Predicting the Risks of Greenhouse Gases at Raising Temperatures in Syria Using Artificial Intelligence Models

Abstract

The research aimed to study the general trend of the evolution of greenhouse gas emissions in Syria during the period (1993-2022) and to predict the risks of greenhouse gas emissions to rising temperatures during the period (2023-2030). The research adopted the descriptive analytical method in estimating the equations of the general time trend of greenhouse gas emissions and calculating the annual growth rate for each of them during the studied period based on the statistics of the World Bank. The amount of gas emissions during the studied period was multi-layered to suit the nature of the data. The neural network used in prediction consisted of three layers: the input layer, the processing layer, and the output layer. The results of the research showed that there is a general trend of increasing temperatures at a rate of 0.16% annually, which is within the internationally permitted limits according to the Paris Agreement in 2015. The amount emitted of nitrous oxide gas occupied the highest relative importance in terms of the effect on temperature rise 100%, followed by the amount emitted of methane gas 94.1%. In contrast, the percentage of carbon dioxide emissions did not exceed 4.3% in the proposed model. The results of the prediction using the neural network model showed that the average temperatures during the coming period (2023-2030) will reach their maximum value in the year (2024) with an average of (14.87) degrees Celsius, with a relatively increasing annual growth rate of 0.07%.