Identifying and quantifying transported vs. local sources of New York City PM2.5 fine particulate matter air pollution

https://doi.org/10.1016/j.atmosenv.2006.04.068Get rights and content

Abstract

New York City (NYC) is presently in violation of the nation's PM2.5 annual mass standard, and will have to take actions to control the sources contributing to these violations. This paper seeks to differentiate the impact of long-range transported aerosols on the air quality of downtown NYC, so that the roles of local sources can more clearly be evaluated. Past source apportionment studies have considered single sites individually in their source apportionment analyses to identify and determine sources affecting that site, often finding secondary sulfates to be an important contributor, but not being able to quantify the portion that is transported vs. local. In this study, a rural site located in Sterling Forest, NY, which is near to the NYC area, but unaffected by local NYC sources, is instead used as a reference to separate the portion of the aerosol that is transported to our Manhattan, NYC site before conducting the source apportionment analysis. Sterling Forest is confirmed as a background site via elemental comparisons with NYC during regional transport episodes of Asian and Sahara sandstorm dusts, as well as by comparisons with a second background site in Chester, NJ. Two different approaches that incorporate Sterling Forest background data into the NYC source apportionment analysis are then applied to quantify local vs. transported aerosols. Six source categories are identified for NYC: regional transported sulfate, trans-continental desert dust, traffic, residual oil, “local” dust and World Trade Center fires pollution. Of these, the transported sulfates and trans-continental desert dust accounted for nearly half of the total PM2.5 mass in Manhattan during 2001, with more than half coming from these transported sources during the summer months. More than 90% of the Manhattan elemental carbon was found to be of local origins. Conversely, roughly 90% of the NYC sulfate mass was identified as transported into the city. Our results indicate that transported pollution has a major role in NYC's fine PM pollution. Reliably meeting the ambient PM2.5 air quality standards in New York will require that upwind sources, outside of the city, will also need to be controlled.

Introduction

Chemically non-specific particulate matter (PM) mass concentrations cannot alone provide a complete indication of PM toxicity. Instead, it is a combination of size and chemical constituents of PM that are proving to be important. In 2000, the US Environmental Protection Agency (EPA) Speciation Trend Network (STN) was initiated to monitor for PM2.5 (particulate matter less than 2.5 μm in diameter) and its elemental constituents at over 200 sites across the US (US EPA, 1999). This provides greater details about the characteristics of PM and using source apportionment tools will allow the identification and estimation of mass contributions from different PM sources (Cooper and Watson, 1980). For a city, such as New York City (NYC), where the impact of “long-range” transport aerosols is well established, speciation data provides an opportunity to apportion NYC PM2.5 into contributions from local and non-local sources.

In the past, multiple studies have compared PM mass and elemental tracers, such as NH3, K, NO3, SO4, carbon, and H+ levels, between sites in the eastern US in order to estimate the impacts of secondary and transported aerosols (Vukovich and Sherwell, 2002; Bari et al., 2003; Dutkiewicz et al., 2004; Khan et al., 2006). The availability of multiple years of STN data, has led to several newer source apportionment studies, allowing the identification of the various sources impacting air quality at these sites, and a comparison of the source's relative impacts. Two recent studies conducted for NYC (Ito et al., 2004; Qin et al., 2006) used every 3-day STN data for multiple sites in the NYC metro area, and conducted independent source apportionment analyses for each of the sites. The types of sources and mass contributions were compared between the sites, and a “secondary sulfate” aerosol source was observed at each of the multiple sites. However, it is unclear exactly how much of the secondary sulfate identified at each site was transported versus local.

In this paper, we investigate an alternate approach to conducting a source apportionment wherein data from an additional “background” site is incorporated into the NYC site source apportionment analysis to help quantitatively estimate the roles of transported vs. local aerosols as a part of the source apportionment. We apply daily elemental data collected during 2001 by New York University (NYU) at two sites located: (1) in downtown Manhattan; and, (2) in Sterling Forest (Tuxedo, NY). Sterling Forest is a rural site surrounded by thousands of acres of largely undeveloped woodland, with an area approximately the size of Manhattan, and it is situated approximately 50 miles west-northwest of NYC. This site was chosen with the goal of comparison with aerosols collected at the Manhattan site. Sterling Forest experiences the same transported aerosols as in NYC, but none of the local sources. In the absence of unique tracers for the local vs. transported portions of each of the different source contributions, we propose incorporating the complementary data from the background (Sterling Forest) site directly into the NYC source apportionment model in order to provide a quantification of the transported aerosol contribution as an integral part of the NYC site apportionment.

