Primary Particulate Matter from Ocean-Going Engines in the

High levels of V and Ni were observed from in-stack emission measurements conducted on the propulsion engines of two different in-use OGVs. The in-sta...
0 downloads 9 Views 1MB Size
Download PDF
Environ. Sci. Technol. 2009, 43, 5398–5402

Primary Particulate Matter from Ocean-Going Engines in the Southern California Air Basin H A R S H I T A G R A W A L , †,‡ R U D Y E D E N , § XINQIU ZHANG,§ PHILIP M. FINE,§ AARON KATZENSTEIN,§ J . W A Y N E M I L L E R , †,‡ J E A N O S P I T A L , § SOLOMON TEFFERA,§ AND D A V I D R . C O C K E R I I I * ,†,‡ Department of Chemical and Environmental Engineering, University of California, Bourns Hall A321, Riverside, California 92521, Center for Environmental Research and Technology, College of Engineering, University of California, 1084 Columbia Avenue, Riverside, California 92507, and South Coast Air Quality Management District, 21865 Copley Drive, Diamond Bar, California 91765

Received December 10, 2008. Revised manuscript received April 6, 2009. Accepted May 21, 2009.

The impact of primary fine particulate matter (PM2.5) from ship emissions within the Southern California Air Basin is quantified by comparing in-stack vanadium (V) and nickel (Ni) measurements fromin-useocean-goingvessels(OGVs)withambientmeasurements made at 10 monitoring stations throughout Southern California. V and Ni are demonstrated as robust markers for the combustion of heavy fuel oil in OGVs, and ambient measurements of fine particulateVandNiwithinSouthernCaliforniaareshowntodecrease inversely with increased distance from the ports of Los Angeles and Long Beach (ports). High levels of V and Ni were observed from in-stack emission measurements conducted on the propulsion engines of two different in-use OGVs. The in-stack V and Ni emission rates (g/h) normalized by the V and Ni contents in the fuel tested correlates with the stack total PM emission rates (g/h). The normalized emission rates are used to estimate the primary PM2.5 contributions from OGVs at 10 monitoring locations within Southern California. Primary PM2.5 contributions from OGVs were found to range from 8.8% of the total PM2.5 at the monitoring location closest to the port (West Long Beach) to 1.4% of the total PM2.5 at the monitoring location 80 km inland (Rubidoux). The calculated OGV contributions to ambient PM2.5 measurements at the 10 monitoring sites agree well with estimates developed using an emission inventory based regional model. Results of this analysis will be useful in determining the impacts of primary particulate emissions from OGVs upon worldwide communities downwind of port operations.

1. Introduction The health effects of particulate matter emitted from diesel engines is a matter of significant concern (1). Further, Corbett et al. (2) recently indicated that shipping-related PM emis* Corresponding author phone: (951) 827-2408; fax: (951) 7815790; e-mail: [email protected]. † Department of Chemical and Environmental Engineering, University of California. ‡ Center for Environmental Research and Technology, University of California. § South Coast Air Quality Management District. 5398

