Chaotic time series analysis of meteorological parameters in some selected stations in Nigeria

Abstract

One of the primary responsibilities of any meteorological and hydrological services is to provide information on weather warnings and forecasts to the general public for necessary precaution and prevention. This necessitated a better understanding of the underlying dynamics in weather parameters for modelling, prediction and control. This research examines chaos in the trend of the meteorological parameters such as air temperature, relative humidity and wind speed in eight meteorological stations in Nigeria using nonlinear time series analysis approach. The meteorological data were obtained and subjected to various components of nonlinear analyses which includes, time series, phase-space reconstruction, actual mutual information (AMI), false nearest neighbours (FNNs), and Lyapunov exponent. The delay time (τ) and embedded dimension (m) are estimated to be 10 and m≥5≥9 produced from actual mutual information (AMI) and false nearest neighbours (FNNs) respectively and are used to obtain the ultimate possible attractor reconstruction. A significant chaotic quantifier, the Lyapunov exponent was also estimated for all the locations. Our result shows positive largest Lyapunov exponent was evident with the parameters across the eight stations. The largest Lyapunov exponent was found to be within the range 0.0034 - 0.1200 for all the parameters; which indicates strong presence of chaos. The results therefore provide a good source of information for constructing weather prediction model.