Atmospheric fossil fuel CO2 traced by 14CO2 and air quality index pollutant observations in Beijing and Xiamen, China.

Radiocarbon (14C) is the most accurate tracer available for quantifying atmospheric CO2 derived from fossil fuel (CO2ff), but it is expensive and time-consuming to measure. Here, we used common hourly Air Quality Index (AQI) pollutants (AQI, PM2.5, PM10, and CO) to indirectly trace diurnal CO2ff variations during certain days at the urban sites in Beijing and Xiamen, China, based on linear relationships between AQI pollutants and CO2ff traced by 14C ([Formula: see text]) for semimonthly samples obtained in 2014. We validated these indirectly traced CO2ff (CO2ff-in) concentrations against [Formula: see text] concentrations traced by simultaneous diurnal 14CO2 observations. Significant (p < 0.05) strong correlations were observed between each of the separate AQI pollutants and [Formula: see text] for the semimonthly samples. Diurnal variations in CO2ff traced by each of the AQI pollutants generally showed similar trends to those of [Formula: see text], with high agreement at the sampling site in Beijing and relatively poor agreement at the sampling site in Xiamen. AQI pollutant tracers showed high normalized root-mean-square (NRMS) errors for the summer diurnal samples due to low [Formula: see text] concentrations. After the removal of these summer samples, the NRMS errors for AQI pollutant tracers were in the range of 31.6-64.2%. CO generally showed a high agreement and low NRMS errors among these indirect tracers. Based on these linear relationships, monthly CO2ff averages at the sampling sites in Beijing and Xiamen were traced using CO concentration as a tracer. The monthly CO2ff averages at the Beijing site showed a shallow U-type variation. These results indicate that CO can be used to trace CO2ff variations in Chinese cities with CO2ff concentrations above 5 ppm.