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 14, 2009

sions are responsible for approximately 60 000 cardiopulmonary and lung cancer deaths annually. Southern California encompasses two of the world’s major ports, the Port of Los Angeles (POLA) and the Port of Long Beach (POLB), which combined form the fifth-largest port complex in the world (3). The Southern California ports are responsible for 32% of the containers moving in and out of the United States (3). In addition, these ports are expecting growth of about 30 million twenty-foot equivalent units (TEUs) (150%) of containerized cargo over the next 20 years (4). This increase in global trade has made emissions from ocean-going vessels (OGVs) a significant and growing source of gaseous and particulate emissions to the local and global environment (2, 5-13). The Southern California Air Basin (SoCAB) exceeded the federal ozone and particulate matter (PM2.5) standards on a total of 89 days at one or more locations in the year 2005 (South Coast Air Quality Management District (SCAQMD), 2007). A recent study (BTH and Cal/EPA, 2005) shows that port-related sources are expected to become the largest source of emissions in the SoCAB by the year 2020. An accurate assessment of the impact of PM from ocean-going engines on air quality downwind is of interest to communities close to port complexes around the world (13-17). The present study aims to assess the impact of emissions from OGVs in the SoCAB by using V and Ni as tracers for the combustion of heavy fuel oil in OGVs. The SoCAB is a highly urbanized area with a population of about 16 million. Many studies have discussed the contribution of different sources on regional air quality in Southern California (18-22). Recently, Vutukuru et al. (13) has illustrated the use of regional air quality models based upon emission inventories to assess the impact of ships in the SoCAB. They estimate a significant increase in concentrations of particulate nitrate and sulfate concentrations (12.8 and 1.7 µg m-3, respectively) in the SoCAB when ship emissions are included. The main propulsion engines on the OGVs run on heavy fuel oil (HFO) also known as residual oil or bunker oil. Heavy fuel oil is rich in trace elements such as V and Ni naturally present in crude oil that become concentrated in the heavy fuel during the refining process (23). V and Ni have been associated with combustion of heavy fuel oil for source apportionment analysis (24, 25). These metals are enriched in the fine particles emitted from the combustion of heavy fuel oil (26). Recently, Querol et al. (27) have observed high V and Ni levels at a site close to heavy ship traffic and hypothesized the possibility of V and Ni as tracers of ship emissions. However, the contribution of ship emissions to primary PM2.5 has not been directly determined using OGV source tests and fine particulate ambient measurements. The only other potential source of heavy fuel oil combustion emissions is power plants (24, 27, 28). Southern California is characterized by a lack of obvious V sources, such as coal/ oil-fueled power plants (29). Thus, ships running on heavy fuel oil are identified as the only significant source of V in the SoCAB. This study for the first time combines in-stack emission measurements from OGVs with ambient measurements made at 10 monitoring locations in Southern California during the Multiple Air Toxics Exposure Study (MATES III) (30). The study is extended to include the dispersion and chemical mass balance modeling platforms to evaluate the results. These data provide an assessment of the impact of primary aerosol from marine engines within the SoCAB. 10.1021/es8035016 CCC: $40.75

 2009 American Chemical Society

Published on Web 06/03/2009

TABLE 1. Location of Monitoring Sites site

latitude

longitude

distance from ports (km)

point source West Long Beach, 1903 Santa Fe Ave. North Long Beach, 3648 N. Long Beach Blvd. Compton, 720 N. Bullis Rd. Anaheim, 1010 S. Harbor Blvd. Huntington Park, 6301 S. Santa Fe Ave. Pico Rivera, 3713 B-San Gabriel River Parkway Los Angeles, 1630 N. Main St. Burbank, 228 W. Palm Ave. Fontana, 14360 Arrow Highway Rubidoux, 5888 Mission Blvd.

33°,38′,48.14′′ N 33°,47′,30.38′′ N 33°,49′,24.70′′ N 33°, 54′,1.44′′ N 33°,49′,13.86′′ N 34°,0′,36.04′′ N 34°,0′,48.86′′ N 34°,4′,1.64′′ N 34°,10′,35.13′′ N 34°,5′,57.42′′ N 34°,0′,3.45′′ N

118°,12′,30.82′′ W 118°12′55.98′′ W 118°,11′,21.54′′ W 118°,12′,24.67′′ W 117°,54′,54.03′′ W 118°,13′,49.19′′ W 118°,3′,34.64′′ W 118°,13′,37.02′′ W 118°,19′,0.40′′ W 117°,29′,32.12′′ W 117°,24′,54.53′′ W

9.7 13.0 21.6 29.1 34.0 35.2 40.2 53.4 78.5 79.5

2. Approach/Methodology The emissions from OGVs have a significant impact on regional air quality. The approach is based on combining the results from in-stack measurements conducted on two different OGVs and results from ambient monitoring of metals and PM2.5 at 10 sites in the SoCAB to estimate the primary PM2.5 from ships at each of the sites. The following sections describe the methodology in detail. 2.1. Ambient Monitoring of Metals and PM2.5. In 2004, the SCAQMD started an intensive ambient air toxics monitoring program, MATES III. The MATES III study covered a period of two years from April 2004 to March 2006, providing data on air toxic levels and exposures, addressing environmental justice issues, establishing an updated toxic emission inventory, and determining cancer risks from air toxics in the SoCAB. As a part of the MATES III monitoring program ambient measurements of PM2.5 and metals were conducted at 10 monitoring locations. The locations of these 10 geographically dispersed monitoring sites are provided in Table 1 and Figure 1. Ambient sampling at each location followed a one-in-three day 24 h integrated sampling schedule from April 2004 to March 2006. The SCAQMD laboratory conducted the mass and metals analysis. Detailed protocols are described in the MATES III report (30). Briefly, PM2.5 samples were collected over a 24 h period using SASS speciation samplers (MetOne, OR). All PM2.5 samples were collected on Teflo filters and are analyzed for total PM2.5 mass and metals. Weighing procedure

