Effect of Greenhouse Gas Emissions on the Predictability of Rainfall, Temperature and Solar Radiation in Port Harcourt, Nigeria

Abstract

This research applies the tools of statistical analysis and nonlinear dynamics on rainfall, average temperature and solar radiation records over Port Harcourt, Nigeria, so as to quantify the signatures of chaos and analyze the trend of predictability of the weather over the last three decades. Quantifying the degree of chaos and the predictability of a dynamical system like weather time series is carried out by computing the largest Lyapunov exponents and then taking the inverse to get the predictability of the system. Rosenstein's algorithm was deployed in evaluating the yearly values of the largest Lyapunov exponents and predictability for each data set and the results obtained were subjected to Correlation analysis, Mann-Kendall test and Sen's slope estimator so as to estimate the trends and relationship between the predictability of these parameters and the total CO 2 emissions recorded in the region over the three decade period. The results obtained show that the trends in the Lyapunov exponents and predictability of rainfall, average temperature and solar radiation in Port Harcourt are all significant; with the relationship between the predictabilities of the meteorological variables and the annual Total CO 2 emissions found to have a decreasing trend. This implies the prevalence of fluctuating and extreme weather as well as failure in the accurate short-term prediction of the weather over Port Harcourt in the near future as a result of greenhouse gas emissions from oil exploration and refining activities.