Atmospheric observations and atmospheric observation-based global emission estimates are presented for the five high molecular weight perfluorocarbons (PFCs): decafluorobutane (C₄F₁₀), dodecafluoropentane (C₅F₁₂), tetradecafluorohexane (C₆F₁₄), hexadecafluoroheptane (C₇F₁₆) and octadecafluorooctane (C₈F₁₈). Their atmospheric histories are based on measurements of 36 Northern Hemisphere and 46 Southern Hemisphere archived air samples, collected between 1973 and 2011, using two of the "Medusa" cryogenic preconcentration gas chromatography-mass spectrometry instruments, which are part of the Advanced Global Atmospheric Gases Experiment (AGAGE). A new calibration scale was prepared for each PFC, with estimated accuracies of 6.8% for C₄F₁₀, 7.8% for C₅F₁₂, 4.0% for C₆F₁₄, 6.6% for C₇F₁₆ and 7.9% for C₈F₁₈. Based on our observations, the 2011 globally averaged dry air mole fractions of these high molecular weight PFCs are: 0.17 parts-per-trillion (ppt, i.e., parts per 1012) for C₄F₁₀, 0.12 ppt for C₅F₁₂, 0.27 ppt for C₆F₁₄, 0.12 ppt for C₇F₁₆ and 0.09 ppt for C₈F₁₈.

Newly measured infrared absorption spectra are presented for C₇F₁₆ and C₈F₁₈, and using these, their radiative efficiencies and global warming potentials (GWPs) are estimated. We find that the radiative eciency of C₈F₁₈ at 0.57 Wm^-2ppb^-1 is similar to that of trifluoromethyl sulfur pentafluoride's (SF5CF3), which has the highest radiative efficiency of any known atmospheric species (Forster et al., 2007). Using their radiative efficiencies, the 2011 observed globally averaged atmospheric mole fractions of the above five high molecular weight PFCs combine to contribute a global average radiative forcing of 0.35mWm^-2, which is 6% of the total anthropogenic PFC radiative forcing (Montzka et al., 2011; Oram et al., 2012).

Global emissions for C₄F₁₀, C₅F₁₂, C₆F₁₄, C₇F₁₆ and C₈F₁₈ were estimated from the observations using a 3-dimensional chemical transport model and a Bayseian inverse method that included a constraint on the annual growth rate of their emissions, consistent with the knowledge of the relevant industries emitting them (Rigby et al., 2011). The derived so-called "top-down" emission estimates show that global emission rates were largest in the 1980s and 1990s for C₄F₁₀ and C₅F₁₂ and in the 1990s for C₆F₁₄, C₇F₁₆ and C₈F₁₈. After a subsequent decline, emission rates have remained relatively stable, within 20% year-to-year variation, for the last 5 years. Using their calculated 100-year time horizon GWPs, the high molecular weight perfluorocarbons studied here contributed up to 15.4% of the total PFC emissions expressed in carbon dioxide (CO2)-equivalents in 1997 and 6% of the total PFC emissions in 2009.

Furthermore, we compare our atmospheric observation-based global emissions to the available so-called "bottom-up" inventories, which are based on production information and end usage. Bottom-up emission estimates are available from the Emission Database for Global Atmospheric Research version 4.2 (EDGARv4.2) for C₄F₁₀, C₅F₁₂, C₆F₁₄ and C₇F₁₆, and emission inventories of C4F10, C5F12 and C6F14 are also reported to the United Nations' Framework Convention on Climate Change (UNFCCC) by Annex 1 countries that have ratied the Kyoto Protocol (European Commission, Joint Research Centre (JRC)/Netherlands Environmental Assessment Agency (PBL), 2009; United National Framework Convention on Climate Change Secretariat, 2011). The atmospheric observation-based emission estimates are 20 times larger than EDGARv4.2 for C₄F₁₀ and over three orders of magnitude larger for C₅F₁₂. In contrast, the top-down emission estimates for C₆F₁₄ largely agree with the bottom-up estimates from EDGARv4.2. Moreover, the top-down C₇F₁₆ emission estimates are comparable to those of EDGARv4.2 at their peak in the 1990s, albeit with signicant underestimation by EDGARv4.2 for the other time periods. There are no bottom-up emission estimates for C₈F₁₈, thus the emission rates reported here are the first for this gas. In general, the emission inventories for C₄F₁₀, C₅F₁₂ and C₆F₁₄ reported to the UNFCCC are five to ten times lower than those estimated in this study from observations. This underreporting to the UNFCCC may be due to only Annex 1 countries reporting inventories and also that some of these countries report a total PFC mixture in CO2-equivalents, instead of individual PFC emissions rates.