guidelines followed the Code of Federal Regulations (31). The metal analysis for particulate samples was determined using energy-dispersive X-ray fluorescence spectrometer following the EPA IO-3 method. The MATES III study also involved regional modeling of known toxic compounds and sources, which include OGV emissions, to determine their impact within Southern California communities. The Comprehensive Air Quality Model with Extensions (CAMx) provided the dispersion modeling platform and the atmospheric chemistry to simulate annual impacts of both gaseous and aerosol toxic compounds on communities in the SoCAB. The modeling was conducted using current emission inventories as input and a grid size of 2 km2 that encompassed the SoCAB and coastal shipping lanes. In addition, a chemical mass balance model (EPA CMB 8.2) was applied to ambient speciated PM2.5 data collected at the 10 monitoring locations. Several different source contributions were determined using monthly measurement data and source profiles; this included ship emissions based primarily upon the Ni and V contents of the filters. At the time the modeling was conducted, a ship emission profile was not available. Therefore, the profile of a stationary boiler burning heavy fuel oil was used as a surrogate for a ship emission profile. Complete descriptions and modeling results are described in the MATES III report (30). 2.2. In-Stack Measurement of Metals and PM2.5. In-stack emission measurements were conducted on the propul-

FIGURE 1. Location of different monitoring sites in the South Coast Air Basin. VOL. 43, NO. 14, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5399

TABLE 2. Engine Description vessel class

manufacturer/model

power, kW

technology

max rpm

Suezmax Panamax

Sulzer 6RTA72 MAN B&W 11K90MC

15750 50270

two-stroke two-stroke

90 104

TABLE 3. Selected Fuel Specifications

fuel type density (kg m-3) at 15 °C viscosity (mm2 s-1) at 50 °C micro carbon residue content (%, m/m) sulfur content (%, m/m) ash content (%, m/m) V content (mg/kg) Ni content (mg/kg)

Suezmax

Panamax

heavy fuel oil 989.7 230.6 13

heavy fuel oil 990.8 296.8 14.5

2.85 0.04 118 52

2.05 0.072 259 26

sion engines of two different in-use OGVs on sea voyages under typical operating conditions. The OGVs tested were a Suezmax class vessel (large ship dimensions still capable of navigating the Suez Canal) and a Panamax class vessel (maximum dimension of a ship allowed through the locks of the Panama Canal) (32, 33). The engines for both vessels operate on a two-stroke cycle and were running on heavy fuel oil when tested; engine and fuel specifications for the vessels are provided in Tables 2 and 3. Emission tests were conducted in triplicate following the modes for the ISO certification cycle (ISO 8178-4). A complete description of comprehensive measurements for both engines is presented elsewhere (32, 33). The mass, metals, and ions collected on PM2.5 filters from the in-stack measurements were acquired by analysis of particulates collected on 47 mm diameter 2 µm pore Teflo filters (Pall Gelman, Ann Arbor, MI). The filters were measured for net mass gain using a Cahn C-35 (Madison, WI) microbalance following the weighing procedure guidelines of the Code of Federal Regulations (31). Before and after collection, the filters were conditioned for 24 h in an environmentally controlled room (RH ) 40%, T ) 25 °C) and weighed daily until two consecutive weight measurements were within 3 µg. The Teflo filters were subsequently analyzed for metals using a PANalytical energy-dispersive X-ray fluorescence spectrometry using the same instrument and method mentioned for ambient filter in section 2.1.

