MeteorologicalPatternsAssociatedwithIntense …downloads.hindawi.com/journals/amete/2012/828301.pdf · 2019. 7. 31. · the regional and synoptic weather forecast models, also able

Hindawi Publishing CorporationAdvances in MeteorologyVolume 2012, Article ID 828301, 17 pagesdoi:10.1155/2012/828301

Research Article

Meteorological Patterns Associated with IntenseSaharan Dust Outbreaks over Greece in Winter

P. T. Nastos

Laboratory of Climatology and Atmospheric Environment, University of Athens, Faculty of Geology andGeoenvironment, Panepistimiopolis, 15784 Athens, Greece

Correspondence should be addressed to P. T. Nastos, [email protected]

Received 14 February 2012; Accepted 21 March 2012

Academic Editor: Dimitris G. Kaskaoutis

Copyright © 2012 P. T. Nastos. This is an open access article distributed under the Creative Commons Attribution License, whichpermits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

The Mediterranean Basin and southern Europe are often affected by Saharan dust outbreaks, which influence the aerosol load andproperties, air quality standards, visibility and human health. The present work examines, mainly of the meteorological point ofview, three intense dust outbreaks occurred over Greece with duration of one or two days, on 4 and 6 February and 5-6 March2009. The synoptic analysis on the dusty days showed the presence of low-pressure systems in the west coasts of Europe and thenorth Tyrrhenian Sea, respectively, associated with a trough reaching the north African coast. The result of these conditions was thestrong surface and mid troposphere winds that carried significant amounts of dust over Greece. During the dusty days extensivecloud cover associated with the dust plume occurred over Greece. The air-mass trajectories showed a clear Saharan origin in allatmospheric levels, while the satellite (MODIS Terra/Aqua) observations as well as the model (DREAM) predictions verified theintense dust outbreaks over eastern Mediterranean and Greece. The ground based particulate matter concentrations in Athens wereexcessively increased on the dusty days (PM10: 150–560 μg/m3), while significant dry and wet deposition occurred as forecasted byDREAM model.

1. Introduction

According to the Earth Observatory website (http://earthob-servatory.nasa.gov/), intense dust outbreaks are considerednatural hazards, which affect the global and regional radia-tive balance, cloud microphysical properties, atmosphericheating and stability, tropical cyclone activity, ecosystems,marine environments and phytoplankton, photolysis rates,ozone chemistry, and human health [1, 2]. Mineral anddesert dust play an important role in radiative forcing, withan estimated top of atmosphere (TOA) radiative forcingin the range −0.6 to 0.4 Wm−2 [2]. However, the radiativeforcing caused by dust particles is very uncertain in bothmagnitude and sign, mainly triggered by the chemicalcomposition of mineral particles [3], by the wavelengthdependence of their optical properties (like single scatteringalbedo, asymmetry factor), as well as by the albedo ofthe underlying surface and also the relative height betweenthe dust layer and the clouds [4, 5]. Desert dust can betransported over long distances from the source regions [6],

with the larger particles to be deposited near the source, whilethe smaller ones to be suspended in the air for a few days orweeks, thus travelling over large distances.

The Saharan desert is the most important dust sourceregion in the world [7]. Exports of dust plumes to theNorth Atlantic and Mediterranean Sea occur throughout theyear [8]. The occurrence of Saharan dust (SD) events aboveeastern Mediterranean has a marked seasonal cycle, witha spring maximum and a winter minimum [9–11]. In thesummer, dust identification over the region is also frequentdue to the longer duration of the dust particles favored bythe stable weather conditions, the absence of depressions andprecipitation that favor their wet deposition. Many studies[8, 12–14] have shown that the Saharan dust events overMediterranean are mainly driven by the intense cyclonescalled Sharav, south of Atlas Mountains (Morocco). Thesecyclones are generated by the thermal contrast between coldAtlantic air and warm continental air that cross North Africaduring spring and summer. Moreover, the thermal lowsdeveloped over the desert regions in the warm period of the

2 Advances in Meteorology

year consist the beginning of the dust erosion and favoring itsuplift at the upper atmospheric levels and its transport overlong distances [15, 16]. In general, in the cold period, neitherthe development of the Sharav cyclones and the thermallows in Saharan are favored nor the dust transport overlong distances. Additionally, due to frequent precipitationand the often presence of depressions and strong winds, thedust particles deposited in the surface near the source regionand their atmospheric lifetime is limited compared to thesummer [14]. However, under specific weather conditions,mainly consisted of strong surface winds over Saharandesert closely associated with low-pressure systems, thedust outbreaks are also observed in the winter [17, 18].Furthermore, the close relation of the dust exposure andtransport with the prevailing local and regional meteorologyis well established. To this respect, Dunion and Velden [19]have shown that an elevated Saharan dust layer may playa crucial role in suppressing tropical cyclone activity inthe Atlantic and the tropical cyclones in the Bay of Bengalcause mineral dust exposure from the continental India overthe Bay of Bengal and Arabian Sea [20, 21]. Focusing inthe Mediterranean several studies [1, 15, 22] have analyzedthe prevailing meteorology during the dust events, whileMeloni et al. [18] and Carmona and Alpert [23] classifiedthe weather types responsible for the dust outbreaks inLampedusa and Israel, respectively.

