Research article
Identification of diverse air pollution sources in a complex urban area of Croatia

https://doi.org/10.1016/j.jenvman.2019.04.024Get rights and content

Highlights

  • CBPF emerged as a reliable tool for source apportionment of gaseous pollutants and PM.

  • PMF is an integral and practical tool to determine pollution sources for PM.

  • An oil refinery was the dominant source of SO2 and H2S.

  • Emissions from city traffic and the highway were major NO2 sources.

  • Peak PM2.5 and NO2 values were observed during stable atmospheric conditions.

Abstract

Pinpointing the contribution of sources in complex urban areas, affected by large point sources such as oil refineries, is important for developing emission control strategies. Receptor models based on the chemical composition of particulate matter (PM), such as chemical mass balance (CMB) and positive matrix factorization (PMF), are useful means for source apportionment, but the inclusion of other gaseous pollutants need further consideration. The results of the multipollutant analyses using temporal variations in pollutant concentrations, chemical PM speciation and receptor modeling, PMF and conditional bivariate polar plots (CBPF), were used for determination of major pollutant sources of fine particulate matter (PM2.5) and less represented pollutants – hydrogen sulfide (H2S), nitrogen dioxide (NO2) and sulfur dioxide (SO2) in an urban area in Slavonski Brod, Croatia influenced by a large point source (an oil refinery) in Brod, Bosnia and Herzegovina. It is found that the composition of PM2.5 is dominated by carbonaceous combustion particles, mainly organic carbon (OC), with maximum values appearing during winter. Summer PM2.5 levels were dominated by sulfate and ammonium, which can be related to the industrial activities i.e., oil refinery. According to PMF analysis, the majority of OC is coming from biomass burning with ∼50% contribution to observed species concentration followed by ∼30% from industry/refinery and ∼10% from traffic. CPBF model showed that urban and highway traffic was the main source of NO2 concentrations while oil refinery was identified as the dominant source of SO2 and H2S. The CBPF receptor model combines concentrations of pollutants and meteorological parameters and emerged as a reliable complementary tool for the identification of sources for considered gaseous pollutants. Limitations of the CBPF method are in the application in stable atmospheric boundary layer conditions (SABL) as wind direction is not representative. Also, larger uncertainty is related to the representation of peak concentrations transported with higher wind speeds (>8  m/s) due to the lower number of events. This work uses various source apportionment methods in the assessment of PM but also for gaseous pollutants, such as NO2, SO2 and H2S that are less represented in the source apportionment studies and can be used for future scientific applications to assure more efficient air quality management.

