Nonlinear dynamics of monthly temperature data set in the Northwestern (Marmara region) Turkey

Abstract

Temperature is one of the essential quantities which affect the agriculture, economy, survival of animal and plant species. Turkey, located in the interconnection of Europe and Asia, contains various climate types. In this research, dynamics of temperature values over northeastern Turkey is investigated via nonlinear time series techniques. Monthly data set, through the period between 1960 and 2006, from various parts of the region is used to understand the underlying dynamics and its spatial distribution. Well-known phase space reconstruction method is used to calculate the maximal Lyapunov exponent for each data set which will indicate the chaotic behaviour. Along with the nonlinear time series analysis, statistical properties of the data are also determined to reveal any possible trends in the data. Calculated positive Lyapunov exponents indicate that the time series of average monthly temperature values have a possible chaotic underlying dynamics over the region.