The performance of HF and DFT/B3LYP in the estimation of the radiative efficiencies of greenhouse gases

In this work, we use electronic structure calculations, at the RHF and DFT/B3LYP level of theory, combined with Pinnock's cloudy sky instantaneous radiative transfer model, to predict radiative efficiency for the whole set of 63 greenhouse gases included in the IPCC Fourth Assessment Report, this set included chlorofluorocarbons, bromofluorocarbons, hydrofluorocarbons, perfluorocarbons, hydrofluoroethers and hydrofluoropolyethers. In order to get better insights, the greenhouse gases included in the IPCC set are divided in four groups and two subsets, and the results of the statistical analysis for each of the groups are presented. We validate our predictive ability using the IPCC accepted values, and obtain an average 20% of the Mean Absolute Percentage Error at the B3LYP/6-31++G∗∗, after an ad hoc uniform scaling of the theoretical vibrational frequencies is carried out.