at SciVerse ScienceDirect

Journal of Arid Environments 77 (2012) 72e83

Contents lists available

Journal of Arid Environments

journal homepage: www.elsevier .com/locate/ jar idenv

The evaluation of PM10, PM2.5, and PM1 concentrations during the Middle EasternDust (MED) events in Ahvaz, Iran, from april through september 2010

A. Shahsavania, K. Naddafia,c,*, N. Jafarzade Haghighifardb, A. Mesdaghiniaa, M. Yunesiana,c,R. Nabizadeha, M. Arahamid, M.H. Sowlata, M. Yarahmadia, H. Sakib, M. Alimohamadia, S. Nazmaraa,S.A. Motevaliane, G. Goudarzib

aDepartment of Environmental Health Engineering, School of Public Health, Tehran University of Medical Sciences, Enghelab St., Tehran, IranbDepartment of Environmental Health, Ahvaz Jondishapoor University of Medical Sciences, Ahvaz, Iranc Institute for Environmental Research, Kargar St., Enqelab Sq., Tehran, IrandDepartment of Civil Engineering, Sharif University of Technology, Azadi St., Tehran, IraneDepartment of Biostatistics and Epidemiology, School of Public Health, Tehran University of Medical Sciences, Enqelab St., Tehran, Iran

a r t i c l e i n f o

Article history:Received 10 February 2011Received in revised form17 August 2011Accepted 10 September 2011Available online 19 October 2011

Keywords:The Middle Eastern Dust eventParticulate matterHealth effectsAir qualityAhvaz

* Corresponding author. Enghlab Square, 16 AzarTel.: þ98 912 2022363; fax: þ98 21 66707315.

E-mail addresses: ashahsavani@gmail.com (A. Sh(K. Naddafi), na.haghighifarad@gmail.com (N.mesdaghinia@sina.tums.ac.ir (A. Mesdaghinia), yunesrnabizadeh@tums.ac.ir (R. Nabizadeh), arhami@sharifgmail.com (M.H. Sowlat), m.yarahmadi86@gmhamed18nov@yahoo.com (H. Saki), m_alimohammadi@snazmara@gmail.com (S. Nazmara), amotevalian@trgoodarzy@gmail.com (G. Goudarzi).

0140-1963/$ e see front matter Crown Copyright � 2doi:10.1016/j.jaridenv.2011.09.007

a b s t r a c t

In this study, PM10, PM2.5, and PM1 concentrations were measured from April through September 2010.These measurements were made every six days and on days with dust events using a Grimm Model 1.177aerosol spectrometer. Meteorological data were also collected. Overall mean values of 319.6 � 407.07,69.5 � 83.2, and 37.02 � 34.9 mg/m3 were obtained for PM10, PM2.5, and PM1, respectively, with corre-sponding maximum values of 5337.6, 910.9, and 495 mg/m3. The presence of the westerly prevailing windimplied that Iraq is the major source of dust events in this area. A total of 72 dust days and 711 dust hoursoccurred in the study area. The dust events occurred primarily during July. The longest dust event duringthe study period occurred in July, lasted five days, and had a peak concentration of 2028 mg/m3. Thesehigh concentrations produced AQI values of up to 500. A total estimated mortality and morbidity of 1131and 8157 cases, respectively, can be attributed to these concentrations. The results of this study indicatedthe importance of dust events in Ahvaz and their possible health impacts. The study also demonstratedthe need to design and implement intergovernmental management schemes to effectively mitigate suchevents.

Crown Copyright � 2011 Published by Elsevier Ltd. All rights reserved.

1. Introduction

Dust events are defined as natural events with substantialparticulate matter (PM) concentrations, usually occurring in arid,semi-arid, or desert areas (Wang et al., 2005) and primarilyresulting from low vegetation cover and strong surface winds(Kurosaki and Mikami, 2003). Dust events produce the large-scaleor even global transport of large amounts of mineral dusts everyyear (Moulin, 1997). It is hypothesized that half of this amount is

Street, TUMS, Tehran, Iran.

ahsavani), knadafi@tums.ac.irJafarzade Haghighifard),

ian@tums.ac.ir (M. Yunesian),.edu (M. Arahami), hsowlat@ail.com (M. Yarahmadi),tums.ac.ir (M. Alimohamadi),ums.ac.ir (S.A. Motevalian),

011 Published by Elsevier Ltd. All

deposited near the sources. The other half can be suspended in theatmosphere, undergoing long-range transport (Zhang, 1995) andstrongly affecting PM concentrations (He et al., 2001), the earth’sradiation budget (Satheesh and Krishna Moorthy, 2005), traffic andagriculture (Kurosaki andMikami, 2003), the chemical compositionof the troposphere (Dentener et al., 1996), the thermal structure ofthe atmosphere, and photochemical and dynamic processes in theatmosphere (Liao et al., 1999). Therefore, the dust transported bythese events can interfere with normal activities.

In addition to the abovementioned impacts, the health effects ofdust events over both the short and the long terms have attractedthe attention of scientists. This scientific interest has producedmany epidemiological studies in susceptible areas. For example,Kwon et al. (2002) found a statistically significant relationshipbetween dust events and mortality due to cardiovascular andrespiratory diseases in Seoul, Korea. Meng and Lu (2007) associateddust events with total respiratory hospitalization, upper respiratorytract infection, pneumonia, hypertension, and cardiovascularhospitalization. Moreover, Perez et al. (2008) hypothesized that

rights reserved.

A. Shahsavani et al. / Journal of Arid Environments 77 (2012) 72e83 73

mineral dust can significantly increase total mortality owing to thecoarse fraction of particulate matter. It has also been hypothesizedthat these health effects are greater in peoplewhose socioeconomicstatus is low (Neidell, 2004). Similar results were obtained by Chanet al. (2008). Additional sources (Brunekreef and Forsberg, 2005;Pope and Dockery, 2006; WHO, 2006) offer more detailedinformation.