Introduction

Identification of the main sources that significantly contribute to the levels of air pollution is a sensitive and crucial task for managing air pollution that requires advanced scientific methods and tools. The source apportionment approach is used to determine pollution sources and their associated contributions to ambient air pollution levels. Commonly used source apportionment techniques include emission inventories (Belis et al., 2014), inverse modeling (Schichtel et al., 2006), artificial neural networks (Reich et al., 1999; Dawidowski et al., 2001), receptor modeling methods (Hopke, 2008; Belis et al., 2013) and air quality models (Fragkou et al., 2012; Belis et al., 2014). Among those techniques, receptor models (RMs) are the most used method for the apportionment of air pollution sources, and these models rely on mass balance analysis (MBA) and positive matrix factorization (PMF) at a receptor site. RMs are mainly applied to particulate matter (Al-Dabbous and Kumar, 2015; Pereira et al., 2017), some are analysing volatile organic compounds (VOCs) Jorquera and Rappengluck, 2004; Leuchner and Rappengluck (2010) while their application to other gaseous pollutants is less represented in the literature (Leuchner and Rappengluck, 2010; Kumar et al., 2013; Sanaei et al., 2017). Different RMs have strengths and weaknesses, and their applicability to very reactive species is limited (Belis et al., 2013; Kumar et al., 2013; Cesari et al., 2016). Factor analysis models incorporating meteorological data are available to better illustrate source contributions associated with wind directions (Chan et al., 2011; Xiang et al., 2012; Argyropoulos et al., 2017). Hopke (2016) provided a review of the receptor modeling methods for source apportionment, mainly of aerosols, including methods that use local wind data: probability density function (PDF) and conditional probability function (CPF) (Kim and Hopke, 2004; Zhou et al., 2010), nonparametric regression (Kim and Hopke, 2004; Yu et al., 2004), and nonparametric wind regression (Henry et al., 2009). Yu et al. (2004) reported that SO2 can be used as a tracer for aircraft emissions, while its other sources can be separated based on wind direction and speed using nonparametric regression. Conditional bivariate polar plots (CBPF) are an advancement in the CPF receptor modeling method that shows how the concentration of a species varies jointly with wind speed and wind direction in polar coordinates. CBPF were successfully applied in a range of settings to characterize airport sources and dispersion characteristics in street canyons (Carslaw et al., 2006; Tomlin et al., 2009; Carslaw and Ropkins, 2012). However, without applying source apportionment techniques such as PMF or CMB, it is not possible to obtain reliable quantitative information on the contribution of specific sources. Therefore, a combination of different approaches might be able to provide holistic information about pollution sources in the complex urban/industrial areas.

Here, the city of Slavonski Brod in Croatia has been chosen since it is known for air quality problems (e.g., Jeričević et al., 2016) over an extended period and hence is in great need of air quality management (detailed information on air quality in Slavonski Brod are provided in Supplementary Information, SI, Section S2).

Urban areas with neighboring large industrial sources (and separated by the state border), such as the one studied in this work, are always challenging for air pollution management and emission control strategies (Kucbel et al., 2017; Li et al., 2017). Although point sources are generally quantitatively the largest emission sources, their contributions to the measured concentrations cannot be easily determined. Considering the substantial problems that exist in the city of Slavonski Brod, the aim of this work is to apply different source apportionment techniques in a polluted complex area with various emission sources, including an oil refinery. The source identification of both PM and much less reported gaseous pollutants here is carried out using (i) a comprehensive temporal analysis of measured pollutants and their relation to meteorological parameters over a four-year period between 2011 and 2014 in the city of Slavonski Brod, (ii) the analysis of quantified emissions from diverse urban sources in the city of Slavonski Brod and from refinery, (iii) the receptor modeling methods based on the CBPF and PMF, and (iv) chemical analysis of the PM2.5 composition during 2015.

Numerous studies have mainly applied RMs to PM. Here, we use standard PMF for PM and aim to further explore the applicability and properties of the CBPF receptor method for gaseous pollutants and investigate the importance of meteorological factors when added to receptor modeling to assess the ambient concentration in the chosen complex urban and industrial area.

Section snippets

Study area and monitoring data

The city of Slavonski Brod has a total population of approximately 60,000 people and is located on relatively flat terrain in eastern Croatia. Surrounded by Dilj Hill (461 m above ground level) to the north and Sava River to the south, the city is also located on the national border with Bosnia and Herzegovina (Fig. 1).

The urban monitoring site, Slavonski Brod (45.6° N and 17.59° E), is a part of a national monitoring network for continuous air quality monitoring maintained by the National

Relationship of diurnal and annual variations in pollutant concentrations with anthropogenic activities

Qualitative evaluation of the impact of anthropogenic sources to all the measured pollutants can be obtained by an understanding of (i) the temporal (e.g., diurnal and annual) variability of pollutant concentrations and (ii) the amount to which the pollutant concentrations follow human activity. It should be pointed that time series of pollutant concentrations may give a hint to sources to some extent, but they are largely modulated by the likely superposition of various sources of different

