Detrended multifractal characterization of Indian rainfall records

Long term rainfall records in seven distinct temperature homogeneous regions of India have been analyzed using two different multifractal analysis techniques, namely the multifractal detrended fluctuation analysis (MFDFA) and multifractal detrended moving average (MFDMA) techniques. In all the series studied a multifractal pattern has been obtained from both the methods, though the degree of multifractality observed for any given series largely depends on the analysis methods. Emphasis is given on the calculation of the generalized Hurst exponent spectra and the singularity spectra for all the series studied. In order to explore the possible source(s) of the multifractality in the data, we also study a set of ten shuffled series corresponding to each original series. We find that the origin of multifractality as obtained from both the methods, is mainly due to the fat-tailed probability distribution of the records in the series, though the contribution of the long-range temporal correlation cannot be ignored in the rainfall records. (C) 2021 Elsevier Ltd. All rights reserved.