The other hazard posed by dust events is that the mineral dustsassociated with the events can potentially carry different infectiveagents and transport themover long distances. According to Kelloggand Griffin (2006), dust events can produce large-scale transport ofpathogens and thereby affect downwind populations and ecosys-tems. Griffin (2007) also found that pathogenic microorganismssuch as Bacillus anthracis, Yersinia pestis,Mycobacterium tuberculosis,Legionella pneumophila, and influenza viruses can be transported bydesert dusts and produce outbreaks at downwind sites. In addition,it has been suggested thatwind-blowndesert dusts commonlycarryhigh amounts of toxins and thereby endanger the organisms andecosystems exposed to the dust (Sandstrom and Forsberg, 2008).Moreover, metal elements can bind to dust particles and thereforeaffect respiratory function (Hong et al., 2010). Dust events have alsobeen related to increased deposition rates of radioactive material(Akata et al., 2007). Therefore, the importance of mineral dustparticles should not be neglected.

Fig. 1. Location of the study area and measurement st

The Sahara Desert, located in North Africa, is considered torepresent the major source of mineral dust in the world, releasingapproximately one billion tons of dust annually (Moulin, 1997).Western China and parts of Mongolia are believed to be the secondmajor source of desert dust (Wang et al., 2008). Dust from thissource is termed “Asian Dust”. Because of their importance, thesetwo sources have been extensively evaluated. For example, studieshave been conducted in Libya (O’Hara et al., 2006), Mongolia(Natsagdorj et al., 2003), South Korea (Kwon et al., 2002), Japan(Akata et al., 2007), Taiwan (Liu et al., 2006), and particularly, inChina (He et al., 2001; Qian et al., 2002; Wang et al., 2005; Zhanget al., 2010). These studies evaluated the concentrations of partic-ulate pollutants during dust events and their possible sources andhealth effects. Accordingly, a substantial body of knowledge isavailable in this field. More detailed information about Asian Dust ispresented by Shao and Dong (2006).

The third major source of desert dust is believed to be theArabian Peninsula, which contributes significantly to the totaltransport of dust particles worldwide (Goudie and Middleton,2001). Our knowledge about this source is, however, sparse andis limited to the studies conducted by Leon and Legrand (2003) andDraxler et al. (2001), which reported PM10 concentrations of up to2500 and >3000 mg/m3 in Kuwait and Saudi Arabia, respectively.Ahvaz, one of the major cites of Iran, is close to Iraq, Kuwait and

ation, showing the nearby sources of dust events.

Table 1Summary statistics of PM10, PM2.5, and PM1 concentrations (mg/m3).

Pollutant Average Min Max Median SD

PM10 April 358.1 16 3086 138 408.7May 415.8 44.7 4025.8 223.9 553.8June 488.9 39.6 5337.6 240.5 695.4July 303.6 81.5 2028.8 210.2 252.3August 185.7 66.3 1855 170.9 98.5September 221.9 66 1240.2 185 145.8Overall 319.6 39.6 5337.6 199.7 407.07

PM2.5 April 65.6 12.3 370.3 49.4 52.7May 78.8 10.5 865.2 47.8 110.8June 105.8 16.2 910.9 43.2 151.1July 65.7 19.8 406 51.4 45.5August 59.7 24.8 163.2 51.9 25.8September 49.3 21.3 209.5 44.3 25.2Overall 69.5 10.5 910.9 47.2 83.2

PM1 April 29.1 4.2 154 19.7 15.2May 35.1 4.4 426.8 21.35 48.9June 45.7 9.3 495.3 22.1 55.6July 37.3 11.6 16.6 27.95 23.1August 46.8 16.2 140.1 39.5 23.8September 31.2 11.4 93.1 27.6 14.9Overall 37.02 4.4 495.3 27.4 34.9

Fig. 2. Temporal trends in daily average PM10, PM2.5, and PM1 concentration

A. Shahsavani et al. / Journal of Arid Environments 77 (2012) 72e8374

Saudi Arabia, the primary sources of dust events in this region. Inparticular, Ahvaz appears to be influenced to a considerable degreeby dust carried by the Shamal wind, a hot northwesterly wind thatis dominant during the spring and carries large quantities of dustfrom southern areas of Iraq (Goudie and Middleton, 2006;Middleton, 1986). In recent years, and especially since 2004, thiscity has been experiencing major desert dust events originatingfrom the sources mentioned. This event is termed “the MiddleEastern Dust (MED) event”. These major dust events have inter-fered significantly with several aspects of normal life and haveproduced environmental and economical problems (Magiran,2008). The most severe dust event occurred during Summer2008. It caused the closure of industrial and educational centersand led to thousands of hospital admissions for cardiovascular andrespiratory diseases. Despite its importance, this phenomenon hasnot been studied, and questions are still being raised about manyaspects of the phenomenon.

Therefore, the present study and have produced sought toevaluate the PM10, PM2.5, and PM1 concentrations during dust andnon-dust days in Ahvaz from April through September 2010, theperiod during which the highest frequency of dust events is ex-pected to occur. The purpose of the study was to determine the

s over the study period in Ahvaz. (a) mean values; (b) maximum values.

A. Shahsavani et al. / Journal of Arid Environments 77 (2012) 72e83 75

general trend of particulate pollution concentrations, to extract thepatterns of diurnal variation of PM concentrations, to define thetemporal trends of PM ratios, to determine the statistical charac-teristics of the dust events, and to evaluate the effect of meteoro-logical conditions on the occurrence of dust events. We also soughtto estimate the impact of particulate pollution on human healthwith AirQ software developed by WHO (2004). The results of thisstudy can provide sound knowledge of the importance and effectsof dust events in the study area. This knowledge can assist policy-makers to design more effective integrated air quality managementschemes.

2. Materials and methods

2.1. Description of the study area

Ahvaz is an arid area located in southwestern Iran (31�200N,48�400E) near Iraq, Saudi Arabia, and Kuwait, the major sources ofdust events in the Middle East (Draxler et al., 2001; Goudie andMiddleton, 2001; Leon and Legrand, 2003). The city has a totalpopulation of 1.3 million and a total surface area of 220 km2. Fig. 1shows the location of the city in relation to the abovementionedsources of dust events.

Fig. 3. Diurnal variation in PM10, PM2.5, and PM1 concentrations over

The sampling station was located on the roof of the HealthResearch Center. This site, at the center of an urban area, waschosen to represent human exposure to particulate matter in thearea. The station was installed 10 m above the ground to minimizethe effects of natural features, such as trees and hills, andmanmadestructures, such as buildings, on PM concentrations. The location ofthe sampling station is shown in Fig. 1.