3. Results and Discussion The following sections describe the results of the in-stack measurements from OGVs and the ambient measurement at ground sites described in Figure 1. 3.1. In-Stack Measurement of Metals and PM2.5. The mass balance of V and Ni collected from in-stack filter samples was determined using the brake-specific fuel consumption of each vessel combined with the V/Ni content in the fuel. This determined the V/Ni emissions expected to be present in the stack emissions on the basis of fuel usage. These estimated V/Ni emissions from the fuel usage were correlated with the actual V/Ni measured on filters collected during the in-stack measurements (Figure 2). Good correlation between the expected V/Ni main engine emissions compared with measured values shows the majority of the V/Ni from the fuel is emitted as particulates from the stack. Additionally, it is observed that the V to Ni ratio in the vessel fuel is preserved in the exhaust on the single Suezmax vessel tested (Figure S3, Supporting Information). The source tests for different engines on OGVs demonstrate a very consistent ratio of PM2.5 emissions to V and Ni 5400

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 14, 2009

FIGURE 2. V and Ni mass balance between the fuel and filters collected from the main engine exhaust in relation to brake-specific fuel consumption.

FIGURE 3. Emission rates of PM and V/Ni normalized by the V/ Ni content in the fuel of the ship being tested. emissions when normalized by the V and Ni contents in the fuel being consumed (Figure 3). This ratio can be utilized to estimate the primary PM2.5 at the regional sites simply by measuring V and Ni concentrations at the monitoring locations and knowing the average V or Ni content of the fuel in use. The V and Ni contents in bunker fuel is known to vary greatly from batch to batch and from different origins of the crude oil (17, 23, 34). This variability of Ni and V in bunker fuel was addressed in the current study by taking the average Ni and V contents of bunker fuel from six different OGVs. From this fuel survey an average of 100 ( 47 ppm V and an average of 35 ( 18 ppm Ni in the heavy fuel oil were used. The average V from the survey is consistent with the data gathered by DNV, which are described in the Supporting Information. These average V and Ni fuel concentrations along with the ratio developed from in-stack sampling (Figure 3) are used in this study to estimate the primary OGV PM2.5 at the monitoring locations. 3.2. Ambient Measurement of Metals and PM2.5. The presence of high levels of Ni and V in heavy fuel oil provided robust tracers of emissions from OGVs. The average V and Ni concentrations collected in the PM2.5 samples collected over two years during the MATES III project increase almost linearly with the inverse of the distance from the ports (Figure 4). Considering this result along with the typical on-shore wind direction suggests strongly that the port complex is the only major regional source of V and Ni in the Southern California Air Basin. This observation should be consistent for any coastal area without coal- or oil-based power plants. While V and Ni ambient concentrations correlate well with the inverse of the distance from the ports, other elements found in OGV emissions such as calcium and sulfur do not. Additionally, Figure 4 shows that V is a more robust marker (R2 ) 0.96) than Ni (R2 ) 0.85). Thus, only results from V as the tracer are described in this paper. The results using Ni as the tracer are provided in the Supporting Information. 3.3. Estimation of Primary PM2.5 from Shipping in the SoCAB. Using the emission factor ratios (r) determined in

r)

FIGURE 4. V and Ni dispersion near the ports. Figure 3 of 8205 for V and 8046 for Ni and applying them to the average ambient V and Ni contents in the ambient air samples yields the estimated primary PM2.5 contribution from OGVs as shown in the following equation (for the case of V): PMa ) 〈r〉

Va

〈FV,HFO〉

(1)

where PMa ) primary OGV PM2.5 concentration estimate (µg/ m3), Va ) ambient V concentration (µg/m3), 〈FV,HFO〉 ) average V content of the heavy fuel oil from all vessels (ppm), and 〈r〉 ) average ratio of PM2.5 to normalized V emitted (ppm) ≈ slope of the line in Figure 3. For any individual source test, this ratio is given by

EPM,stack EV,stack/FV,HFO

(2)

