Mesoscale modeling of transport and radiative impacts of Central American smoke aerosols

Under specific meteorological conditions, smoke aerosols produced by the Central American Biomass Burning (CABB) can be transported northward across the Gulf of Mexico and intrude into the southeastern United States (SEUS), thousands of kilometers from the source region. These widely dispersed smoke aerosols not only degrade the air quality and visibility, but also impact the meteorological and photochemical processes through their radiative effects. One of key components in modeling the smoke transport is the realistic specification of smoke source function, including its spatial and temporal variations. A wealth of ground-based and space-borne observations has consistently indicated that biomass burning emissions has a distinct diurnal variation with the peak near local noon time. To date, the consideration of such diurnal variation in smoke emissions is still very limited in the chemistry transport models, not to mention the smoke radiative effects that highly depend on the solar zenith angle.

In this study, we present a coupled aerosol-radiation-meteorology mesoscale model, RAMS-AROMA, which is suitable to investigate the smoke transport and radiative impacts simultaneously. RAMS-AROMA originates from RAMS but has newly developed capabilities of Assimilation and Radiation Online Modeling of Aerosols (AROMA). The smoke radiative effect and its feedback on meteorology are considered at each model time step and model grid. We focus on the simulation of the CABB smoke events in April – May 2003 that were the largest since 1998. Specification of the diurnal variation of smoke emissions is made possible by using an hourly smoke emission inventory from the Fire Locating and Modeling of Burning Emissions (FLAMBE) geostationary database in RAMS-AROMA. Comprehensive evaluation of model performance by using datasets from EPA PM2.5 network, IMPROVE, and ARM SGP showed that RAMS-AROMA well simulated the smoke front timeline and smoke spatial distribution. A top-down analysis showed that about 1.3Tg smoke particles were emitted in the study time period, among which 55% was transported to the SEUS. The extinction of solar radiation by smoke resulted in the decrease of 2-meter air temperature by more than 0.2K in the smoke-impacted region, but the smoke absorption of solar radiation increased the air temperature in the upper boundary layer. The model simulation numerically verified a positive feedback mechanism in which the increase of atmospheric stability caused by the smoke radiative effects further traps more smoke particles in the lower boundary layer.