2.2. Instrumentation

The concentrations of coarse (PM10), fine (PM2.5), and ultrafine(PM1) particulate matter were measured using a Grimm model1.177 aerosol spectrometer (Grimm Aerosol Technik GmbH, Ainr-ing, Germany). The device, an optical particle counter (OPC), is aninstrument for real-time measurement of particulate matter in the1.0e10.0 mm range. The advantages of this instrument over otherreal-time measurement instruments, such as TEOM or DMAdevices, are its convenience, its low maintenance requirements,and its ability to run for long periods without specific supervision(Burkart et al., 2010). The ability to measure particulate matterconcentrations in time intervals ranging from 1 s to 60 min isconsidered to be another advantage offered by the instrument. Thevalidity of the measurements performed by the instrument has

the study period in Ahvaz. (a) mean values; (b) maximum values.

Fig. 4. Trends in daily average PM2.5/PM10, PM1/PM10, and PM1/PM2.5 ratios during the study period in Ahvaz.

Fig. 5. SEMphotograph (a) andX-ray spectra (b) of the particles collected on 25May2010.

A. Shahsavani et al. / Journal of Arid Environments 77 (2012) 72e8376

been evaluated by several studies (Burkart et al., 2010; Grimm andEatough, 2009; Heim et al., 2008; Peters et al., 2006). These studieshave demonstrated high correlations between the outputs of theinstrument and the results of standard gravimetric methods. Toassess the morphology and elemental composition of the particlesin dust events, PM10 samples were collected on a fiberglass filterusing a high-volume sampler. The filter was then analyzed forparticle morphology and elemental composition using the SEM-EDX system at the Razi Metallurgical Research Center. Theparticle number-size distributions per liter of air sampled were alsoobtained with the Grimm model 1.108 OPC. The ranges of particlesizes included >0.30 mm, >0.40 mm, >0.50 mm, >0.65 mm,>0.80 mm, >1.0 mm, >1.6 mm, >2.0 mm, >3.0 mm, >4.0 mm, >5.0 mm,>7.5 mm, >10.0 mm, >15.0 mm, and >20.0 mm.

2.3. Measurement times

Measurements were made from April through September 2010.These measurements were made every six days and on dust days(days when PM10 concentrations were higher than 250 mg/m3). Thefrequency of dust events is believed to be highest during thesemonths (Natsagdorj et al., 2003; Wang et al., 2006), particularly onthe Arabian Peninsula (Leon and Legrand, 2003). On all themeasurement days, the concentrations of PM1, PM2.5, and PM10were recorded every 30 min during a 24-h period.

2.4. Meteorological data

The wind speed, wind direction, temperature, and relativehumidity (RH) in Ahvaz during the study period were obtainedfrom the Iranian Meteorological Organization website (www.weather.ir). The data were examined for missing values and foroutliers and then input toMicrosoft Excel 2003 to plot thewindroseand the temporal trends in the other parameters.

2.5. Data analysis

The PM concentrations were examined for outliers and thenanalyzed using SPSS Ver.15. All graphs were plotted usingMicrosoftExcel 2003. The mortality and morbidity rates associated with thePM10 and PM2.5 concentrations were estimated with AirQ Ver.2.2.3., which was developed by the World Health Organization

Table 2Quantitative estimates of elemental compositions of the particles collected.

Element Series Unn. C[wt.-%]

Norm. C[wt.-%]

Atom. C[wt.-%]

Oxygen K series 11.02 12.58 21.78Magnesium K series 5.04 5.76 6.57Aluminum K series 11.88 13.56 13.93Silicon K series 36.15 41.28 40.72Sulfur K series 2.39 2.73 2.36Chlorine K series 0.72 0.82 0.64Potassium K series 2.57 2.93 2.08Calcium K series 11.05 12.62 8.72Iron K series 3.37 3.85 1.91Copper K series 1.44 1.65 0.72Zinc K series 0.51 0.58 0.24Barium L series 1.45 1.65 0.33

A. Shahsavani et al. / Journal of Arid Environments 77 (2012) 72e83 77

(WHO, 2004). In addition, AQI values for PM10 and PM2.5 concen-trations over the study period were calculated according to thedocument Guidelines for the Reporting of Daily Air Quality e theAir Quality Index (USEPA, 2006).

3. Results and discussion

3.1. Particulate matter concentrations

The concentrations of PM10, PM2.5, and PM1 were measured on82 days from April through September 2010. These measurementsproduced 3936 data points for the concentration of each pollutant.Table 1 shows the summary statistics for particulate matterconcentrations during different months of the study period. Theoverall mean values of PM10, PM2.5, and PM1 were 319.6 � 407.07,69.5 � 83.2, and 37.02 � 34.9 mg/m3, respectively, and the corre-sponding maximum values were 5337.6, 910.9, and 495 mg/m3

(Table 1).These results are similar to those found by other studies. For

example, PM10 concentrations higher than 3000 mg/m3 andproduced by dust events were observed in a study conducted inIraq, Kuwait, and Saudi Arabia (Draxler et al., 2001), a region locatednear our study area that is believed to be the major source of dustevents in the Middle East. Similar results were obtained by studiesof PM2.5 concentrations conducted in China (considered the

Fig. 6. Monthly averages of particle number-size dis

primary source of dust events in East Asia, i.e., “Asian Dust”). Forexample, Ye et al. (2003) reported a mean PM2.5 concentration of67.6 mg/m3. He et al. (2001) reported mean PM2.5 concentrations of89 and 76 mg/m3 during the spring and summer, respectively.

In contrast, Meng and Lu (2007), observed mean PM2.5concentrations of up to 216.7 � 96.5 mg/m3 in December. Thesevalues are at least three times higher than the PM2.5 mean valuesobserved in the present study. The explanation for this substantialdifference is that the PM2.5 concentrations were measured duringthe autumn and winter in the study cited. Fossil fuel combustion(the primary source of fine particles) is highest during theseseasons. In addition, adverse meteorological conditions in autumnand winter can inhibit particle dispersion and produce highconcentrations near the ground. In contrast, the major source of theparticulate matter examined in the present study is mineral desertdust, which consists primarily of coarse particles (Fig. 3). It shouldbe noted that studies of PM1 are rare despite the considerableimportance of this fraction for human health. The health impor-tance of this fraction is a result of its high potential for penetratingthe lungs (Colls and Tiwary, 2010).

