The Role Played by the Bulk Hygroscopicity on the Prediction of the Cloud Condensation Nuclei Concentration Inside the Urban Aerosol Plume in Manaus, Brazil: From Measurements to Modeled Results

Abstract

This work use observed and modeled data to trace a basic model that describes how the bulk chemical composition can be used to determine the aerosol size-dependency of hygroscopicity inside the urban aerosol pollution plume from Manaus, Brazil. The procedure is then used to predict the Cloud Condensation Nuclei (CCN) concentration (NCCN) from the Aerosol Size Distribution (ASD). The work briefly described hourly averaged data of ASD, Black Carbon size distribution (BCSD), Black Carbon mass concentration (BC), NCCN, and aerosol size hygroscopicity (κ) observed about 70 km downwind of the City of Manaus, inside the amazon rain forest, during 60 days of 2014, in the course of the rainy season. It uses a description of aerosol chemical composition obtained previously in the city of São Paulo, Brazil, as starting point to formulate a model that describes how particles in the pollution plume act as CCN as a function of their sizes and bulk chemical composition. A parameterization based on a polynomial fitting is proposed to determine BCSD from the BC. The measured and parameterized BCSD agree very well with more than 98% of the observed BCSDs, presenting a correlation coefficient larger than 0.9 with observation. The parameterized BC number concentration is also very well represented, with a correlation coefficient of 0.91 and a slope of 0.99 in comparison with observations. This work also studies how soot particles are incorporated in the CCN population and propose a parameterization to represent it. Simulations suggest that the larger BC particles are only partially incorporated as CCN and BC is a constraint to the fraction of aerosols that can act as CCN in all size ranges. The simulations also suggest that the incorporation of BC as CCN is probably only effective through coagulation in the presence of the wet phase. In addition, the work compares different strategies to predict observed NCCN from ASD. It is shown that the strategy presented can substantially improve the results of comparison of modeled and measured data from as bad as 2.14 ± 0.65 to as good as 1.02 ± 0.17. This study shows that the knowledge of the bulk chemical composition can be used to predict both the size-resolved chemical composition and mixing state, reducing the number of parameters needed to predict NCCN.