Elsevier

Aeolian Research

Volume 9, June 2013, Pages 153-160
Aeolian Research

Assessing sources of airborne mineral dust and other aerosols, in Iraq

https://doi.org/10.1016/j.aeolia.2013.02.003Get rights and content

Abstract

Most airborne particulate matter in Iraq comes from mineral dust sources. This paper describes the statistics and modeling of chemical results, specifically those from Teflon® filter samples collected at Tikrit, Balad, Taji, Baghdad, Tallil and Al Asad, in Iraq, in 2006/2007. Methodologies applied to the analytical results include calculation of correlation coefficients, Principal Components Analysis (PCA), and Positive Matrix Factorization (PMF) modeling. PCA provided a measure of the covariance within the data set, thereby identifying likely point sources and events. These include airborne mineral dusts of silicate and carbonate minerals, gypsum and salts, as well as anthropogenic sources of metallic fumes, possibly from battery smelting operations, and emissions of leaded gasoline vehicles. Five individual PMF factors (source categories) were modeled, four of which being assigned to components of geological dust, and the fifth to gasoline vehicle emissions together with battery smelting operations. The four modeled geological components, dust-siliceous, dust-calcic, dust-gypsum, and evaporate occur in variable ratios for each site and size fraction (TSP, PM10, and PM2.5), and also vary by season. In general, Tikrit and Taji have the largest and Al Asad the smallest percentages of siliceous dust. In contrast, Al Asad has the largest proportion of gypsum, in part representing the gypsiferous soils in that region. Baghdad has the highest proportions of evaporite in both size fractions, ascribed to the highly salinized agricultural soils, following millennia of irrigation along the Tigris River valley. Although dust storms along the Tigris and Euphrates River valleys originate from distal sources, the mineralogy bears signatures of local soils and air pollutants.

Highlights

► Statistics and modeling of chemical results from Teflon filter samples of aerosols collected at six sites in Iraq. ► Principal Components Analysis (PCA), and Positive Matrix Factorization (PMF). ► Four of the five PMF factors were assigned to components of geological dust. ► Baghdad has the highest proportions of salinized agricultural soils. ► Aerosol mineralogy bears signatures of distal and local soils.

Introduction

Airborne mineral dust severely impacts human health (Bu-Olayan and Thomas, 2011, Engelbrecht et al., 2009a, Engelbrecht et al., 2009b) as well as daily operations and activities in Middle Eastern countries, including Iraq. Major particulate matter (PM) contributions come from natural dust sources such as soils (Buringh, 1960), exacerbated by agricultural (Al-Awadhi et al., 2005, Al-Awahdi et al., 2003) and other human activities (Al-Awadhi, 2001), as well as diesel vehicle emissions, oil refineries, construction, and industry in the region. Desertification over the past decades and wind erosion in the Sistan region of southeastern Iran is responsible intense dust storms, affecting regional air quality and severely impacting on human health (Rashki et al., 2013). From an electron microscopic study on individual aerosol particles collected over Riyadh in Saudi Arabia, Pósfai et al. (2013) established the relationship between mineral dust particles in the atmosphere and cloud droplet nucleation.

The Enhanced Particulate Matter Surveillance Program (EPMSP) provided measurements of the chemical and physical properties of airborne mineral dusts and other aerosols at 15 sites in the Middle East (Engelbrecht et al., 2008, Engelbrecht et al., 2009a, Engelbrecht et al., 2009b). Observed chemical variations point to spatial and temporal differences amongst the sites over the year-long sampling period (Kutiel and Furman, 2003), related to differences in regional soil mineralogy (Kahlaf et al., 1985), transport patterns, local industries, and seasonal meteorology. PM concentration levels often far exceed those of international health standards during severe dust storms (Draxler et al., 2001, Engelbrecht et al., 2008, Engelbrecht et al., 2009a). The major goal of the above mentioned surveillance was to determine levels of potentially harmful trace elements such as lead, arsenic, and other heavy metals. The elemental and ion compositions, together with X-ray diffractometry and SEM based individual particle-chemistry also provided supporting evidence of aerosol mineralogy (Engelbrecht et al., 2009a).

We show that statistical analysis of the chemical data provides a better understanding of chemical interrelationships and mineral assemblages, typical of airborne dust and pollutant sources impacting in the region. Methodologies applied include the calculation of correlations, Principal Components Analysis (PCA), and Positive Matrix Factorization (PMF) modeling. Chemical data from filter sets collected at the six military sites in Iraq were statistically analyzed – data mining results from the Teflon® filter set are presented in this paper.

