Nonlinear behavior of organic aerosol in biomass burning plumes: amicrophysical model analysis

Abstract. Organic aerosol (OA) is a major component of smoke plumes from open biomass burning (BB). Therefore, adequate representation of the atmospheric transformations of BB OA in chemistry-transport and climate models is an important prerequisite for accurate estimates of the impact of BB emissions on air quality and climate. However, field and laboratory studies of atmospheric transformations (aging) of BB OA have yielded a wide diversity of observed effects. This diversity is still not sufficiently understood and thus not addressed in models. As OA evolution is governed by complex nonlinear processes, it is likely that at least a part of the observed variability of the BB OA aging effects is due to the factors associated with the intrinsic nonlinearity of the OA system. In this study, we performed a numerical analysis in order to gain a deeper understanding of such factors. We employ a microphysical dynamic model that represents gas-particle partitioning and OA oxidation chemistry within the volatility basis set (VBS) framework and includes a schematic parameterization of BB OA dilution due to dispersion of an isolated smoke plume. Several VBS schemes of different complexity, which have been suggested in the literature to represent BB OA aging in regional and global chemistry-transport models, are applied to simulate BB OA evolution over a five-day period representative of a BB aerosol lifetime in the dry atmosphere. We consider the BB OA mass enhancement ratio (EnR), which is defined as the ratio of the mass concentration of BB OA to that of an inert tracer and allows us to eliminate the linear part of the dilution effects. We also analyze the behavior of the hygroscopicity parameter, κ, that was simulated in a part of our numerical experiments. As a result, five qualitatively different regimes of OA evolution are identified, which comprise (1) a monotonic saturating increase of EnR, (2) an increase of EnR followed by a decrease, (3) an initial rapid decrease of EnR followed by a gradual increase, (4) an EnR increase between two intermittent stages of its decrease, or (5) a gradual decrease of EnR. We find that the EnR for BB aerosol aged from a few hours to a few tens of hours typically increases for larger initial sizes of the smoke plume (and therefore, smaller dilution rates) or for lower initial OA concentrations (and thus more organic gases available to form secondary OA). However, these dependencies can be weakened or even reversed, depending on the BB OA age and on the ratio between the fragmentation and functionalization oxidation pathways. Nonlinear behavior of BB OA is also exhibited in the dependencies of κ on the parameters of the plume. Application of the different VBS schemes results in large quantitative and qualitative differences between the simulations, although our analysis suggests also that the main qualitative features of OA evolution simulated with a complex two-dimensional VBS scheme can also be reproduced with a much simpler scheme. Overall, this study indicates that the BB aerosol evolution may strongly depend on parameters of the individual BB smoke plumes (such as the initial organic aerosol concentration and plume size) that are typically not resolved in chemistry transport models.