where EPM,stack ) in-stack PM2.5 emissions (g/h), EV,stack ) instack particulate V emissions (g/h), and FV,HFO ) V content of heavy fuel oil of the vessel being tested (ppm). The resulting primary PM2.5 estimates from OGVs at the 10 monitoring sites within the SoCAB using the V emission factor are shown in Table 4. Average primary PM2.5 from ships is highest at the Wilmington monitoring station (1.61 ( 0.28 µg/m3, 8.8% of the total PM2.5) located closest to the ports. The primary emissions from ships have an impact on the areas up to 80 km from the coast (0.32 ( 0.05 µg/m3, 1.4% of the total PM2.5). 3.4. Evaluation of the Primary PM2.5 Estimates with Estimates from Regional Air Quality Models. The OGV contribution to ambient measurements agrees well with the estimates made during the MATES III study by both chemical mass balance modeling and CAMx. OGV contribution estimates determined using the CAMx model are solely determined from emission inventories and transport calculations which are entirely independent of the monitoring and emissions data used in this study. Figure 5 shows reasonable agreement between the CAMx modeling results and CMB modeling results with the ratio method results presented here. CAMx results appear to be approximately 20% higher than the ratio method, but given the uncertainties in the emissions inventories assumed in CAMx, this small difference and the high level of correlation are remarkable. This suggests that both independent methods yield reasonably accurate results and that the ratio method may provide a quicker estimate of primary PM2.5 OGV impacts than resource-intensive modeling runs. It appears that the CMB analysis slightly underestimates OGV contributions (Figure 5). This underestimate was likely the result of using a stationary residual fuel boiler profile and not having an appropriate OGV source profile at the time the CMB analysis was performed.

FIGURE 5. Comparison of primary PM2.5 estimates in this study with CAMx and CMB modeling for the year 2005.

TABLE 4. Primary PM2.5 Estimates distance from ports (km)

measured annual average V emission (ng/m3)

primary PM2.5 emission from ships, this study (µg/m3)

measured annual average total PM2.5 emission (µg/m3)

percentage of total PM2.5 emission, this study

primary PM2.5 emission estimate, CMB (µg/m3)

primary PM2.5 emission estimate, CAMx (µg/m3)

9.7 13.0 21.6 29.1 34.0 35.2 40.2 53.4 78.5 79.5

19.47 ( 3.24 11.45 ( 1.91 7.48 ( 1.25 6.83 ( 1.14 5.82 ( 0.97 5.29 ( 0.88 4.71 ( 0.78 4.17 ( 0.69 3.70 ( 0.62 3.85 ( 0.64

1.61 ( 0.28 0.95 ( 0.16 0.62 ( 0.11 0.57 ( 0.10 0.48 ( 0.08 0.44 ( 0.08 0.39 ( 0.07 0.35 ( 0.06 0.31 ( 0.05 0.32 ( 0.05

18.35 ( 3.06 17.84 ( 2.97 18.87 ( 3.14 17.55 ( 2.92 21.36 ( 3.56 20.87 ( 3.47 18.70 ( 3.11 20.94 ( 3.49 21.51 ( 3.58 22.88 ( 3.81

8.80 5.32 3.29 3.22 2.26 2.10 2.09 1.65 1.42 1.40

1.3 0.82 0.51 0.52 0.43 0.39 0.33 0.29 0.28 0.27

1.70 1.02 0.46 0.66 0.21 0.22 0.16 0.11 0.15 0.17

VOL. 43, NO. 14, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5401

The ratio method described and evaluated above provides a relatively simple method for estimating primary OGV contributions to ambient PM2.5 levels if no other major sources of V are present. The data presented above suggest that the source tests conducted in this study are indicative of PM2.5 emission rates from HFO-burning OGVs in general and that the V in the HFO is conserved in the emissions. Thus, the average normalized ratio, 〈r〉, derived in this paper could be universally applied to other locations with HFO-burning OGV emissions. Then, using eq 1, a measurement of ambient V levels combined with an estimate of the average V content of in-use HFO will yield an estimate of the primary contribution of OGVs to any given PM2.5 sample.

Acknowledgments This study would have not been possible without the analytical support from Ms. Kathy Cocker, Ms. Varalakshmi Jayaram, and Dr. Abhilash Nigam, Funding support from CARB and generous helping hands on the ship from the shipping company crew. We thank Dr. Aniket Sawant and Dr. Kanok Boriboonsomsin for useful discussions.

Note Added after ASAP Publication The Acknowledgments were modified in the version of this paper published ASAP June 3, 2009; the corrected version published ASAP June 5, 2009.