3.2. Temporal trends of PM concentrations

Fig. 2(a) and (b) illustrates the temporal trends in the mean andmaximum values of the PM10, PM2.5, and PM1 concentrations overthe study period. It is apparent from the figure that the mean andmaximum values of all of the particulate matter fractions reachedpeaks in MayeJune, when the wind speeds were also the highest(Fig. 7). This finding is consistent with the results of Draxler et al.(2001). In that study, the maximum PM10 concentrations inKuwait, Iraq, and Saudi Arabia were observed on calendar days 120through 180, corresponding to May and June. Accordingly, mostdust events in this region (the Middle East) are expected to occurduring the late spring and early summer (Note that the PM10concentrations in all figures must be read on the right-hand axis.)This result can also indicate the role of the Shamal wind, a hotnorthwesterly wind that is dominant during the spring and bringslarge amounts of dust from southern areas of Iraq (Goudie andMiddleton, 2006; Middleton, 1986).

The diurnal variations in the mean and maximum concentra-tions of PM10, PM2.5, and PM1 are shown in Fig. 3(a) and (b). Twopeaks in the mean values (Fig. 3(a)) occurred during the day. A

tributions per liter of air over the study period.

Fig. 7. Percentage of dust days ((a), out of 72) and dust hours ((b), out of 711) duringdifferent months of the study period.

A. Shahsavani et al. / Journal of Arid Environments 77 (2012) 72e8378

sharp peak occurred at midday (1 PM), and a lower peak occurredbefore midnight (11 PM). The trends in the mean values for allparticulate matter followed the same pattern. For the maximumvalues (Fig. 3(b)), however, three major peaks occurred. The PMconcentrations were lower during the rest of the day. The first peak

Table 3Statistical characteristics of dust event duration in Ahvaz.

Item Month Dust event duration (min)

0e120 121e240

Frequency (%) of dustevent occurrence

April 57.1 21.4May 47 11.7June 41.1 11.7July 36.6 16.6August 70.5 23.5September 42.8 9.5Overall 47.4 15.5

occurred at midday (1 PM), the sharpest peak occurred during theafternoon (7 PM), and the final peak occurred before midnight (11PM). The high concentrations of PM observed during the afternooncan be explained by unstable meteorological conditions thatproduce strong winds, believed to be the primary cause of dustevents (Schwanghart and Schütt, 2008).

3.3. PM ratios

The trends in the daily averages of PM2.5/PM10, PM1/PM10, andPM1/PM2.5 ratios over the study period in Ahvaz are shown in Fig. 4.The PM2.5/PM10 ratios ranged from 0.1 to 0.5. A gradual increasewas observed from April through September. The mean PM2.5/PM10

ratio for the entire study period was 0.23. The mean values of theseratios during the spring and summer were 0.2 and 0.27, respec-tively. Similarly, increasing patterns were also observed for PM1/PM10 (ranging from 0.05 to 0.4) and PM1/PM2.5 (ranging from 0.05to 0.8), with corresponding mean values of 0.14 and 0.55 for theentire study period.

The average PM2.5/PM10 ratio observed here is much lower thanthose found by other studies. For example, He et al. (2001) reporteda mean PM2.5/PM10 ratio of 0.64 for the entire study period. Inaddition, Ye et al. (2003) reported amean PM2.5/TSP ratio of 0.55 forthe entire study period. The primary reason for this difference isthat in the two studies cited, the PM2.5/PM10 or PM2.5/TSP ratioswere calculated for the entire year, including the autumn andwinter. During these seasons, the fine fraction of particulate matter(i.e., PM2.5) is highest because various sources primarily releasefiner particles during these seasons and therefore produce anincrease in the PM2.5/PM10 ratio. Accordingly, in the study of Yeet al. (2003), the PM2.5/TSP ratio in winter increased to 0.73,whereas the ratio decreased to 0.35 in the late spring. These resultsare highly consistent with our findings. In the present study, thePM2.5/PM10 ratio was the lowest during the spring but graduallyincreased by the end of the summer. Therefore, these low ratios canbe considered to represent dust events, the main component ofwhich is expected to be coarse particles. High PM1/PM2.5 ratiosimply that PM2.5 mainly consists of ultrafine particles (PM1) andindicate the greater health impacts that can be caused by suchparticles.

3.4. Characteristics of particles in the dust events

A SEM-EDX photograph for 25 May 2010 is shown as anexample in Fig. 5(a). The SEM-EDX spectra of the particlescollected are shown in Fig. 5(b). Quantitative estimates of theelemental composition of the particles are given in Table 2. Asshown in the table, Si (41.28%) is the major component of theMiddle Eastern dusts. However, other elements, such as Al(13.56%), Ca (12.62%), Mg (5.76%), Fe (3.58%), and K (2.93%), arealso of great importance. Although the Middle Eastern dusts, likethe Saharan, Harmattan, Chinese, and North American dusts, are

241e360 361e480 481e600 >600

7.1 7.1 0 7.15.8 0 5.8 29.4

11.7 11.7 11.7 11.720 6.6 3.3 16.60 5.8 0 0

14.2 19 9.5 4.711.2 8.6 5.1 12

Table 4Statistical characteristics of dust event occurrence time interval in Ahvaz.

Item Month Time interval (h) Day time9e12

Night time21e9

0e3 3e6 6e9 9e12 12e15 15e18 18e21 21e24

Frequency (%) of dustevent occurrence

April 11.3 8.2 8.8 8.9 8.7 17 22.8 8.4 60 40May 10 10.5 10.9 9.3 8.2 12.2 24.5 14.3 54.3 45.7June 8.5 9.3 7 7.2 17.5 18.3 17.9 14.3 60.9 39.1July 8.8 8.9 14.7 15.7 13.7 17.6 9.8 10.2 56.8 43.2August 5.5 4.5 20 5.3 4.7 20 30 10 60 40September 9.8 4.9 14.5 9.8 14.7 14.6 22 9.8 61.1 38.9Overall 8.7 8.2 13.1 9.9 11.6 16.6 20.08 11.2 58.9 41.1

Fig. 8. Windrose (a) and PM10 rose (b) for the study period in Ahvaz.