Section snippets

Terrain

The Syrian Desert occupies much of western Iraq, and the Arabian Desert much of the country to the south and southwest. Iraq is composed of broad arid plains, with two regularly flooded valleys of the Tigris and the Euphrates Rivers, transecting the country from northwest to southeast (Fig. 1). These river valleys are narrow for the first third of their flows in Iraq, before opening into broad flood plains north of Baghdad, creating a well-watered agricultural terrain. Elevations along the

Sampling and analysis

Filter (Engelbrecht et al., 2009a) and bulk soil (Engelbrecht et al., 2009b) sampling was performed over a period of a year, largely in 2006 and the beginning of 2007. At each site, ambient sampling was performed with three collocated Airmetrics MiniVol® particulate samplers mounted on poles, with inlets approximately 2 m off the ground. Each sampler was fitted with a size selective inlet, either for total suspended (TSP), less than 10 μm in aerodynamic diameter (PM10), or less than 2.5 μm in

Correlations

Correlation coefficients provide a measure of the interdependence of pairs of chemical variables. High correlations signifying chemicals being emitted by similar or collocated sources while small correlation coefficients represent single species emitters, or bear no relationship to specific sources. Linear correlation coefficients were calculated for each of the size fractions (TSP, PM10, and PM2.5) and also on the combined Teflon® filter set, for all sampled sites in the Middle East (

Discussion

From linear correlation coefficients (Table 2, Supplement A), chemical assemblages associated with previously established emission sources in the region (Engelbrecht et al., 2009a) were identified. These included soils that generally contain silicon (Si), aluminum (Al), potassium (K), calcium (Ca), magnesium (Mg), and iron (Fe) as their major components, with minor amounts of manganese (Mn), vanadium (V), and titanium (Ti), and trace amounts of rubidium (Rb), yttrium (Y), and zirconium (Zr).

Conclusions

The statistical analysis of the chemical results from ambient aerosols collected as part of the EPMSP (Engelbrecht et al., 2009a), provided an understanding of species interrelationships, mineral assemblages, and polluting sources impacting six monitoring sites in Iraq. Chemical data sets from Teflon® membrane filter sets and three size fractions (TSP, PM10, PM2.5) could be interpreted by applying exploratory statistics such as correlation coefficients and PCA, followed by the PMF receptor

Acknowledgements

The assistance by Greta Engelbrecht with preparation of the graphical presentations is greatly appreciated. The research is based on results from the Enhanced Particulate Matter Surveillance Program (EPMSP) performed for the United States Army Public Health Command (USAPHC), funded under Contract Numbers W9124R-05-C-0135/SUBCLIN 000101-ACRN-AB 028, and W9124R-07-C-0028 modification number P00016.

References (20)

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

Cited by (34)

  • Mineralogical, geochemical, and textural characteristics of soil and airborne samples during dust storms in Khuzestan, southwest Iran

    2022, Chemosphere
    Citation Excerpt :

    The concentrations of Ba and Ni are related to industrial activities such as oil extraction in the Persian Gulf area (Alastuey et al., 2016; Theodosi et al., 2018; Attiya and Jones, 2020). Furthermore, high levels of Strontium (Sr) were detected in the dust samples that are related to high concentrations of calcite in Khuzestan and surrounding areas (Engelbrecht and Jayanty, 2013), also affecting the marine environment of the Arabian Sea (Suresh et al. 2021). The mean V/Ni ratio was found to be 0.37, significantly lower than the V/Ni ratio of 2–4 reported for shipping emissions in the Meditteranean (Pandolfi et al., 2011; Becagli et al., 2012; Alastuey et al., 2016), indicating different sources and only a weak influence from shipping in the examined samples.

  • Evaluation of minerals being deposited in the Red Sea using gravimetric, size distribution, and mineralogical analysis of dust deposition samples collected along the Red Sea coastal plain

    2021, Aeolian Research
    Citation Excerpt :

    Most of the quartz (61.93%) is contained in Factor 4 together with some clay minerals, and can be labelled as quartz rich (sandy) soils. These mineral factors can be considered to represent fingerprints of the different dust types impacting the sampling sites (Engelbrecht and Jayanty, 2013). We calculate the source contributions, which reveal the contribution of each factor to each sample (Supplement S7).

  • Spatial and temporal variations of airborne dust fallout in Khorasan Razavi Province, Northeastern Iran

    2018, Geoderma
    Citation Excerpt :

    This was attributed to increasing distance from the presence of recently deposited sediments and the formation of loose salt minerals on the playa surface which are susceptible to wind erosion following desiccation. In recent years, many studies have been carried out on the temporal variability of dust to better understand aeolian and land surface processes (Engelbrecht and Jayanty, 2013; Flagg et al., 2014; Wang et al., 2009). For example, Groll et al. (2013) reported that peak dust rates in Central Asia between 2003 and 2010 occurred in February and June.

View all citing articles on Scopus
View full text