Supporting Information Available Text and figures describing the uncertainty analysis for PM2.5 estimates and comparison of PM2.5 estimates from V and Ni. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Pope, C. A.; Burnett, R. T.; Thun, M. J.; Calle, E. E.; Krewski, D.; Ito, K.; Thurston, G. D. Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. JAMA, J. Am. Med. Assoc. 2002, 287 (9), 1132–1141. (2) Corbett, J. J.; Winebrake, J. J.; Green, E. H.; Kasibhatla, P.; Eyring, V.; Lauer, A. Mortality from ship emissions: A global assessment. Environ. Sci. Technol. 2007, 41, 8512–8518. (3) American Association of Port Authorities (AAPA). Port Industry Statistics, 2008. AAPA Web site. http://www.aapa-ports.org/ home.cfm, accessed August 8, 2008. (4) California Marine and Intermodal Transportation System Advisory Council (CALMITSAC). Growth of California Ports: Opportunities and Challenges; Los Angeles, 2006. (5) Corbett, J. J.; Fischbeck, P. S. Emissions from ships. Science 1997, 278, 823–824. (6) California Air Resources Board. Goods movement action plansphase I, 2007. California Air Resources Board Web site. http://www.arb.ca.gov/gmp/docs/finalgmpplan090205.pdf, accessed March 25, 2008. (7) Cofala, J.; Amann, M.; Heyes C.; Klimont, Z.; Posch, M.; Scho¨pp, W.; Tarasson, L.; Jonson, J. E.; Whall, C.; Stavrakaki, A. Final Report: Analysis of Policy Measures To Reduce Ship Emissions in the Context of the Revision of the National Emissions Ceilings Directive; International Institute for Applied Systems Analysis: Laxenburg, Austria, 2007; p 74. (8) Cooper, D. A. Exhaust emissions from ships at berth. Atmos. Environ. 2003, 37, 3817–3830. (9) Capaldo, K.; Corbett, J. J.; Kasibhatla, P.; Fischbeck, P. S.; Pandis, S. N. Effects of ship emissions on sulfur cycling and radiative climate forcing over the ocean. Nature 1999, 400, 743–746. (10) Endresen, Ø.; Sørgard, E.; Sundet, J. K.; Dalsøren, S. B.; Isaksen, I. S. A.; Berglen, T. F. Emissions from international sea transportation and environmental impact. J. Geophys. Res. 2003, 108, 4560–4578. (11) Streets, D. G.; Guttikunda, S. K.; Carmichael, G. R. The growing contribution of sulfur emissions from ships in Asian waters 1988-1995. Atmos. Environ. 2000, 34 (26), 4425–4439. (12) Lawrence, M. G.; Crutzen, P. J. Influence of NOx emissions from ships on tropospheric photochemistry and climate. Nature 1999, 402, 167–170. (13) Vutukuru, S., Dabdub D. Modeling the effects of ship emissions on coastal air quality: A case study of Southern California. Atmos. Environ. 2008, 42 (16), 3751-3764.