A. Shahsavani et al. / Journal of Arid Environments 77 (2012) 72e83 79

dominated by Si and Al, they are relatively less rich in Si (41.28%vs. the world average of 59.9%) and Fe (3.58% vs. the world averageof 6.85%). However, the Middle Eastern dusts are richer in Ca(12.26% vs. the world average of 3.94%) and Mg (5.76% vs. theworld average of 2.60%) than the abovementioned dusts. For Aland K, the Middle Eastern dusts show the same characteristics asthe Harmattan (11.02% Al and 2.81% K), Saharan (13.91% Al and2.63% K), Chinese (11.40% Al and 2.13% K), and North American(12.21% Al and 2.63% K) dusts (Goudie and Middleton, 2006).However, the concentration of Cl (chloride) is negligible. Thisfinding indicates that sea salt particles contribute little to the totalmass of the particles collected. A possible explanation of this resultis that very high concentrations of crustal dusts during dust eventscan neutralize the effects of other sources such as marine aerosols,even if their contributions are substantial during non-dustperiods. The monthly averages of the number-size distributionsof particles during different months of the study period areillustrated in Fig. 6. These results indicate that nearly 90% of theparticles are smaller than 1 mm. However, because they are sosmall, they do not contribute heavily to the total particleconcentrations. Therefore, the concentrations of coarse particlesare higher than fine particles. This difference produces low PM2.5/PM10 ratios, as outlined in section 3.3.

3.5. Statistical characteristics of dust events

Fig. 7(a) and (b) illustrates the percentages of dust days and dusthours during different months of the study. The greatest number ofdust days (16 days) occurred in July (Fig. 7(a)). May, with 13 dustdays, and June, with 12, were the months with the next highestnumbers of dust days. July also had the greatest number of dusthours (245 h) (Fig. 7(b)). June and May were again the next highestcontributors, with 142 and 124 dust hours, respectively. A total of72 dust days and 711 dust hours occurred in Ahvaz over the entirestudy period.

These findings are consistent with the results of previousstudies. For example, Leon and Legrand (2003) investigated theprimary sources of dust events near the Indian Ocean and foundthat the frequency of dust events was highest in July and August.Orlovsky et al. (2005) studied dust events in Turkmenistan andreported that the frequency of dust events was highest in April,May, June, and July. Additional studies (Kurosaki andMikami, 2003;Natsagdorj et al., 2003;Wang et al., 2005) obtained similar findings.The number of dust days reported by the abovementioned studiesranged from 61 to 170. An important finding is that the greatestamount of dust activity, and therefore the maximum number ofdust events, temporally matches the occurrence of the north-westerly Shamal wind (Goudie and Middleton, 2006; Middleton,1986). This finding could imply a role for Iraq as the origin of thedust events in the study area.

The statistical characteristics of the occurrence of dust eventsduring different months of the study are given in Table 3. Almost

half of the dust events lasted less than 2 h. The frequency of long-lasting dust events (>600 min) was highest in May and lowest inAugust, whereas the frequency of short dust events (0e120 min)was highest in August and lowest in July. The durations of the dust

Fig. 9. Trends in weekly average wind speed, RH, and temperature in Ahvaz from April through September 2010.

A. Shahsavani et al. / Journal of Arid Environments 77 (2012) 72e8380

events during June had an even frequency distribution. This resultmight explain the high contributions of May, June, and July to thetotal number of dust days and hours. It is also noteworthy that thelongest dust event during the study period occurred in July andlasted five days, with a peak concentration of 2028 mg/m3. Thehighest amount of pollution associated with a dust event occurredin June. This event lasted two days, with a peak concentration of5337.6 mg/m3.

Fig. 10. AQI values in Ahvaz for PM10 (a) and PM2.5 (b) concentrati

These results agreewith the findings ofWang et al. (2005) on thepatterns of dust events in China. That study found that approxi-mately 60%of the durations of dust events ranged from0 to 120min.The frequency of dust events was also found to decrease as theduration of dust events increased, as observed in the present study.

Table 4 shows the statistical characteristics of the occurrence ofdust events during different hours of the day. Dust events are mostfrequent from 3 through 9 PM and least frequent from midnight to

ons in Ahvaz on all the days when measurements were made.

A. Shahsavani et al. / Journal of Arid Environments 77 (2012) 72e83 81

6 AM. The data also indicate that the frequency of dust eventsduring the day is considerably higher than the frequency at night.These findings are consistent with the results reported by Wanget al. (2005) and Natsagdorj et al. (2003). In both of the studiescited, the frequency of dust events was highest from 3 to 9 PM(35e45%) and lowest frommidnight-6 AM (5e15%). In addition, theoccurrence of dust events during the day was considerably higher(60e90%) than the occurrence at night (9e35%). This result can beexplained by the occurrence of unstable meteorological conditionsduring certain time intervals. These conditions result in theformation of strong winds and thereby produce dust events.

3.6. Meteorological parameters

Fig. 8(a) illustrates the prevailing wind (windrose) in Ahvaz overthe study period. Western and southwestern winds are the mostfrequent winds in this area. Because the primary sources of duststorms, such as Kuwait, Iraq, and Saudi Arabia, are west of Ahvaz,these sources represent the primary cause of dust events in Ahvaz.As previously explained, these events transport large quantities ofmineral dust and produce several kinds of adverse impacts in thestudy area. Furthermore, Fig. 8(b) shows that these winds had thehighest PM10 concentrations. This finding also indicates that thecountries mentioned are the major sources of the Middle EasternDust (MED) in this region. In particular, Iraq should be considered inthis context. Iraq contains extensive areas of sand desert plateau(almost 40% of the country’s total area). The severe drought of 1999,deforestation actions, inappropriate land use, political instability,war impacts, and improper internal decision-making haveincreased the rate of desertification in this country (UNEP, 2003). Inturn, this process can highlight the importance of this source andits contribution to the total PM concentrations observed duringsand dust events.

