Comparison of Top-Down and Bottom-Up Methods for Compiling Motor-Vehicle Emission Inventories: A Case Study of Kaifeng, China

Urban vehicle emission inventories are strongly influenced by the representation of vehicle activity patterns and fleet composition. This study developed a high-spatiotemporal-resolution bottom-up inventory for Kaifeng, China, using 2018 hourly traffic-monitoring data, road-network information, and locally corrected guideline emission factors, and compared it with a registration-based top-down inventory. Nine exhaust species were quantified: CO, NO X , VOCs, NH 3 , SO 2 , PM 10 , PM 2.5 , black carbon, and organic carbon. Relative to the top-down inventory, the bottom-up estimates were 9.9% lower for CO, 11.7% higher for NO X , 46.6% lower for PM 2.5 , and 164.6% higher for SO 2 . These opposing differences reflected the combined effects of activity allocation, cross-boundary traffic, fleet composition, and emission-standard-dependent emission factors. China III and China IV vehicles accounted for 97.95% of the observed on-road heavy-duty truck fleet, whereas pre-China III vehicles represented 66.84% of locally registered heavy-duty trucks. Light-duty passenger vehicles and taxis together accounted for 5.22 t yr⁻¹ (99.4%) of the 5.25 t yr⁻¹ difference in SO 2 . Among the tested spatial proxies, road-length density showed the strongest agreement with the bottom-up gridded CO inventory ( r = 0.854), followed by population density ( r = 0.733) and building density ( r = 0.426). These results highlight inherent structural discrepancies between the two methodologies, indicating that neither approach systematically dominates. The study aims to compare different methods of compiling emission inventories and provide insights for policy makers and environmental managers to enhance pollution control strategies.