strong connections with the regional precipitation series in the study area. This study provided a solid foundation for further extraction of hydrological information in Harbin and proposed a new method for complexity analysis. The novel MFS–EMD–DFA approach could also be applied to the analysis of multifractal characteristics as well as complexity measurement in various other disciplines.