Discussion

According to the results of all analyses, major identified sources of air pollution in Slavonski Brod are biomass burning (domestic heating), industry (mainly refinery), traffic, and natural sources (resuspension and long-range transport). Meteorological conditions (SABL) that significantly affect the ambient level of pollutants are added as an important factor as in these conditions accumulation of concentration from different local sources is enhanced contributing to the occurrence of peaks

Summary and conclusions

We used a combination of different source apportionment approaches to identify the sources of PM2.5, H2S, NO2 and SO2 in a polluted complex urban area of Croatia that are important for successful air pollution assessment and management. In addition to CBPF and PMF receptor models, source apportionment was carried out using other methods such as (i) a comprehensive temporal analysis of measured pollutants and their relation to meteorological parameters over a four-year period between 2011 and

Acknowledgements

The Meteorological and Hydrological Service in Croatia and the Croatian Agency for Environment and Nature are thanked for providing the data. Prashant Kumar acknowledge the support received through the Research England funding under the Global Challenge Research Fund (GCRF) programme for the project CArE-Cities: Clean Air Engineering for Cities.

References (67)

  • R.M. Harrison et al.

    Field study of the influence of meteorological factors and traffic volumes upon suspended particle mass at urban roadside sites of different geometry

    Atmos. Environ.

    (2004)
  • H. Jorquera et al.

    Receptor modeling of ambient VOC at Santiago, Chile

    Atmos. Environ.

    (2004)
  • E. Kim et al.

    Comparison between conditional probability function and nonparametric regression for fine particle source directions

    Atmos. Environ.

    (2004)
  • M.L. Kreider et al.

    Physical and chemical characterization of tire-related particles: comparison of particles generated using different methodologies

    Sci. Total Environ.

    (1 January 2010)
  • M. Kucbel et al.

    Temporal and seasonal variations of black carbon in a highly polluted European city: apportionment of potential sources and the effect of meteorological conditions

    J. Environ. Manag.

    (2017)
  • P. Kumar et al.

    Nanoparticle emissions from 11 non-vehicle exhaust sources - a review

    Atmos. Environ.

    (2013)
  • M. Leuchner et al.

    VOC source-receptor relationships in Houston during TexAQS-II

    Atmos. Environ.

    (2010)
  • Y. Li et al.

    Monitoring and source apportionment of trace elements in PM2.5: Implications for local air quality management

    J. Environ. Manag.

    (2017)
  • A.R. Malby et al.

    Conditional extraction of air-pollutant source signals from air-quality monitoring

    Atmos. Environ.

    (2013)
  • K. Na et al.

    Primary and secondary carbonaceous species in the atmosphere of western riverside county, California

    Atmos. Environ.

    (2004)
  • C. Pio et al.

    OC/EC ratio observations in Europe: Re-thinking the approach for apportionment between primary and secondary organic carbon

    Atmos. Environ.

    (2011)
  • S.L. Reich et al.

    Artificial neural network for the identification of unknown air pollution sources

    Atmos. Environ.

    (1999)
  • S. Prateek S. Sharma et al.

    Statistical behavior of ozone in urban environment

    Sustainable Environment Research

    (2016)
  • R. Strader et al.

    Evaluation of secondary organic aerosol formation in winter

    Atmos. Environ.

    (1999)
  • A.S. Tomlin et al.

    A field study of factors influencing the concentrations of a traffic-related pollutant in the vicinity of a complex urban junction

    Atmos. Environ.

    (2009)
  • I. Uria-Tellaetxe et al.

    Conditional bivariate probability function for source identification

    Environ. Model. Softw

    (2014)
  • M. Viana et al.

    Source apportionment of particulate matter in Europe: a review of methods and results

    Aerosol Science

    (2008)
  • Y. Xiang et al.

    Development of a methodology examining the behaviours of VOCs source apportionment with micro-meteorology analysis in an urban and industrial area

    Environ. Pollut.

    (2012)
  • K.N. Yu et al.

    Identifying the impact of large urban airports on local air quality by nonparametric regression

    Atmos. Environ.

