DOI

10.17077/etd.qvtgdzfr

Document Type

Dissertation

Date of Degree

Spring 2018

Access Restrictions

Access restricted until 07/03/2020

Degree Name

PhD (Doctor of Philosophy)

Degree In

Chemical and Biochemical Engineering

First Advisor

Stanier, Charles O.

First Committee Member

Bradley, A. Allan

Second Committee Member

Carmichael, Gregory R.

Third Committee Member

Nuxoll, Eric E.

Fourth Committee Member

Spak, Scott N.

Abstract

New particle formation (NPF) from nucleation and subsequent nuclei growth are frequently observed phenomena in the troposphere. Previous studies show that nucleation increases condensation nuclei (CN) and cloud condensation nuclei concentrations (CCN). CCN can change cloud properties and affect meteorology indirectly. Meteorology, in turn, affects concentrations of air pollutants. The Midwestern United States hosts about 21% of the U.S. population. It has high levels of anthropogenic emissions of atmospheric aerosols and sulfur dioxide (SO2, a widely acknowledged gas precursor for nucleation). Thus, it is of great importance to understand the concentration, composition, size distribution and other properties of particles in this area. This thesis aims to improve the current understanding of physical and chemical properties of particles in the Midwest via the use of Chemical Transport Model (CTM).

In this work, the fully coupled NPF-explicit WRF-Chem model was used to perform nucleation related simulations in the Midwest. This model has the capability of linking NPF to cloud properties, and to changes in both meteorology and air quality. Results show that NPF leads to more shortwave radiation reaching the surface, higher surface temperatures, and a deeper boundary layer (PBL). For air pollutants, the most pronounced influence of PBL nucleation is a reduction in PM2.5 concentrations, which is mainly caused by decreases in secondary sulfate. Sensitivity tests demonstrate that changes in cloud processing of SO2 are primarily responsible for SO42- reduction when PBL nucleation is modeled. Thus, a feedback loop was discovered for the first time in which large CCN concentration reduction in turn suppresses aqueous SO2 oxidation.

Model-measurement comparison shows the NPF-explicit WRF-Chem model is able to reproduce observed NPF events in spring (April) and late fall (November) in Bondville, IL, a representative Midwestern rural site. By applying different anthropogenic emissions scenarios, simulation results show that particle number concentrations in the Midwest decreased significantly from year 2005 to 2011. CCN concentrations at the low water supersaturation were found to be impacted by a combination of factors, i.e. aqueous chemistry, gas-phase chemistry, emissions, .etc. Meanwhile, the model lacks the ability to reproduce NPF events in September (early fall) and February (winter). Possible causes are lack of SOA treatment, simulated meteorology biases as well as the uncertainty of nucleation scheme being used.

Last but not the least, a high spatial resolution (0.44 km) WRF-CMAQ modeling framework has been designed to study the impact of a Midwestern power plant to local air quality. Preliminary modeling results are satisfactory and the remaining work is to be completed.

Pages

xv, 151 pages

Bibliography

Includes bibliographical references.

Copyright

Copyright © 2018 Can Dong

Available for download on Friday, July 03, 2020

Share

COinS