Research Projects

Project at a Glance

Project Status: complete

Title: Organic and elemental carbon size distributions of Los Angeles aerosols measured during SCAQS

Principal Investigator / Author(s): McMurray, Peter H

Contractor: University of Minnesota

Contract Number: A732-075

Research Program Area: Atmospheric Processes


This report summarizes measurements of size-resolved residual organic carbon (ROC)and elemental carbon (EC) made by researchers from the University of Minnesota during the Southern California Air Quality Study (SCAQS). Measurements were made with MOUDI impactors. Impactors were located at Claremont and Rubidoux during summer SCAQS sampling, and at Long Beach and Los Angeles during the fall.

It was found that the average mass mean diameter of ROC ranged from 0.43m at Los Angeles to 0.62m at Rubidoux. Mass mean diameters of EC were systematically smaller than for ROC, ranging from 0.39m at Los Angeles to 0.45m at Rubidoux. The Rubidoux ROC mean sizes were systematically and substantially larger than sizes at the other three sites.

The ROC/EC ratio varied with site, season, and particle size. Average ROC/EC ratios were 2.66 (Claremont), 2.12 (Rubidoux), 1.51 (Long Beach) and 1.26 (Los Angeles). Although there was no systematic dependence of ROC/EC ratios with particle size, there was a tendency for values to be higher for particless at the large (~1m ) and small (~0.05-0.1m ) extremes of the measured size spectrum.

The data were 'inverted ' to account for the effect of size-dependent collection efficiencies on measured size distributions. All data were inverted with two inversion schemes: the modified Twomey routine developed by Walter John and coworkers at the Air and Industrial Hygiene Laboratory (Berkeley, CA), and the MICRON code developed by Wolfenbarger and Seinfeld (Caltech). Integral moments of inverted size distributions obtained with these two inversion schemes were compared. It was found that for the integral moment corresponding to particle mass (the 3rd moment) the two inversion methods were in good agreement. This is not surprising since the techniques are constrained to return the measured mass. However, for smaller or larger moments the integral moments can be up to several orders of magnitude different. Although resolving differences between inversion schemes is beyond the scope of the present research, the discrepancies in integral moments is a serious concern. Because size distributions are usually measured with the objective of calculating some integral moment (light scattering of absorption, dry deposition, lung deposition, etc.), it is essential to have some means of estimating most likely values and uncertainties of such integrals. This result suggests that uncertainties for integrals other than total mass can be large.

A limited number of comparisons between EC concentrations obtained with the MOUDI with other samplers was done for the Claremont site. As with previous studies it is found that correlations are poor, with R2 values ranging down to 0.35. MOUDI EC data were found to be somewhat better correlated with the Ford Spectrometer (an optical absorption device) data than with data from a quartz filter. A similar result was found during the 1986 Carbonaceous Species Methods Comparison Study where the MOUDI agreed best with the LBL aethalometer, another optical absorption device. McMurry and Zhang (Aerosol Sci. Technol. 10:430-437) discussed discrepancies for carbon measurements among different samplers and conclude that impactors may be superior to filters for sampling carbon-containing particulate matter.


For questions regarding research reports, contact: Heather Choi at (916) 322-3893

Stay involved, sign up with ARB's Research Email Listserver