Source Apportionment of Ambient Volatile Organic Compounds from Petroleum and Non-Petroleum Emissions

Abstract

Volatile Organic Compounds (VOCs), precursors of photochemical smog and toxic substances, have attracted widespread attention of environmental communities for decades [1]. Petroleum products are often considered as the major sources of VOCs in the atmosphere. However, some VOCs may be emitted from non-petroleum products such as biogenic emissions and biomass burning. Hence, identification of emission sources and quantification of source contributions to ambient VOCs are prerequisite for the formulation and implementation of VOC related air pollution control measures and strategies. Source apportionment may also help to identify potential oilfield in a location through obtaining the concentrations of the tracers. Previous research suggests that high ozone (O 3 ) production rates in many locations are strongly associated with high anthropogenic VOC emissions [2-3]. Hence, effort has been made to identify and quantify source emissions of VOCs in different cities and regions so that photochemical air pollution could be controlled [4-5]. Receptor-oriented source apportionment models are the main tools for the source identification of pollutants and the estimation of source contributions to pollutant concentrations. The most widely used models are the Chemical Mass Balance (CMB) [4], Principal Component Analysis (PCA)/Absolute Principal Component Scores (APCS) [5], Positive Matrix Factorization (PMF) [6], and Graphical Ratio Analysis for Composition Estimates (GRACE)/Source Apportionment by Factors with Explicit Restriction (SAFER), incorporated in the UNMIX model [7]. Watson et al. [4] reviewed VOC source apportionment by CMB in more than 20 urban areas. Gasoline vehicle exhaust, liquid gasoline, and gasoline evaporation contributed over 50% of the ambient VOCs in many of these studies. Vega et al. [8] estimated VOC source apportionment in Mexico City in 1996–1997 and reported that vehicle exhaust made contribution of 54.9–63.8% and handling and distribution of Liquefied Petroleum Gas (LPG) 20.0–28.5%. Sosa et al. [9] re-conducted source apportionment in Mexico City using VOC data collected in 2000– 2001 and found that handling and distribution of LPG was the major source (42–52%), followed by vehicle exhaust (25–28%), asphalt works (12–14%) and cooking (5–10%). In Houston, refinery (26–35%), petrochemical and evaporative emissions (20–22%) and natural gas (13–17%) were the dominant sources of VOCs in 2003, while the major VOC sources in 2006 became natural gas/crude oil (32–39%), LPG (26%), fuel evaporation (20–23%) and vehicular exhausts (11–13%). All these studies have provided robust results for VOCs in the study areas.