Forecasting CO2 Emissions Due To Electricity Power Generation In Bahrain

Abstract

Global warming is one of the biggest challenges among the leaders and scientists from developed and developing countries. Rapid industrialization and urbanization have given the boost to the amount of greenhouse gases’ emission. Carbon dioxide (CO2) is a significant of greenhouse gases and is the major contributing factor for global warming. CO2 emissions concentrations in the atmosphere have increased by 47% over the past 170 years due to human activities. As per Doha amendment of the Kyoto protocol in 2012, the target for maximum CO2 emission per capita for Bahrain was set to 20.96 metric ton for 2020. However, the current amount of CO2 emission per capita is 21.64 metric ton as of 2019. This research has applied multiple methods such as neural network time series nonlinear autoregressive, Gaussian Process Regression and Holt’s methods for CO2 emission forecasting. It attempts to forecast the CO2 emission of Bahrain. These methods are evaluated for performance. Neural network model has the RMSE of merely 0.206 while the GPR-RQ model has RMSE of 1.0171 and Holt’s method has RMSE of 1.4096. Therefore, it can be concluded that neural network time series nonlinear autoregressive model has performed better for forecasting the CO2 emission of Bahrain.