The temporal trends in wind speed, relative humidity (RH), andtemperature are shown in Fig. 9. The wind speeds are high duringAprileJuly and are lower during August and September. The windspeeds are highest in June. These high winds correspond to the

Table 5Estimation of mortality and morbidity rates due to dust events in Ahvaz over the study

Health effect PM10

Relative risk(rr)

Total mortality Mean ¼ 1.06a

Lower ¼ 1.008Upper ¼ 1.114

Respiratory mortality Mean ¼ 1.12b

Lower ¼ 1.08Upper ¼ 1.37

Cardiovascular mortality e

e

e

Hospital admissions forcardiovascular diseases

Mean ¼ 1.009b

Lower ¼ 1.006Upper ¼ 1.013

Hospital admissions forrespiratory diseases

Mean ¼ 1.137c

Lower ¼ 1.096Upper ¼ 1.178

Acute Respiratory Infection Mean ¼ 1.162c

Lower ¼ 1.144Upper ¼ 1.181

a Adopted from the study of Breitner et al. (2009).b Existed as default values in the AirQ software.c Adopted from the study of Sanhueza et al. (2009).d Adopted from the study of Staniswalis et al. (2009).e Adopted from the study of Brunekreef et al. (2009).f Adopted from the study of Ignotti et al. (2010).

sharp peaks in PM concentrations observed during this month(Fig. 2(b)). The temperature gradually increases from April throughAugust and subsequently begins to decrease slightly. In contrast,the RH increases from April through July and subsequentlycontinues to increase. Therefore, the maximum temperatures andminimum RHs are observed in June. This combination of hightemperature, low RH, and high wind speed creates ideal conditionsfor the occurrence of dust events during June.

3.7. Air quality index (AQI) and health effects

Fig. 10(a) and (b) illustrates the air quality index (AQI) values forPM10 and PM2.5 concentrations in Ahvaz on all the days whenmeasurements were made. The high PM10 concentrations(Fig. 10(a)) during the dust events in May, June, and July producedAQI values of up to 500. These values imply severe impacts onhuman health (USEPA, 2006). The AQI values for the other days arelower. However, they still range from 100 to 200. These valuesimply that the air is unhealthy and may cause the aggravation ofcardiovascular and respiratory diseases, increase prematuremortality in sensitive groups, and increase hospital admissions forrespiratory diseases in the general population (USEPA, 2006).

For PM2.5, the maximum AQI value is 439, but the index valuestend to be lower on the non-dust days than those for PM10.ThePM10 AQI values generally range from100 to 150, whereas the PM2.5AQI values on many days may be less than 100. These results implythat the air quality associated with PM2.5 is relatively high and thatthis pollutant poses lower health risks to the population. The lowerAQI values for PM2.5 can be explained by the low PM2.5/PM10 ratios.These values result from the higher contribution of the coarsefraction to the total mass of dust events. This hypothesis is alsosupported by the observation that during the first half of themeasurement period, when the PM2.5/PM10 ratios were lowest, theAQI values for PM10 were considerably higher than those of PM2.5.However, during the second half of the measurement period, thePM2.5/PM10 ratio increased to approximately 0.5. The AQI values ofthe two pollutants were then closer.

period.

PM2.5

Number ofcases

Relative risk(rr)

Number ofcases

934 Mean ¼ 1.024d 197258 Lower ¼ 1.012 105

1153 Upper ¼ 1.037 285642 Mean ¼ 1.22e 490571 Lower ¼ 0.99 0773 Upper ¼ 1.5 644

e Mean ¼ 1.04e 167e Lower ¼ 0.95 0e Upper ¼ 1.13 3371036 e e

735 e e

1384 e e

4522 Mean ¼ 1.06f 15584122 Lower ¼ 1.014 4574771 Upper ¼ 1.108 23121041 e e

1016 e e

1063 e e

A. Shahsavani et al. / Journal of Arid Environments 77 (2012) 72e8382

Estimates of the mortality and morbidity rates resulting fromdust events in Ahvaz over the study period are given in Table 5. Thetotal mortalities associated with the PM10 and PM2.5 concentrationsare 1153 and 285, respectively. The corresponding values formorbidity are 6599 and 1558, respectively. These results imply thatthe health impacts represented by cardiovascular and respiratorydiseases owing to dust events in Ahvaz are considerable andimpose substantial costs on society. The lower rates of mortalityand morbidity for PM2.5 can be explained by the lower concentra-tions of PM2.5 relative to those of PM10. These values correspond tothe low PM10/PM2.5 ratios that are characteristic of dust events.

It should be recognized that the current study did not includeepidemiological investigations. The estimates of the health effectscited above were based on the relative risks and baseline inci-dences furnished by AirQ software. The default values of thesevariables were used because no actual values are available for Iran.In addition, the values of relative risks were obtained fromepidemiological studies if the AirQ software did not provide thevalues needed for the analysis. Of course, this approach addsconsiderable uncertainty to the estimates provided by the soft-ware because the characteristics of the target population differedfrom those of the populations investigated in these epidemiolog-ical studies. However, these estimates can still be highly usefulbecause they give policymakers valuable information about theimportance of dust events and the substantial impacts of theseevents on society.

4. Conclusions

The present study, the first to characterize the dust eventsoccurring in the study area, produced the following findings:

� The average concentrations of PM10, PM2.5, and PM1 in Ahvazover the study period were 319.6 � 407.07, 69.5 � 83.2, and37.02 � 34.9 mg/m3, respectively. The corresponding maximumvalues were 5337.6, 910.9, and 495 mg/m3.

� The highest daily averages and the hourly peaks of the PM10,PM2.5, and PM1 concentrations were observed in late May andearly July, when the wind speeds were also highest. Thesefindings indicated the role played by the Shamal wind in thedust events in this area.

� The diurnal variations in the PM concentrations exhibitedpeaks at midday (1 PM) and before midnight (11 PM).

� Low PM2.5/PM10 ratios were characteristic of dust events. Themain component of these events was the coarse fraction of theparticles. The high PM1/PM2.5 ratios showed that PM1 consti-tutes the main fraction of the total mass of PM2.5.

