Black carbon (BC) aerosol, as a typical optical absorption aerosol, is of great significance to the study of climate and radiation. The China Atmosphere Watch Network (CAWNET), established by the China Meteorological Administration (CMA), contains 35 BC-monitored stations, which have been collecting data using commercial Aethalometer instruments (AEs) since 2006. Element carbon (EC) data measured from the thermal/optical reflectance (TOR) method was used to correct the BC monitoring data from the AEs, which are affected by various sampling and analytical artifacts. The average difference before and after the revision was about 17.3% (±11.5%). Furthermore, we analyzed the variations of BC in China from 2006 to 2017 using a revised dataset. The ten-year averaged concentration of BC would have been applicable for climate analysis, and can be a comparison sample in future research. The concentrations of BC across the stations in China showed a general downward trend, with occasional fluctuations, and the concentrations at urban sites decreased more significantly. The average concentrations of BC in urban sites are higher than rural and remote sites. The 10-year averaged concentration of BC ranges from 11.13 μg m−3 in Gucheng to 0.19 μg m−3 in Shangri-La, showing a strong spatial variation; the proportion of BC aerosol in PM2.5 is generally less than 20%. The BC showed obvious seasonal and diurnal variation; and the highest concentration occurred in winter, with more dramatic diurnal variation, followed by autumn and spring. There was a significant increase in concentration between local time 7:00–9:00 and 18:00–0:00. The distribution and trend of BC concentration in China showed a consistency with emissions of BC.

Carbon Bond Mechanism IV (CBM-IV) is a widely used reaction mechanism in which VOCs are grouped according to the molecular structure. In the present study, we applied a sensitivity analysis on the CBM-IV mechanism to clarify the importance of each reaction under two different initial conditions (urban and low-NO scenarios). The reactions that exert minor influence on the reaction system are then screened out from the mechanism, so that a reduced version of the CBM-IV mechanism under specific initial conditions can be obtained. We found that in a typical urban condition, 11 reactions can be removed from the original CBM-IV mechanism, and the deviation is less than 5% between the results using the original CBM-IV mechanism and the reduced mechanism. Moreover, in a low-NO initial condition, two more reactions, both of which are nitrogen-associated reactions, can be screened out from the reaction mechanism, while the accuracy of the simulation is still maintained. It is estimated that the reduction of the CBM-IV mechanism can save 11–14% of the computing time in the calculation of the chemistry in a box model simulation.

No abstract