Grantee Research Project Results
2005 Progress Report: New Technologies for Source Apportionment
EPA Grant Number: R832157Title: New Technologies for Source Apportionment
Investigators: Henry, Ronald C. , Fine, Philip M.
Institution: University of Southern California
EPA Project Officer: Chung, Serena
Project Period: January 20, 2005 through January 19, 2008 (Extended to January 19, 2009)
Project Period Covered by this Report: January 20, 2005 through January 19, 2006
Project Amount: $450,000
RFA: Source Apportionment of Particulate Matter (2004) RFA Text | Recipients Lists
Research Category: Air , Air Quality and Air Toxics , Particulate Matter
Objective:
The objective of this research project is to develop new receptor-oriented methods of source apportionment based on the source information in wind speed and direction and the periodic variations of concentrations as extracted from the data by nonparametric regression and Fourier analysis. These methods will be applied to airborne particulate and other air quality datasets nationwide. Also, the source apportionment aspects of a particulate health-effects study to be carried out in Long Beach, California, will be strengthened by additional analysis of particulate samples for organic and inorganic species funded by this project. The newly developed receptor models will be augmented with the chemical mass balance (CMB) model and the Unmix multivariate receptor model.
Progress Summary:
Size-fractionated particulate matter (PM) samples and meteorological data have been collected in the Long Beach, California, area. Samples were analyzed for mass concentrations, organic carbon (OC) and elemental carbon (EC) concentrations, and elemental concentrations of Na, Mg, Al, S, K, Ca, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Sr, Cd, Ba, and Pb. Preliminary results showed that stagnant meteorological conditions created relatively higher OC and EC concentrations in the winter compared to the summer sampling campaigns. In general, concentrations of PM mass, OC, EC, S, and trace elements often were not significantly different among the various sampling sites, suggesting the regional nature of these species. Intersite correlations and coefficients of divergence resulted in mostly homogenous spatial distributions in mass, OC, and elements in all three size fractions. Higher spatial variability, however, was observed in the smallest aerosol size fractions, indicating the effect of localized sources and fresh PM emissions. The size-resolved datasets also were analyzed using positive matrix factorization to identify possible source categories. The ultrafine fraction was resolved into three factors, whereas four factors characterized the accumulation and coarse aerosols. All three size fractions exhibited a distinct oil combustion source, with S, V, and Ni as the principal elements. Results from this study have identified V as a possible indicator of ship emissions from nearby seaports in that area. In addition to the elemental analysis, size-fractionated PM samples have been spiked with internal standard compounds and extracted. Future work will include derivatization and analysis by gas chromatography/mass spectrometry for trace organic species. This information will be used in conjunction with the elemental data in source apportionment calculations, and results can be compared to the other statistical analyses of highly time-resolved continuous data based on wind speed, direction, and pollutant levels.
Significant development and improvements were made to nonparametric source apportionment methods and Fourier methods for missing and irregularly spaced data. These methods were applied to novel datasets, including 1-minute average data for criteria pollutants and meteorological data from the primary air quality monitoring site in Long Beach, California, and data on ultrafine particle numbers from four sites in southern California.
A new nonparametric trajectory source apportionment method was invented that uses local trajectories constructed with observations of wind speed and direction taken every minute. This new method clearly identifies local sources and quantifies the impact of these on the monitoring site. Using this method, many of the thousands of continuous air quality monitors now operating in the United States could be put to much better use simply by saving the 1-minute data, not aggregating it to 1-hour averages. Applying the method to Long Beach for May, June, and July of 2005 results in the local source areas shown in red in the figure below. The results are superimposed on an aerial photo of the Long Beach area. The monitoring site is located at (0,0) on the grid, is less than a quarter mile north of a major freeway, and less than 1 mile east of the intersection of two major freeways. The freeways are not seen to be major sources of high PM concentrations in the figure. The Port of Los Angeles, however, does seem to be a major contributor to high PM10, along with other sources off the grid to the north. Other nonparametric methods were applied to 1-hour average ultrafine particle data. This analysis showed that at an isolated site near a freeway the ultrafine particles are associated with the freeway emissions. At sites with a more complex mix of sources, however, the sources of ultrafine particles also are complex. Nonparametric methods have shown that sometimes nearby sources (such as freeways) are important contributors to high pollutant concentrations and, surprisingly, sometimes they are not.
A method was developed to estimate the Fourier transform (not just the periodogram) for time-series with missing or irregularly spaced data. When applied to the 1-minute observations of criteria pollutants at Long Beach, no significant periodicities less than 24 hours were found. Thus, the 1-minute data at Long Beach used to create Figure 1 seemed to be free of periodic behavior that could complicate its use in source apportionment models. The method also was applied to a time-series of more than 13 years of 24-hour average particulate composition data from the Interagency Monitoring of Protected Visual Environments (IMPROVE) site in Washington, DC, and a shorter 3-year period of similar data from Phoenix, Arizona. The first 10 years of the DC data are impossible to analyze by standard Fourier methods because the data were taken every Wednesday and Sunday, giving an irregular spacing of 3 then 4 days between samples (not to mention missing data). Several species, including EC, showed strong weekly periodicity, indicating that these species have significant local sources. In the case of EC this is consistent with a diesel source. There is a very large volume of historical IMPROVE particulate composition data from many sites that now can be analyzed for periodic behavior that could have significance for source apportionment.
Figure 1. Average Concentration of PM10 in µg/m3 at the Long Beach Monitoring Site When the Air Passes Over the Areas Shown in Color. The average PM10 concentration is 25 µg/m3.
Future Activities:
Additional 1-minute data must be obtained for multiple sites, preferably near Long Beach, California, for further development of the nonparametric trajectory model. Additional datasets with short time average particulate and air quality data will be identified and if these can be obtained, the nonparametric and Fourier methods will be applied. The major objective of the organic analysis/CMB work is to determine the chemical composition of size-fractionated PM at various sites in Long Beach to evaluate PM components and their relationship to various PM sources on a community scale. This will be accomplished by quantifying individual organic compounds in size-fractionated samples from the various sampling locations, identifying previously studied source profiles for major sources categories in the Long Beach area, and using CMB methods to estimate contributions of different source types to the measured pollutant concentrations.
Journal Articles on this Report : 1 Displayed | Download in RIS Format
Other project views: | All 15 publications | 6 publications in selected types | All 6 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Dilmaghani S, Henry IC, Soonthornnonda P, Christensen ER, Henry RC. Harmonic analysis of environmental time series with missing data or irregular sample spacing. Environmental Science & Technology 2007;41(20):7030-7038. |
R832157 (2005) R832157 (2007) R832157 (Final) |
Exit Exit Exit |
Supplemental Keywords:
Air quality, source apportionment, chemical mass balance, organic aerosol, statistical analysis, nonparametric regression, data analysis, Southern California, CA, health effects,, RFA, Scientific Discipline, Air, Ecosystem Protection/Environmental Exposure & Risk, particulate matter, Air Quality, Environmental Chemistry, Monitoring/Modeling, Environmental Monitoring, Environmental Engineering, particulate organic carbon, atmospheric dispersion models, atmospheric measurements, model-based analysis, source apportionment, chemical characteristics, emissions monitoring, environmental measurement, airborne particulate matter, air quality models, air quality model, air sampling, speciation, particulate matter mass, analytical chemistry, modeling studies, real-time monitoring, aerosol analyzers, chemical speciation sampling, particle size measurement, atmospheric chemistryProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.