Original Article
Chemical speciation of PM2.5 in Tehran: Quantification of dust contribution and model validation

https://doi.org/10.1016/j.apr.2020.07.020Get rights and content

Highlights

  • BSC-DREAM8b and NMMB/BSC-Dust were evaluated using chemical speciation of PM.

  • Monthly average modeled and observed data agree well.

  • There was a poorer agreement for daily modeled data and observations.

  • Agreement between models and observations was sensitive to sampling frequency.

  • Spatial variability between grid cells had no significant effect on model-observation correlation.

Abstract

Each year, considerable levels of dust particles coming from arid regions of the earth contribute to the atmosphere. Because dust has serious environmental and human health effects, predictions of dust concentrations and their contribution to PM surface concentrations are essential for atmospheric research and the implementation of air quality programs and rules. This study aims to assess dust contributions to PM2.5 in Tehran in order to provide guidance for air quality management systems as well as validate the updated Dust Regional Atmospheric Model version 2 (BSC-DREAM8b) and the NMMB/BSC-Dust model using chemical speciation of ground-based measurements of PM2.5. Accurate and reliable measurements are necessary to determine the sources of pollutants as well as to confirm, validate, and improve models. For this purpose, dust concentration and contribution were calculated from chemical speciation of PM2.5 samples that were obtained on a 6 day basis for a whole year from February 2014 to February 2015. The comparison of the dust observations to the model products revealed that model outputs were in good agreement for the monthly averaged values. However, for the daily averaged values, models and ground observation showed a considerable difference. In order to understand possible reasons for the discrepancy, differences between neighboring grid cells were investigated, as well as the effect of sampling frequency. It is found that agreement between models and observations was sensitive to sampling duration and frequency but not to spatial variability between grid cells. This study is the first validation of model outputs with the calculated dust concentration from ground-based chemical speciation of PM2.5 rather than using total PM or aerosol optical depth measurements as a proxy for dust concentrations.

Introduction

The majority of the world's population lives in metropolitan areas. Consequently, air pollution, and more specifically particulate matter (PM), has become a critical problem for human health (Anderson et al., 2012; Pope and Dockery, 2006). In recent decades, the serious health effects of particulate matter on human health have been well documented. There are numerous studies which have demonstrated a clear correlation between particulate matter and daily mortality and hospitalizations due to respiratory and cardiovascular disease (Crooks et al., 2016; U.S. Environmental Protection Agency, 2009; Brunekreef and Stephen, 2002; Hoek et al., 2002; Atkinson et al., 2001). Recently, there has been a particular increasing interest in the study of PM air pollution due to mineral dust. Significant effects on the physical environment and the world's human inhabitants have attracted attention to air pollution due exclusively to dust sources (Arhami et al., 2017, 2018; Ghafarian; Owlad, 2016; Neophytou et al., 2013)

With the increased interest in dust air pollution, numerous studies have been conducted in order to demonstrate a correlation between dust and daily mortality, as well as hospitalizations due to respiratory and cardiovascular diseases. Perez et al. (2008) found that a significant increase in daily mortality was observed in Barcelona, Spain due to coarse particles during dust events: they report that a 10 μg/m3 increase in PM2.5-10 causes an 8.4% increase in daily mortality during the dust intrusion days. In a study conducted by Mallone et al. (2011) on mortality effects of PM during dust days, it was found that the increase in coarse particles during the dusty days caused a 4.9% increase in cardiac mortality, compared to 0.5% on non-dust days. By contrast, some studies have reported that dust is not associated with health effects (Barnett et al., 2012; Goto et al., 2010; Hong et al., 2010).

