10th International Aerosol Conference September 2 - September 7, 2018 America's Center Convention Complex St. Louis, Missouri, USA
Abstract View
Physical and Model-based Characterization of NPF Events and Sensitivity of CN and CCN to Changes in Anthropogenic Emissions in the Midwestern United States
Can Dong, Robert Bullard, Ashish Singh, Scott N. Spak, Hitoshi Matsui, CHARLES STANIER, University of Iowa
Abstract Number: 913 Working Group: Aerosol Modeling
Abstract New particle formation (NPF) is a frequently observed phenomenon worldwide. Ten months (July 2013 to June 2014) of continuous measurements of particle number size distribution were performed at a rural Midwestern United States location (Bondville, Illinois), spanning sizes from 3 nm to 2.5 microns. Bondville is representative of Midwestern United States and other perturbed continental locations with moderate sulfur dioxide (SO2) and fine particulate matter (PM2.5) concentrations (i.e. SO2 of 0.5-3 ppb, PM2.5 of 5-15 µg/m3) at background sites.
New particle formation behavior at the site is similar to other mid-latitude sites, with NPF in 2013-2014 occurring on 44.1% of classifiable days, highest NPF frequency in spring and fall, followed by summer and lowest in winter. Observed growth rates for measured nucleation mode particles (3-25 nm) were highest (8.97 nm/h) in summer and lowest (1.45 nm/h) in winter.
SO2 and sulfate concentrations are dropping in the region, and the fractional importance of secondary organic aerosol is increasing (SOA). For example, at Bondville, SO2 fell from 2.1 to 0.8 ppb from 2005 to 2013, a 62% decrease. As a test of the nucleation explicit version of WRF-Chem, which simulates nucleation using a 20-bin sectional model down to 1 nm, each season was simulated using 2005 NEI (higher SO2 emissions), 2011 NEI (lower SO2 emissions), and sensitivity tests were done with inclusion and exclusion of growth by SOA.
NPF events were captured well by the NPF-explicit WRF-Chem model in April and November without considering secondary organic aerosol (SOA) formation. The two anthropogenic emission inventories, i.e. 2005 NEI and 2011 NEI, were used to study sensitivities of CN (condensation nuclei) and CCN (cloud condensation nuclei) to emissions in the Midwest. Modeling results show that SO2 emission reduction is the primary cause for CN number concentration change. Switching SO2 emissions from 2005 NEI to 2011 NEI led to significant decrease of CN10 (particles with sizes larger than 10 nm) number concentrations, especially in April, September and November (over 20% decrease). On average, model simulated SO2 concentrations by using 2011 NEI were approximately half of the values simulated by using 2005 NEI (2.17 ppb vs 4.32 ppb). Changes of primary aerosol and NH3 emissions also play a role, but the influence is not as significant as that of SO2. Simulated changes of CCN at lower water supersaturations (0.2%) were complicated due to interactions between meteorology and chemistry. At higher water supersaturations (1%), CCN changes were consistent with that of CN10.
Comparison to independent measurements of ultrafine particle number with height (by numerous light aircraft flights) showed that the model’s aerosol number predictions do not decrease fast enough with height, an indication of the model having too much vertical mixing, insufficient particle removal, or too strong of a free tropospheric nucleation source.
During the February and September simulation periods, bursts of new particles were reproduced but the subsequent growth was not. The sensitivity of model performance on meteorology and SOA is discussed. Finally, simulations with very low SO2 emissions are performed to show the impacts on air quality, CCN, and the number size distribution under future such scenarios.