Abstract
Today, global high levels of industrialization, rapid population growth, unconsciously consumed energy resources and high energy demand have brought about an increase in the amount of carbon emissions. The increase in greenhouse gas emissions has come to a position that threatens economic development, sustainable growth, climate conditions, urbanization and most importantly economic power in Turkey as well as in the world. In the study, the Adaptive Neural Fuzzy Inference System (ANFIS) method, in which both fuzzy logic and artificial neural networks are integrated, is used for the estimation of carbon dioxide emissions. In the first part of the application, the monthly data set covering the years 1998-2018 for the explanatory variables of Gross Domestic Product (GDP), population, exports, and imports and retrospective 2019-2020 carbon emissions (R2=0.964) are estimated. In the second part, using the data obtained in the first part, the variables (t-12, t-11, t-10, mean-t-12, mean-t-11, mean-t-10) for monthly carbon emission values (t) are used. Forecasting success (R2=0.99) is estimated for 2021, 2022 and 2023. In this context, the findings of the study are evaluated within the scope of the Kyoto Protocol and the Paris Climate Agreement. It is expected that the results obtained would help decision makers in their projects and plans to reduce carbon emissions.