The details of Bingqing Zhang's Ph.D. Thesis defense are listed below:
Title of Thesis: Constraining Atmospheric Aerosol Emissions and Properties Across Temporal and Spatial Scales Using Modeling Techniques
Defense Date: June 25, 2025
Defense Time: 11:00 am
Defense Location: ES&T: L1114
Names of Advisor and Committee members: Dr. Pengfei Liu (EAS, advisor), Dr. Greg Huey (EAS), Dr. Rodney Weber (EAS), Dr. Jennifer Kaiser (CEE&EAS), Dr. Da Pan (CEE).
Abstract: Atmospheric aerosols can be emitted directly from natural or anthropogenic processes or formed secondarily through chemical reactions. Understanding these processes is crucial for assessing their potential impacts on atmospheric chemistry, climate forcing, and human health. This thesis uses a combination of emission reconstruction, atmospheric modeling, and machine learning to constrain aerosol emissions and properties across different temporal and spatial scales. In the first part, we developed a comprehensive global emission inventory of hydrogen chloride (HCl) and particulate chloride (pCl) from continental sources using a bottom-up approach. These sources have traditionally been considered negligible compared to the dominant contribution from sea salt aerosols. This inventory provides gridded, model-ready emissions at 0.1°×0.1° resolution for the period 1960-2014. It has been used by other researchers to simulate its impact on secondary organic aerosol (SOA) formation, nitrate chemistry, and atmospheric oxidative capacity. In the second work, we reconstructed biomass burning emissions from preindustrial period (i.e., 1750) to the present day using black carbon (BC) as a tracer. This work employed an inverse modeling framework that leveraged BC deposition fluxes from a global array of 31 ice core records, along with two different historical emission inventories as a priori estimates, and GEOS-Chem chemical transport model simulated emission-deposition sensitivities. The reconstructed emissions exhibit greater temporal variability and higher preindustrial aerosol levels compared to existing inventories, with important implications for understanding historical aerosol radiative forcing and evaluating the magnitude of aerosol-driven climate cooling. In the third work, we examined aerosol properties, focusing on aerosol liquid water content (ALWC) and aerosol acidity (indicated by aerosol pH) over the contiguous United States. We developed high resolution datasets of ALWC (1km×1km, daily) and aerosol pH (1km×1km, monthly) for the period 2000-2019 by training machine learning models using GEOS-Chem outputs to capture underlying thermodynamic relationships, and then integrating them with high-resolution, observation-constrained aerosol mass and composition datasets. To demonstrate their utility, we incorporated these datasets into a simplified iron solubility model, illustrating their value for studying aerosol-driven metal dissolution, spatial variability in soluble iron, and associated human exposure and health risks.