FRACTAL APPROACH TO ANALYSE THE INSTRUMENTAL TIME SERIES OF TEMPERATURE OBSERVATIONS
The fractal approach allows to extend the knowledge about the nonlinear dynamics evolution caused by the growth of capacity and adequacy of information about regularities, uncertainties and abnormalities. The paper applies the fractal approach to analyze the instrumental time series observing the air temperature of the land surface during 1902-2012 years. An annual warming-cooling cycle is considered as a unit, so the peculiarities of the viewpoint on the temperature dynamics are connected with the analysis of the time series profiles. The corresponding fractal model – described the annual warming-cooling cycle dynamics - is adapted to the considered subject. Then it becomes possible to study comparatively the averaged, extremely and amplitude-frequency characteristics of the annual warming-cooling cycles. The observations – completed in the forty meteorological stations – are used in the six quasi-homogeneous climate regions. The individual and comparative study of the local climate systems exhibits both over-regional dominant tendencies and local uncertainties that enable to learn the local climate dynamics in more details.