Abstract:
This study investigated the scaling characteristics of daily rainfall time series over India and their spatio-tem poral variability using Multifractal Detrended Fluctuation Analysis (MF-DFA) method. In the study, fine re solution gridded (1o
x1o
) dataset of daily rainfall for the period 1951–2016 was used for the analysis. The scaling
characterization using MF-DFA shows that rainfall data of most of the grid points (over 87%) display short term
persistence. The analysis of spatial variability of multifractal characteristics shows that the multifractal strength
is strongest in the western and central India while the strength is the lowest in the north east region. Further, the
evaluation of multifractal properties of rainfall time series of pre and post 1976/77 period of Pacific climate shift
shows that there is a clear reduction in the multifractal strength and complexity for the post 1976/77 with
contrasting behavior for the persistence. Finally, the association of daily rainfall with mean, maximum,
minimum temperature values and the diurnal temperature range (DTR) time series of the 1951–2016 period
were investigated using Multifractal Detrended Cross Correlation analysis (MF-DCCA). The nature and strength
of association between the two variables of rainfall and temperature differs with time scales and it was found
that the joint persistence of these variables lies between individual persistence property. There is a distinct
difference in the persistence cross-correlation properties at the Peninsular region and coastal belts when com pared with other regions in India and the difference is most perceptible in the Tmin-rainfall link
