Are National HFC Inventory Reports Accurate?

Hydrofluorocarbons (HFCs) were introduced as replacements for ozone depleting chlorinated gases due to their negligible ozone depletion potential. As a result, these potent greenhouse gases are now rapidly increasing in atmospheric mole fraction. However, at present, less than 50% of HFC emissions, as inferred from models combined with atmospheric measurements (top-down methods), can be accounted for by the annual national reports to the United Nations Framework Convention on Climate Change (UNFCCC). There are at least two possible reasons for the discrepancy. Firstly, significant emissions could be originating from countries not required to report to the UNFCCC (“non-Annex 1” countries). Secondly, emissions reports themselves may be subject to inaccuracies. For example the HFC emission factors used in the ‘bottom-up’ calculation of emissions tend to be technology-specific (refrigeration, air conditioning etc.), but not tuned to the properties of individual HFCs. To provide a new top-down perspective, we inferred emissions using high frequency HFC measurements from the Advanced Global Atmospheric Gases Experiment (AGAGE) and the National Institute for Environmental Studies (NIES) networks. Global and regional emissions information was inferred from these measurements using a coupled Eulerian and Lagrangian system, based on NCAR’s MOZART model and the UK Met Office NAME model. Uncertainties in this measurement and modelling framework were investigated using a hierarchical Bayesian inverse method. Global and regional emissions estimates for five of the major HFCs (HFC-134a, HFC-125, HFC-143a, HFC-32, HFC-152a) from 2004-2012 are presented. It was found that, when aggregated, the top-down estimates from Annex 1 countries agreed remarkably well with the reported emissions, suggesting the non-Annex 1 emissions make up the difference with the top-down global estimate. However, when these HFC species are viewed individually we find that emissions of HFC-134a are over-reported, with all other species being under-reported. Therefore, the approximate agreement between top-down and bottom-up aggregated Annex 1 emissions appears to be the result of a fortuitous cancellation of errors in the individual HFC emissions reports.