� A total of 72 dust days and 711 dust hours occurred in the studyarea over the entire study period. The month of July includedmost of these occurrences (16 dust days and 245 dust hours).

� The longest dust event during the study period occurred in July.It lasted five days and had a peak concentration of 2028 mg/m3.The dust event associated with the highest level of pollution,a peak concentration of 5337.6 mg/m3, occurred in June andlasted two days.

� The frequency of dust events was highest from 3 to 9 PM andlowest from midnight to 6 AM. Approximately 60% of the dustevents occurred during the day.

� The occurrence of prevailing west and southwest winds andthe PM10 rose supported the hypothesis that Iraq, Kuwait, andSaudi Arabia are the major sources of the Middle Eastern Dust(MED) events in the study area.

� High concentrations of PM owing to dust events over the studyperiod produced high AQI values of up to 500. Therefore,adverse effects, such as the estimated total mortality and

morbidity cases of 1131 and 8157, respectively, can be expectedfrom such high particulate concentrations during dust events.

The present study demonstrated the scale and importance ofdust events in Ahvaz and the characteristics and possible healthimpacts of these events on society. Therefore, the results of thisstudy can be used by policymakers and local authorities to designand implement appropriate control measures. Of course, the miti-gation of such a large-scale event requires intergovernmentalagreements. Otherwise, these efforts may not be effective.

Acknowledgments

The authors wish to thank the Tehran University of MedicalSciences (research project number #9742) and the Institute forEnvironmental Research (IER). We also thank the Iranian HealthResearch Center for providing a site for the sampling station.

References

Akata, N., Hasegawa, H., Kawabata, H., Chikuchi, Y., Sato, T., Ohtsuka, Y., Kondo, K.,Hisamatsu, S.i., 2007. Deposition of 137Cs in Rokkasho, Japan and its relation toAsian dust. Journal of Environmental Radioactivity 95, 1e9.

Breitner, S., Stölzel, M., Cyrys, J., Pitz, M., Wölke, G., Kreyling, W., Küchenhoff, H.,Heinrich, J., Wichmann, H.E., Peters, A., 2009. Short-term mortality rates duringa decade of improved air quality in Erfurt, Germany. Environmental HealthPerspectives 117, 448e454.

Brunekreef, B., Forsberg, B., 2005. Epidemiological evidence of effects of coarseairborne particles on health. European Respiratory Journal 26, 309e318.

Brunekreef, B., Beelen, R., Hoek, G., Schouten, L., Bausch-Goldbohm, S., Fischer, P.,Armstrong, B., Hughes, E., Jerrett, M., van den Brandt, P., 2009. Effects of long-term exposure to traffic-related air pollution on respiratory and cardiovascularmortality in the Netherlands: the NLCS-AIR study. Research report (HealthEffects Institute) 139, 5e71.

Burkart, J., Steiner, G., Reischl, G., Moshammer, H., Neuberger, M., Hitzenberger, R.,2010. Characterizing the performance of two optical particle counters (GrimmOPC1.108 and OPC1.109) under urban aerosol conditions. Journal of AerosolScience 41, 953e962.

Chan, C.-C., Chuang, K.-J., Chen, W.-J., Chang, W.-T., Lee, C.-T., Peng, C.-M., 2008.Increasing cardiopulmonary emergency visits by long-range transported Asiandust storms in Taiwan. Environmental Research 106, 393e400.

Colls, J., Tiwary, A., 2010. Air Pollution Measurement, Modelling and Mitigation,third ed. Routledge, London & New York.

Dentener, F., Carmichael, G., Zhang, Y., Lelieveld, J., Crutzen, P., 1996. Role of mineralaerosol as a reactive surface in the global troposphere. Journal of GeophysicalResearch 101, 869.

Draxler, R.R., Gillette, D.A., Kirkpatrick, J.S., Heller, J., 2001. Estimating PM10 airconcentrations from dust storms in Iraq, Kuwait and Saudi Arabia. AtmosphericEnvironment 35, 4315e4330.

Goudie, A.S., Middleton, N.J., 2001. Saharan dust storms: nature and consequences.Earth Science Reviews 56, 179e204.

Goudie, A.S., Middleton, N.J., 2006. Desert dust in the global system. Springer,Germany.

Griffin, D.W., 2007. Atmospheric movement of microorganisms in clouds of desertdust and implications for human health. Clinical Microbiology Reviews 20,459e477.

Grimm, H., Eatough, D.J., 2009. Aerosol measurement: the use of optical lightscattering for the determination of particulate size distribution, and particulatemass, including the semi-volatile fraction. Journal of Air & Waste ManagementAssociation 59, 101e107.

He, K., Yang, F., Ma, Y., Zhang, Q., Yao, X., Chan, C.K., Cadle, S., Chan, T., Mulawa, P.,2001. The characteristics of PM2.5 in Beijing, China. Atmospheric Environment35, 4959e4970.

Heim, M., Mullins, J.M., Umhauer, H., Kasper, G., 2008. Performance evaluation ofthree optical particle counters with an efficient “multimodal” calibrationmethod. Journal of Aerosol Science 39, 1019e1031.

Hong, Y.-C., Pan, X.-C., Kim, S.-Y., Park, K., Park, E.-J., Jin, X., Yi, S.-M., Kim, Y.-H.,Park, C.-H., Song, S., Kim, H., 2010. Asian dust storm and pulmonary function ofschool children in Seoul. Science of the Total Environment 408, 754e759.

Ignotti, E., de Souza Hacon, S., Junger, W.L., Mourao, D., Longo, K., Freitas, S.,Artaxo, P., Ponce de Leon, A.C.M., 2010. Air pollution and hospital admissions forrespiratory diseases in the subequatorial Amazon: a time series approach. Cad.Saúde Pública 26, 747e761.

Kellogg, C.A., Griffin, D.W., 2006. Aerobiology and the global transport of desertdust. Trends in Ecology & Evolution 21, 638e644.

Kurosaki, Y., Mikami, M., 2003. Recent frequent dust events and their relation tosurface wind in East Asia. Geophysical Research Letters 30, 1736e1739.