Significant influences of dust on the environment and human health have increased the demand for understanding and forecasting the dust cycle in the atmosphere. Dust models describing the complete atmospheric dust cycle have become essential tools to understand its complex production, transformation, transport, and removal processes. This includes wet and dry deposition and its impact on air quality (Tegen, 2003). In this regard, numerous dust prediction models have been developed, including regional models, such as BSC-DREAM8b (Pérez et al., 2006a, 2006b; Nickovic et al., 2001)]), SKIRON (Ničković and Dobričić, 1996), MOCAGE (Martet et al., 2009), and NMMB/BSC-Dust (Pérez et al., 2011) and global models, such as Navy Aerosol Analysis and Prediction System (Westphal et al., 2009). These models can be used to provide operational dust forecasts that can generate full 3-D areas of dust concentrations and offer a birds-eye view of dust in the atmosphere. Models rely on the choice of parameterization, the actual physical explanation of atmospheric methods, and the tuning of specific elements in the model to perform these simulations; thus, modeling outputs need to be supervised against in situ and remote sensing data to assess their accuracy (Binietoglou et al., 2015). All of the published studies compare PM2.5, PM10 in situ measurements, or aerosol optical depth (AOD) values with model outputs in order to assess the model's performance.

Considering the various outcomes for PM2.5, the present work is a comparison of regional dust modeling outputs with specific dust concentrations calculated from the chemical characterization of PM2.5. The main focus of this study is to compare outputs of the two commonly used regional dust models, BSC-DREAM8b and NMMB/BSC-Dust, using dust concentrations calculated from chemical analysis of PM2.5 samples collected in Tehran. Furthermore, the study investigates the effect of the temporal frequency of measurement and the spatial variability of model output on the difference between modeled and observed daily concentrations of dust. Dust models are often validated using in situ surface PM measurements and/or remote sensing observations. Here a different validation method is described that uses the dust contribution in PM2.5 from the ground observation site as a more accurate comparison between simulated and measured concentrations. The comparison between the two indicate that monthly average simulations of both models agree with the monthly observed concentration, but there was a poorer agreement between the daily simulation of both model and measurements. Most studies in model-observation comparison have tended to focus on model validation and statistical analysis of the correlation between model and observed values, rather than investigating reasons that cause a disparity between model and observation. The findings of this study will help to understand the effect of temporal frequency of measurement and spatial variability of models on discrepancies between numerical simulations and observations. To achieve this, 24-h PM2.5 samples were collected at a central noncommercial station in Tehran six times per month throughout a full twelve month period from February 2014 to February 2015. The samples were analyzed to determine the major mass components of PM2.5. Dust-originated components were identified by chemical mass closure. We used outputs of BSC-DREAM8b and NMMB/BSC-Dust regional dust models, which have been employed for a variety of studies concerning dust transport, deposition, and prediction.

Tehran often faces episodes of substantially high levels of pollutants and PM2.5 stands out as the primary cause of serious episodes of air pollution. Dust is responsible for 45% of PM2.5 mass in Tehran as it is part of the world's dust belt in the Middle East (Arhami et al., 2018; Ashrafi et al., 2014). There have been only a limited number of studies aiming to quantify dust intrusion, which plays a major role in increasing PM2.5 levels in Tehran.

Section snippets

Sampling and chemical analysis

PM2.5 samples have been collected as explained in Arhami et al. (2017). In brief, every six days for a complete 12 months from February 2014 to February 2015, 24-h samples of PM2.5 were collected on the roof of the primary residential air quality monitoring station, which was located at the Sharif Faculty of Technology campus in Tehran (35.7 N and 51.4 E). Two low-volume ambient air samplers (PQ200 by BGI, Inc., USA) were used to collect two sets of samples at the same time on quartz fiber

Results and discussion

This section starts with a comparison between dust estimates calculated by different methods. Fig. 2 shows the dust concentration estimation using the different approaches described in the methodology. Although the same ICP-MS chemical measurements were used, different concentrations were obtained following the same trend. A maximum concentration, with the value of 57 μg/m3, was estimated using the approach of subtracting the sum of organic carbon, elemental carbon, sulfate, nitrate, and

Conclusion

Dust storms are one of the natural hazards that have seriously affected the economy, public health, and environment of Iran throughout the past decade. The principal causes of dust resuspension are atmospheric circulations, lack of moisture in the soil, and land surface cover. Studying the mechanisms of dust storm occurrence, trajectory, velocity, and concentration are important for developing dust preparedness plans.

