Spatialization of Chinese R-410A emissions from the room air-conditioning sector

Hydrofluorocarbons (HFCs) are strong greenhouse gases and regulated by the Montreal Protocol as substitutes of ozone depletion substances. Currently, Chinese HFC emissions keep increasing, and the inventory is only on a national or city level. A high-resolution gridded HFC emission inventory is needed to develop HFC reduction policy and phase-down schedule. We developed a method by integrating point sources with longitude and latitude information and area sources using the proxy factor to explore the distribution of R-410A [a mixture of HFC-32 (CH 2 F 2 ) and HFC-125 (C 2 HF 5 )] emissions from the room air-conditioning sector on a 10 × 10 km 2 grid scale. Variety of regression models (including the principal component analysis, multiple linear regressions, stepwise regressions, and linear regression), analysis scale (national level and provincial level), and data dimensions (the proxy factor and unit-area value) were tested. The gross domestic product was found as the optimal proxy factor and used to spatialize R-410A emissions at a high-resolution scale. Compared to the national-level analysis, model evaluation parameters were largely improved for the provincial-level regression analysis, including root-mean-square error (from 20.96 to 11.35), normalized mean bias (from 0.16 to − 0.01), normalized mean error (from 0.45 to 0.20), mean absolute error (from 11.27 to 4.97), correlation coefficient (from 0.91 to 0.97), and relative error (from 39% to 76%), suggesting a better performance for the provincial-level analysis. This study provides a cost-effective method to establish fine-resolution HFC inventory. Meanwhile, high-resolution emissions grid data could be further applied to implement site-specific management of low-carbon development.