Kwon, H.J., Cho, S.H., Chun, Y., Lagarde, F., Pershagen, G., 2002. Effects of the Asiandust events on daily mortality in Seoul, Korea. Environmental Research 90, 1e5.

A. Shahsavani et al. / Journal of Arid Environments 77 (2012) 72e83 83

Leon, J.-F., Legrand, M., 2003. Mineral dust sources in the surroundings of the NorthIndian Ocean. Geophysical Research Letters 30, 1309e1312.

Liao, H., Yung, Y.L., Seinfeld, J.H., 1999. Effect of aerosols on tropospheric photolysisrates in clear and cloudy atmospheres. Journal of Geophysical Research 104,23697e23707.

Liu, C.-M., Young, C.-Y., Lee, Y.-C., 2006. Influence of Asian dust storms on air qualityin Taiwan. Science of the Total Environment 368, 884e897.

Magiran, 2008.www.magiran.com/npview.asp?ID¼1897593 (accessed 21.11.2010).Meng, Z., Lu, B., 2007. Dust events as a risk factor for daily hospitalization for

respiratory and cardiovascular diseases in Minqin, China. Atmospheric Envi-ronment 41, 7048e7058.

Middleton, N.J., 1986. Dust storms in the Middle East. Journal of Arid Environments10, 83e96.

Moulin, C.e.a., 1997. Control of atmospheric export of dust from North Africa by theNorth Atlantic oscillation. Nature 387, 691e694.

Natsagdorj, L., Jugder, D., Chung, Y.S., 2003. Analysis of dust storms observed inMongolia during 1937e1999. Atmospheric Environment 37, 1401e1411.

Neidell, M.J., 2004. Air pollution, health, and socio-economic status: the effect ofoutdoor air quality on childhood asthma. Journal of Health Economics 23,1209e1236.

O’Hara, S.L., Clarke, M.L., Elatrash, M.S., 2006. Field measurements of desert dustdeposition in Libya. Atmospheric Environment 40, 3881e3897.

Orlovsky, L., Orlovsky, N., Durdyev, A., 2005. Dust storms in Turkmenistan. Journal ofArid Environments 60, 83e97.

Perez, L., Tobias, A., Querol, X., Künzli, N., Pey, J., Alastuey, A., Viana, M., Valero, N.,González-Cabré,M., Sunyer, J., 2008. Coarse particles fromSaharandust and dailymortality. Epidemiology 19, 800e807. 810.1097/EDE.1090b1013e31818131cf.

Peters, M.T., Ott, D., O’Shaughnessy, P.T., 2006. Comparison of the Grimm 1.108 and1.109 portable aerosol spectrometer to the TSI aerodynamic particle sizer fordry particles. The Annals of Occupational Hygiene 50, 843e850.

Pope III, C.A., Dockery, D.W., 2006. Health effects of fine particulate air pollution:lines that connect. Journal of Air & Waste Management Association 56,709e742.

Qian, W., Quan, L., Shi, S., 2002. Variations of the dust storm in China and itsclimatic control. Journal of Climate 15, 1216e1229.

Sandstrom, T., Forsberg, B., 2008. Desert dust an unrecognized source of dangerousair pollution? Epidemiology 19, 808e809.

Sanhueza, P.A., Torreblanca, M.A., Diaz-Robles, L.A., Schiappacasse, L.N., Silva, M.P.,Astete, T.D., 2009. Particulate air pollution and health effects for cardiovascularand respiratory causes in Temuco, Chile: a wood-smoke-polluted urban area.Journal of Air & Waste Management Association 59, 1481e1488.

Satheesh, S.K., Krishna Moorthy, K., 2005. Radiative effects of natural aerosols:a review. Atmospheric Environment 39, 2089e2110.

Schwanghart, W., Schütt, B., 2008. Meteorological causes of Harmattan dust in westAfrica. Geomorphology 95, 412e428.

Shao, Y., Dong, C.H., 2006. A review on East Asian dust storm climate, modelling andmonitoring. Global and Planetary Change 52, 1e22.

Staniswalis, J.G., Yang, H., Li, W.W., Kelly, K.E., 2009. Using a continuous time lag todetermine the associations between ambient PM2.5 hourly levels and dailymortality. Journal of Air & Waste Management Association 59, 1173e1185.

UNEP, 2003. Desk Study on the Environment in Iraq. United Nations EnvironmentProgramme, Switzerland.

USEPA, 2006. Guidelines for the reporting of daily air quality e the air quality index(AQI).

Wang, S., Wang, J., Zhou, Z., Shang, K., 2005. Regional characteristics of three kindsof dust storm events in China. Atmospheric Environment 39, 509e520.

Wang, S., Yuan, W., Shang, K., 2006. The impacts of different kinds of dust events onPM10 pollution in northern China. Atmospheric Environment 40, 7975e7982.

Wang, X., Huang, J., Ji, M., Higuchi, K., 2008. Variability of East Asia dust events andtheir long-term trend. Atmospheric Environment 42, 3156e3165.

WHO, 2004. Air Quality Health Impact Assessment Software AirQ 2.2.WHO, 2006. Air Quality Guidelines: Global Update 2005. World Health Organiza-

tion, Copenhagen.Ye, B., Ji, X., Yang, H., Yao, X., Chan, C.K., Cadle, S.H., Chan, T., Mulawa, P.A., 2003.

Concentration and chemical composition of PM2.5 in Shanghai for a 1-yearperiod. Atmospheric Environment 37, 499e510.

Zhang, D., 1995. Paleoclimate and environmental records available from Chinesehistorical documents. In: Mikami, T., Matsumoto, E., Ohata, S., Sweda, T. (Eds.),Paleoclimate and Environmental Variability in Austral-Asian Transect duringthe Past 2000 Years. Nagoya University, Japan, pp. 20e26.

Zhang, X.X., Shi, P.J., Liu, L.Y., Tang, Y., Cao, H.W., Zhang, X.N., Hu, X., Guo, L.L.,Lue, Y.L., Qu, Z.Q., Jia, Z.J., Yang, Y.Y., 2010. Ambient TSP concentration anddustfall in major cities of China: spatial distribution and temporal variability.Atmospheric Environment 44, 1641e1648.