The main focus of this study was to validate two different regional dust

CRediT authorship contribution statement

Muge Yasar: Conceptualization, Software, Methodology, Investigation, Formal analysis, Visualization, Data curation, Writing - original draft. Alexandra M. Lai: Data curation, Investigation, Writing - review & editing. Benjamin de Foy: Conceptualization, Methodology, Writing - review & editing. James J. Schauer: Supervision, Conceptualization, Methodology, Writing - review & editing. Mohammad Arhami: Investigation, Writing - review & editing. Vahid Hosseini: Investigation, Writing - review &

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors would like to acknowledge the Air Quality Company manager and staff for providing the samplers and granting access to the Sharif air quality station and its data. We would also like to thank Maryam Zare Shahne and Navid Roufigar Haghighata for their important contributions to the field campaign.

References (41)

  • M. Arhami et al.

    Seasonal trends, chemical speciation and source apportionment of fine PM in Tehran

    Atmos. Environ.

    (2017)
  • G. Hoek et al.

    Association between mortality and indicators of traffic-related air pollution in The Netherlands: a cohort study

    Lancet

    (2002)
  • J.O. Anderson et al.

    Clearing the air: a review of the effects of particulate matter air pollution on human health

    J. Med. Toxicol.

    (2012)
  • M. Arhami et al.

    Seasonal Trends in the Composition and Sources of PM 2.5 and Carbonaceous Aerosol in Tehran

    (2018)
  • K. Ashrafi et al.

    Dust storm simulation over Iran using HYSPLIT

    J. Environ. Heal. Sci. Eng.

    (2014)
  • R.W. Atkinson et al.

    Acute effects of particulate air pollution on respiratory admissions: results from APHEA 2 project

    Am. J. Respir. Crit. Care Med.

    (2001)
  • A.G. Barnett et al.

    The effects of the 2009 dust storm on emergency admissions to a hospital in Brisbane, Australia

    Int. J. Biometeorol.

    (2012)
  • S. Basart et al.

    Aerosols in the CALIOPE air quality modelling system: evaluation and analysis of PM levels, optical depths and chemical composition over Europe

    Atmos. Chem. Phys.

    (2012)
  • Sara Basart et al.

    Development and evaluation of the BSC-DREAM8b dust regional model over northern Africa, the mediterranean and the middle east

    Tellus Ser. B Chem. Phys. Meteorol.

    (2012)
  • BGI by Mesa Labs, n.d. PQ200 Air Sampler User Manual...
  • I. Binietoglou et al.

    A methodology for investigating dust model performance using synergistic EARLINET/AERONET dust concentration retrievals

    Atmos. Meas. Tech.

    (2015)
  • J.R. Brook et al.

    The relationship among TSP, PM10, PM2.5, and inorganic constituents of atmospheric particulate matter at multiple Canadian locations

    J. Air Waste Manag. Assoc.

    (1997)
  • B. Brunekreef et al.

    Air pollution and health

    Lancet

    (2002)
  • J.L. Crooks et al.

    The association between dust storms and daily non-accidental mortality in the United States, 1993-2005

    Environ. Health Perspect.

    (2016)
  • F. de Longueville et al.

    Desert dust impacts on human health: an alarming worldwide reality and a need for studies in West Africa

    Int. J. Biometeorol.

    (2013)
  • P. Ghafarian et al.

    Temporal Variations in the Frequency and Concentration of Dust Events over Iran Based on Surface Observations 2062

    (2016)
  • K. Goto et al.

    Relationship between influx of yellow dust and bronchial asthma mortality using satellite data

    Sci. Res. Essays

    (2010)
  • K. Haustein et al.

    Regional dust model performance during SAMUM 2006

    Geophys. Res. Lett.

    (2009)
  • Y.C. Hong et al.

    Asian Dust Storm and pulmonary function of school children in Seoul

    Sci. Total Environ.

    (2010)
  • J. Jaafari et al.

    Human and Ecological Risk Assessment: an International Journal Study of PM 10 , PM 2.5 , and PM 1 Levels in during Dust Storms and Local Air Pollution Events in Urban and Rural Sites in Tehran

    (2017)
  • Cited by (0)

    Peer review under responsibility of Turkish National Committee for Air Pollution Research and Control.

    View full text