While progress has been made in characterizing theimportance of mineral dust in global-scale processes, therehas been less progress in identifying the sources of dust, theenvironmental processes that affect dust generation in thesource regions and the meteorological factors that affect thedust transport. This can be achieved by the development ofthe regional and synoptic weather forecast models, also ableto predict the emission, the amount, the transport, as well asthe deposition of the dust [15, 22].

Saharan dust outbreaks also affect human health [24–27],because such episodes are closely associated with increases inparticulate matter concentrations on the surface, especiallywhen the synoptic conditions favor advection of the dustwithin the boundary layer. The health burden due toparticulate matter (PM) air pollution is one of the biggestenvironmental health concerns, especially over areas directlyaffected by intense dust storms [28].

The present study focuses on the analysis of three dustevents occurred over Greece in the cold period of the year(February 4, 6, 2009 and March 5-6, 2009), mainly fromthe meteorological aspect based on NCEP-NCAR reanalysis.The dust transport is also monitored by the satellite remotesensing, through observations from MODIS, while the use ofregional atmospheric models, such as DREAM, is establishedas a powerful tool for the dust forecasting. Ground-based PMconcentrations recorded in the University Campus of Athensare in close agreement with satellite observations and modelforecasts showing that the three dust events affect stronglythe aerosol concentrations at the ground.

2. Data Analysis

The Saharan dust episodes over Greece on February 4, 6 andMarch 5-6, 2009 were analyzed using satellite observations

(MODIS Terra/Aqua, TRMM) model forecasting (DREAM),air mass trajectories, NCEP-NCAR reanalysis datasets, andground-based PM measurements.

2.1. Satellite Observations. The data used in this studyinclude both Terra and Aqua MODIS aerosol products(AOD550, Angstrom exponent at 550–865 nm over oceanand aerosol mass concentration), calculated using separatealgorithms over land and ocean. The “Level 3” MODISproducts are available on daily and monthly intervals,globally, on a 1◦ × 1◦ grid. Further details of the MODISaerosol algorithm, products, and validation are presented inthe studies of Remer et al. [29] and Levy et al. [30]. In thepresent work, Level 3 C005 retrievals were used centered overGreater Athens Area covering the period February 1, 2009 toMarch 10, 2009.

Except of the aerosol retrievals, satellite data of theaccumulated precipitation over eastern Mediterranean dur-ing the dusty days were obtained via Tropical RainfallMeasurement Mission (TRMM) with a spatial resolution of0.25◦ × 0.25◦. Since the dust events are associated with low-pressure systems and extensive cloudiness above the studyarea, precipitation has also taken place. The TRMM spatialdistribution of the precipitation can also be compared withthe predictions of DREAM regarding the wet depositionof dust. Therefore, the TRMM observations can also be avalidation tool for the DREAM predictions.

2.2. The DREAM Model. In the last years, an integratedmodeling system, the Dust Regional Atmospheric Mod-eling (DREAM) model (http://www.bsc.es/projects/earth-science/DREAM/) [31], is widely used to simulate the 3-dimensional field of the dust concentration and its cycle inthe atmosphere. It is based on the SKIRON/Eta modelingsystem (http://forecast.uoa.gr/) and the Eta/NCEP regionalatmospheric model [16]. The dust model takes into accountall the major processes of dust life cycle, such as dustproduction, convection and advection, as well as wet anddry deposition. It also includes the effects of the particle-size distribution on aerosol dispersion, lifetime, and trans-portation from their source regions. The dust productionis parameterized using near-surface wind and thermalconditions, as well as soil features. The dust-productionmechanism is based on the turbulent mixing, shear-freeconvection diffusion, and soil moisture [31]. In addition tothese mechanisms, very high-resolution databases, includingelevation, soil properties, and vegetation cover are alsoutilized. In the operational version, for each soil texturetype, fractions of clay, small silt, large silt, and sand areestimated with typical particle size radii of 0.73, 6.1, 18, and38 μm, respectively [22]. Transport mixing and depositionprocesses are online driven by an atmospheric model andthe predicted meteorological parameters. The atmosphericmodel is updated every 24 h with newly observed data,but the simulated dust-concentration field produced in theprevious-day run initializes the dust model at the sametime intervals. This model has been extensively used for theidentification of dust events in the Mediterranean [10, 22, 31,32].

Advances in Meteorology 3

(a) (b)

Figure 1: Satellite images from Aqua-MODIS sensor on February 4, 2009 (a) and February 6, 2009 (b).

