Multifractal analysis of measure representation of flood/drought grade series in the Yangtze Delta, China, during the past millennium and their fractal model simulation

摘要

Based on reconstructed flood/drought grade series, IFS (iterated function systems) and RIFS [(recurrent) iterated function systems] models were used to reconstruct their measure representations. The goodness-of-fit of the simulation indicated good performance of IFS and RIFS models in simulation of the measure representations of the historical flood/drought grade series in the Yangtze Delta. Multifractal properties of the measures of the flood/drought grade series were analyzed using multifractal detrended fluctuation analysis (MF-DFA) technique. The DFA1 exponents of 0.6348 ± 0.013 and 0.7215 ± 0.013 indicated that the measures of the flood/drought grade series of the Yangtze Delta were characterized by long memory or persistence. Furthermore, the h(q) curves of the measure representations indicated that the measure representations of the flood/drought grade series in the Yangtze Delta were multifractal measures. The finding of this paper highlights possibility of prediction for future climatic changes based on historical records in the Yangtze Delta region.