Abstract
In this study, Multifractal Detrended Fluctuation Analysis (MF-DFA) is applied to daily temperature time series (mean, maximum and minimum values) from 22 Greek meteorological stations with the purpose of examining firstly their scaling behavior and then checking if there are any differences in their multifractal characteristics. The results showed that the behavior is the same at almost all stations, i.e., time series are positive long-term correlated and their multifractal structure is insensitive to local fluctuations with large magnitude. Moreover, this study deals with the spatial distribution of the main characteristics of multifractal (singularity) spectrum: the dominant Hurst exponent, the width of the spectrum, the asymmetry and the truncation type of the spectrum. The spatial distributions are discussed in terms of possible effects from various climatic features. In general, local atmospheric circulation and weather conditions are found to affect the shape of the spectrum and the corresponding spatial distributions. Furthermore, the intercorrelation of the main multifractal spectrum parameters resulted in a well-defined group of stations sharing similar multifractal characteristics. The results indicate the usefulness of the non-linear analysis in climate research due to the complex interactions among the natural processes.
Highlights
Complex systems consist of many components that interact with each other in very complicated ways, which are determined by nonlinear laws
Analysis, characteristics the Seasonal and Trend decomposition using Loess (STL) decomposition method is applied to the temperature time series
Prior to Multifractal Detrended Fluctuation Analysis (MF-Detrended Fluctuation Analysis (DFA)) analysis, the STL decomposition method is applied to the temperature time
Summary
Complex systems consist of many components that interact with each other in very complicated ways, which are determined by nonlinear laws. A characteristic example of a complex system is the atmosphere and the natural processes that take place in it The result of such complex interaction among atmospheric processes is the fluctuation of the values of meteorological parameters in an almost random way at many scales in time and space. These nonlinear processes are described by nonlinear partial differential equations [1]. Researchers can analyze processes that obey nonlinearity and they are able to reveal some properties, which could not be detected by linear methods One such nonlinear method is Detrended Fluctuation Analysis (DFA), which was developed by [3].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