    (2004)
  • A.N. Al-Dabbous et al.

    Source apportionment of airborne nanoparticles in a Middle Eastern city using positive matrix factorization

    Environ. Sci.: Processes & Impacts

    (2015)
  • C.A. Belis et al.

    European Guide on Air Pollution Source Apportionment with Receptor Models

    (2014)
  • D.C. Carslaw

    The OpenAir Manual – Open-Source Tools for Analysing Air Pollution Data

    (2012)
  • V. Celo et al.

    Concentration and source origin of trace metals in PM2.5 collected at selected Canadian sites within the Canadian national air pollution surveillance program

    Urban Airborne Particulate Matter: Origin, Chemistry, Fate and Health Impacts

    (2011)
  • Cited by (30)

    • Exhaust emission characteristics study of light and heavy-duty diesel vehicles in India

      2022, Case Studies in Thermal Engineering
      Citation Excerpt :

      Other pollutants such as NH3, PAHs, dioxins and furans have been shown to have been increased gradually, since 1999 onwards. This has been due to the unregulated and unchecked vehicular emissions by the authority and hence there has been increase in pollutants from vehicular exhaust [11]; Yildez et al., 2019; [12–18]. The impacts of new European driving cycle on the vehicular emissions have been studied using biodiesel-diesel blends on a EURO 5 compatible vehicle.

    • An estimation method for regional transport contributions from emission sources based on a high-mountain site: a case study in Zhumadian, China

      2021, Atmospheric Environment
      Citation Excerpt :

      Thus, it is important to estimate the contributions of regional pollutant transport to accurately characterize the impacts of local emission sources (Diamantopoulou et al., 2016; Skyllakou et al., 2014; Zong et al., 2018) and thereby develop effective control strategies. Currently, regional transport analysis methods to determine the contributions to ambient PM2.5 at a given location were mainly air quality models (Li et al., 2019b; Wang et al., 2018, 2019; Wei et al., 2019), coupling field observations and backward trajectory analysis (Molnar et al., 2017; Tshehla and Djolov, 2018; Turap et al., 2019), potential source contribution function (PSCF) combined with the positive matrix factorization (PMF) apportionments (Liu et al., 2017; Nicolás et al., 2011; Zhang et al., 2019; Zong et al., 2018), other approaches of the combination of backward trajectories and PMF source apportionments (Liu et al., 2020b; Tian et al., 2015), and the conditional bivariate probability function (CBPF) (or CPF) with PMF contributions (Jain et al., 2020; Jericevic et al., 2019; Morera-Gómez et al., 2018; Rai et al., 2016). However, the accuracy of emission inventories generally limits the accuracy of the simulation results when using air quality models to estimate regional transport contributions (An et al., 2013; Bergin et al., 1999).

    • Microbial degradation of pyrene in holm oak (Quercus ilex) phyllosphere: Role of particulate matter in regulating bioaccessibility

      2021, Science of the Total Environment
      Citation Excerpt :

      The highest air pollution concentrations are generally found in urban areas with high population density. The pollutants involved could be either organic (e.g., polycyclic aromatic hydrocarbons, PAHs) or inorganic chemicals (e.g., O3, SO2, NOx, CO), as well as particulate matter (PM) (Cichowicz and Stelęgowski, 2019; Jeričević et al., 2019; Madruga et al., 2019). Improving the air quality in cities is therefore one of the main challenges for the European Union (EU) (EEA, 2019) and in general for many countries.

    • Improving source inversion performance of airborne pollutant emissions by modifying atmospheric dispersion scheme through sensitivity analysis combined with optimization model

      2021, Environmental Pollution
      Citation Excerpt :

      Ambient air pollution caused by accidental or furtive emissions has aroused widespread concerns (Bildirici, 2017; Jeričević et al., 2019).

    View all citing articles on Scopus
    View full text