Two approaches that incorporate background data into the model are applied in this research. Since numerous past studies have shown a major impact of transported sulfates into NYC air quality, Sterling Forest sulfur data is included in the NYC source apportionment as an additional variable that should help more clearly define “transported aerosols” in the first apportionment approach. That is, the sulfur measurements taken at the background site would essentially act as a tracer for aerosols transported into the NYC metro area, with the remainder of the NYC sulfur (i.e. the difference in sulfur levels between the two sites) then being more clearly attributed by the model to local sources in the source apportionment analysis. In the second approach, we assume Sterling Forest is a good background site for all transported aerosols (not just sulfates), and therefore it incorporates all the background data into the NYC source apportionment model. In this second approach, two separate source apportionments are conducted on: (1) the Sterling Forest data; and (2) the “remainder” elemental data at NYC after “subtracting off” the Sterling Forest concentrations, in order to then provide separate apportionments for the transported vs. local sources affecting the NYC site.

Section snippets

Data collection

NYU set up two PM sampling sites in late 2000, and began sampling PM2.5 and its chemical constituents on a daily basis in January 2001. At both sites, 24-h (midnight-to-midnight) PM2.5 filter samples were collected using an R&P ACCU sampler (operating at 13.7 LPM), and every half-hour continuous real-time PM2.5 mass concentration data were simultaneously collected using an R&P TEOM (Thermo Electron Corporation, 2005). An R&P 5400 carbon analyzer using “thermal-CO2” was also used to measure

Evaluating sterling forest as a NYC background site

Before applying the Sterling Forest data as a background pollution site for NYC, we first evaluated this assumption. To test the suitability of Sterling Forest as a background site, the elemental and carbon data collected at the site were compared with both an STN “background” site located in Chester NJ and NYU's Manhattan NYC site (a map indicating the location of these sites is provided in Fig. 1). It is apparent from the results in Table 1 that, while PM2.5, sulfur and OC are highly

PMF source apportionments

Case 1: PMF analysis separating transported and local sulfur.

This source apportionment analysis of NYC data (that incorporates the Sterling Forest sulfur) identifies six sources of PM2.5, as shown in Table 3. The two sulfur variables considered in the NYC source apportionment analysis now clearly defines sources as being either transported or local. Two of the sources, namely, “transported sulfates” and “transported desert dust” are from non-local sources. On the other hand, “residual oil

Discussion

The major role of aerosols transported into NYC indicated by our analysis highlights the significance of the upwind sources on the impact of the city's air quality. As mentioned previously, this transport phenomenon has been well documented by several past studies that identify transported aerosols as a major part of PM problem in the northeastern US. Some of these past studies have made estimations as to the contributions of transported aerosols in NYC through external site-to-site comparisons

Acknowledgements

The authors wish to thank the following NYU-NIEHS Center members for their contributions to the collection of the data discussed in this paper: M. Blaustein, J. Gorczynski, S.I. Hsu, and Dr. M. Lippmann of the NYU SOM. We would like to thank Gordon Cook for providing the map used in Figure 1. We also thank Hunter College and the NYU Downtown Hospital for allowing us to locate our particulate matter air samplers at their facilities. This research was supported by a grant from the New York State

References (18)

There are more references available in the full text version of this article.

Cited by (58)

  • Metals and air pollution

    2021, Handbook on the Toxicology of Metals: Fifth Edition
  • Synergistic aircraft and ground observations of transported wildfire smoke and its impact on air quality in New York City during the summer 2018 LISTOS campaign

    2021, Science of the Total Environment
    Citation Excerpt :

    With the prevailing western and northwest winds, the site at Newburgh (41.499° N, 74.099° W, 83-km away in the north of CCNY) located in the northwest NYC generally represent the upwind rural area of NYC, where the hourly PM2.5, CO and BC are observed. The correlation of the pollutants in the urban and upwind rural areas can be an indicator of regional transport (Lall and Thurston, 2006). Further details about these sites can be found at the website (http://www.dec.ny.gov/chemical/8406.html).

  • Long-term trends in local and transported PM<inf>2.5</inf> pollution in New York City

    2021, Atmospheric Environment
    Citation Excerpt :

    Secondary pollutant content in PM2.5 can be used to estimate the fraction of local PM2.5 that originates from regional transport. Though some secondary pollutants are formed locally, the proportion of regional contribution can be approximated by the concentrations observed at a nearby rural area, where local emissions are negligible compared to regional input (Lall and Thurston 2006; Qin et al., 2006). PSP is a rural location approximately 310 km northwest of NYC and has been identified in other studies as a suitable site for the estimation of transported PM in the region due to its exposure to the primary direction of long-range transport from the west (Rattigan et al., 2016; Schwab et al., 2009).

  • Metals and air pollution

    2021, Handbook on the Toxicology of Metals: Volume I: General Considerations
View all citing articles on Scopus
View full text