This paper examines the relationship between economic growth and carbon dioxide (CO2) emissions, the dominant contributor to the greenhouse effect, in Turkey. For time series data for the period 1980-2009 taking into account the Environmental Kuznets Curve (EKC) was modeling. The results show that there are N-shaped relationship between CO2 emissions and economic growth, so these findings do not support EKC hypothesis with inverted-U shaped curve. At this point, because of the presence of a cubic polynomial, regression spline method is used as a different approach and a new model has been proposed. The model gives better results according to different model selection criteria.