Existence of fractal behaviour in ozone time series

Abstract

Fractal has received a wide attention and has been used in many areas such as meteorology, stock market and also in medical field. It can be viewed as an object which has similar appearance when viewed at different scales which are known as self-similar. Thus, the scale invariance and scaling properties of the time series may be explored using fractal techniques. The aim of this study is to investigate the presence of fractal behaviour in the ozone time series. The presence of fractal behaviour can lead to the possibility of implementing the fractal approach in order to examine the properties of a time series. In this study, the daily average of ozone concentration from six selected air monitoring stations with different types of backgrounds in Peninsular Malaysia are used. From the autocorrelation function (ACF) plot obtained for the six stations, the slow decay in ACF values indicates that self-similarity are present where it shows that the series is having the property of fractal behaviour. In this paper, the existence of fractal behaviour is investigated by using the power spectrum method and the empirical probability distribution function. The result shows that the daily average of ozone concentration exhibits fractal behaviour for all the six monitoring stations considered.