5402

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 14, 2009

(14) Deniz, C.; Durmusoglu, Y. Estimating shipping emissions in the region of the Sea of Marmara, Turkey. Sci. Total Environ. 2008, 390, 255–261. (15) Li, Q. Transatlantic transport of pollution and its effects on surface ozone in Europe and North America. J. Geophys. Res. 2002, 107, 4166–4179. (16) Skjølsvik, K. O.; Andersen, A. B.; Corbett, J; Skjelvik, J. M. Study on Greenhouse Gas Emissions from Ships; MT Report Mtoo A23038; MARINTEK: Trondheim, Norway, 2000. (17) Church, T. M.; Arimoto, R.; Barrie, L. A.; Dehairs, F.; Dulac, F.; Jickells, T. D.; Mart, L.; Sturges, W. T.; Zoller, W. H. The longrange atmospheric transport of trace elements: A critical evaluation. In The Long-Range Atmospheric Transport of Natural and Contaminant Substances; Knap, A. H., Ed.; Kluwer Academic Publishers: Dordrecht, The Netherlands, 1988; pp 37-58. (18) Rodriguez, M. A.; Brouwer, J.; Samuelsen, G. S.; Dabdub, D. Air quality impacts of distributed power generation in the South Coast Air Basin of California 2: Model uncertainty and sensitivity analysis. Atmos. Environ. 2007, 41, 5618–5635. (19) Carreras-Sospedra, M.; Dabdub, D.; Rodriguez, M.; Brouwer, J. Air quality modeling in the South Coast Air Basin of California: What do the numbers really mean? J. Air Waste Manage. Assoc. 2006, 56 (8), 1184–1195. (20) Kleeman, M. J.; Hughes, L. S.; Allen, J. O.; Cass, G. R. Source contribution to the size and composition distribution of atmospheric particles: Southern California in Septemeber 1996. Environ. Sci. Technol. 1999, 33, 4331–4341. (21) Hughes, L. S.; Allen, J. O.; Kleeman, M. J.; Johnson, R. J.; Cass, G. R.; Gross, D. S.; Gard, E. E.; Galli, M. E.; Morrical, B. D.; Fergenson, D. P.; Dienes, T.; Noble, C. A.; Silva, P. J.; Prather, K. A. Size and composition distribution of atmospheric particles in Southern California. Environ. Sci. Technol. 1999, 33, 3506–3515. (22) Russell, A. G.; McCue, K. F.; Cass, G. R. Mathematical-modeling of the formation of nitrogen-containing air-pollutants 0.1. Evaluation of an eulaerian photochemical model. Environ. Sci. Technol. 1988, 22, 263–270. (23) Nelson, W. L. How much metals on crude oils? Oil Gas J. 1972 (Aug 7). (24) Olmez, I.; Sheffield, A. E.; Gordon, G. E.; Houck, J. E.; Pritchett, L. C.; Cooper, J. A.; Dzubay, T. G.; Bennett, R. L. Compositions of particles from selected sources in Philadelphia for receptor modeling applications. J. Air Pollut. Control Assoc. 1988, 38, 1392–1402. (25) Vallius, M.; Lanki, T.; Tittanen, P; Koistinen, K.; Ruuskanen, J.; Pekkanen, J. Source apportionment of urban ambient PM2.5 in two successive measurement campaigns in Helsinki, Finland. Atmos. Environ. 2003, 37, 615–623. (26) Jang, H. N.; Seo, Y. C.; Lee, J. H.; Hwang, K. W.; Yoo, J. I.; Sok, C. H.; Kim, S. H. Formation of fine particles enriched by V and Ni from heavy fuel oil combustion: Anthropogenic sources and drop-tube furnace experiments. Atmos. Environ. 2007, 41, 1053–1063. (27) Querol, X.; Vianaa, M.; Alastueya, A.; Amatoa, F.; Morenoa, T.; Castilloa, S.; Peya, J.; Rosab, J.; Sa´nchez de la Campab, A.; Artı´n ˜anoc, B.; Salvadorc, P.; Garcı´a Dos Santosd, S.; Ferna´ndez-Patierd, R.; Moreno-Graue, S.; Negrale, L.; Minguillo´naf, M. C.; Monfortf, E.; Gilg, J. I.; Inzag, A.; Ortegag, L. A.; Santamarı´ah, J. M.; Zabalzahet, J. Source origin of trace elements in PM from regional background, urban and industrial sites of Spain. Atmos. Environ. 2007, 41, 7219–7231. (28) Gao, N.; Cheng, M. D.; Hopke, P. K. Receptor modeling for airborne ionic species collected in SCAQS, 1987. Atmos. Environ. 1994, 28, 1447–1470. (29) South Coast Air Quality Management District (SCAQMD). Emissions from diesel combustion by major source category. Air Quality Management Plan; South Coast Air Quality Management District: Diamond Bar, CA, 2007; Appendix III, Attachment F. (30) South Coast Air Quality Management District (SCAQMD). Multiple Air Toxics Exposure Study in the South Coast Air Basin (MATES III); South Coast Air Quality Management District: Diamond Bar, CA, 2008. (31) Protection of the Environment. Code of Federal Regulations, 40 CFR 86. (32) Agrawal, H.; Malloy, Q.; Welch, W. A.; Miller, J. W.; Cocker, D. R. In-usegaseousandparticulatematteremissionsfromamodernocean going container vessel. Atmos. Environ. 2008a, 42, 5504–5510. (33) Agrawal, H.; Welch, W. A.; Miller, J. W.; Cocker, D. R. Emission measurements from a crude oil tanker at sea. Environ. Sci. Technol. 2008b, 42, 7098–7103. (34) Zaki, S. N.; Barbooti, M. M.; Baha-uddin, S. S.; Hassan, E. B. Determination of trace metals and their distribution in heavy crude oil distillates (350° C +) by atomic absorption spectroscopy. Appl. Spectrosc. 1989, 43, 7–15.

ES8035016