Due to tremendous increase in the level of carbon dioxide (CO2) emissions in the last several decades, a number of studies in the energy-growth-environment literature have attempted to identify the determinants of CO2 emissions. A major criticism related to the existing studies, we realize, is the selection of panel estimation techniques. Almost all studies use panel methods that ignore the issue of cross-sectional dependence even though countries in the panel are most likely heterogeneous and cross-sectionally dependent In addition, the majority of existing studies use aggregate energy consumption, and thus fail to identify the impacts of energy consumption by sources on the environment In order to fulfill the mentioned gaps in the literature, this empirical study analyzes the influence of the real income, renewable energy consumption, non-renewable energy consumption, trade openness and financial development on CO2 emissions in the EKC model for the top countries listed in the Renewable Energy Country Attractiveness Index by employing heterogeneous panel estimation techniques with cross-section dependence. We find that the analyzed variables become stationary at their first-differences by using the CADF and the CIPS unit root tests, and the analyzed variables are cointegrated by employing the LM bootstrap cointegration test By using the FMOLS and the DOLS, we also find that increases in renewable energy consumption, trade openness and financial development decrease carbon emissions while increases in non-renewable energy consumption contribute to the level of emissions, and the EKC hypothesis is supported for the top renewable energy countries. (C) 2016 Elsevier Ltd. All rights reserved.