Constraining radiative forcing of Asian carbonaceous aerosols with Observations and CESM1/CAM5

Carbonaceous aerosols, including black carbon (BC) and organic carbon (OC), are significant contributors to the anthropogenic climate change. BC is considered as the second largest warming agent. However, the direct radiative forcing of carbonaceous aerosols is still quite uncertain, in particular over Asia. To better constrain the present-day Asian carbonaceous aerosol forcing, we utilize both a top-down approach that is primarily based on ground-based and satellite observations over the first decade of 21st century, as well as a bottom-up approach that is based on the latest global climate model coupled with an interactive chemistry and aerosol module (CESM1/CAM5/MAM3). (1) The comparisons of top-down observational estimates with bottom-up model simulations suggest the model considerably underestimates atmospheric heating of BC by at least a factor of three over Asia. The major source of discrepancy between observations and models are speculated to be emission inventory, which is developed based on limited economic activity data reported by developing countries. A series of sensitivity test in which BC anthropogenic emission are increased by difference factors are conducted, which suggested BC emission sources are underestimated by a factor of three to five over Asia. (2) By applying a new partitioning scheme to the observed aerosol optical properties, we show that OC can contribute up to 20% of atmospheric heating, and thus the overall TOA cooling of OC is previously overestimated. The biases from the model can be attributed to the model assumption of the BC and OC refractive indexes primarily developed from laboratory measurements, rather than ambient environment. In particular, the model currently cannot sufficiently account for OC absorption, leading to a factor-of-two underestimation of its atmospheric heating and consequently overestimation of TOA cooling. The adjustment of the OC refractive index to match the empirically derived single scattering albedo improved the agreement between observations and the model.