2.3. Air Mass Back Trajectories. The surface-level aerosolcharacteristics during dust events can be quite differentcompared to the columnar ones as different types of aerosolsvary depending on their scale heights [33]. For the abovereasons, the 72-hour air mass back trajectories for eachone of the examined SD events were calculated using theHYSPLIT-4 model of Air Resources Laboratory of NOAA(http://www.arl.noaa.gov/HYSPLIT info.php). The back tra-jectories analysis was carried out for 3 distinct levels, namely,500 m, to give representative origins of air masses nearthe surface, 1500 m, which can serve as a representativeheight for the boundary layer, and 4000 m, representativeof the free troposphere, where the Saharan dust is usuallytransported [1]. The Air Resources Laboratory’s HybridSingle-Particle Lagrangian Integrated Trajectory (HYSPLIT)model is a complete system for computing both simple airparcel trajectories and complex dispersion and depositionsimulations. The model calculation method is a hybridbetween the Lagrangian approach, which uses a movingframe of reference as the air parcels move from theirinitial location, and the Eulerian approach, which uses afixed three-dimensional grid as a frame of reference. In themodel, advection and diffusion calculations are made in aLagrangian framework following the transport of the airparcel, while pollutant concentrations are calculated on afixed grid. The model is designed to support a wide rangeof simulations related to the atmospheric transport anddispersion of pollutants and hazardous materials, as well asthe deposition of these materials (such as mercury) to theEarth’s surface. Some of the applications include tracking andforecasting the release of radioactive material, volcanic ash,wildfire smoke, and pollutants from various stationary andmobile emission sources [34, 35].

2.4. NCEP-NCAR Reanalysis. The National Center for Envi-ronmental Prediction (NCEP) and National Center forAtmospheric Research (NCAR) have cooperated in a project

(denoted “Reanalysis”) to produce a record of global analysesof atmospheric fields in support of the needs of theresearch and climate monitoring communities. This effortinvolved the recovery of land surface, ship, radiosondeaircraft, satellite and other data, then quality controllingand assimilating these data with a data assimilation systemwhich is kept unchanged over the reanalysis period [36].NCEP uses satellite-based temperature retrievals from theNational Environmental Satellite Data and InformationService (NESDIS). These retrievals commence in November1978 and are assimilated over all ocean areas, and above100 hPa over land areas with poor radiosonde coverage [36].

2.5. Ground-Based PM Measurements. Ground-based PMconcentrations were collected using a real-time photometricsampler (Aerocet 531-MetOne instruments), based on light-scattering method at 0.780 μm. The data consists of 3-minute recordings (operation process for 2 minutes followedby 1 minute interval) of particulate matter concentrationswith different aerodynamic diameters: PM10, (with diameterless than 10 μm), PM7 (with diameter less than 7 μm),PM2.5 (with diameter less than 2.5 μm), and PM1 (withdiameter less than 1 μm). The accuracy of the Aerocet531 detector is ±10% to calibration aerosol. The samplingsite was located at the Laboratory of Climatology andAtmospheric Environment (UOA) in the campus of theUniversity of Athens (longitude: 23◦47′E, latitude: 37◦58′N,altitude: 257 m a.m.s.l.), a suburban region with low trafficand far from inhabited areas. This location is at the east edgeof the Athens basin, at the foothills of Hymettus Mountain.Calibrations, including flow rate and zero tests, were carried-out weekly, according to the manufacturer’s instructions.

3. Results

Remotely sensing aerosols via satellite images gives gen-eral picture and can provide information of the spatial


(a) (b)

Figure 2: Satellite images from Aqua-MODIS sensor on March 5, 2009 (a) and March 6, 2009 (b).

distribution of the aerosol properties. The Aqua-MODIStrue color images on 4 and 6 February are shown inFigure 1. Both images show extensive cloud cover overGreece and, especially, over Balkan countries (February 4,2009) associated with the cyclonic conditions. However, agreat part of the images is free of clouds and, therefore,the dust plumes on both days are clearly depicted. Morespecifically, on February 4 (Figure 1(a)), the dust plume isobserved over Aegean Sea, while it is more intense south ofCrete in the Libyan Sea. Towards northern Greece, the dustplume intensity decreases due to the gravitational depositionof the largest particles. After the passage of a clear day (notshown), a new dust outbreak starts from the western Libyandesert and transported towards Greece (north-northeasterly)following the cyclonic circulation. This dust plume was moreintense as clearly can be seen from the Aqua-MODIS imageon February 6 (Figure 1(b)). It is extended from the Libyancoast towards the south Ionian and Libyan Seas also coveringthe western and central Greece. In this image, the easternAegean Sea is free from dust aerosols, but the dust plumewas transported easterwards affecting these areas in the nexthours and day, February 7 (not shown). In Athens, thevisibility was limited to few hundred meters on February 4and February 6-7, while February 5 was a day with clear skyand good visibility. In the midday hours of February 6, aslight precipitation (0.7 mm) took place, but not capable todillute the atmosphere, which was remained very turbid dueto dust presence until the evening hours of February 7, whenan intense rainfall (29 mm) took place.

In Figure 2, the analysis is focused on the dust eventon March 5-6, 2009, which was one of the most intenseabove Greece in the last years, with maximum values ofAOD550 and PM (Figures 6 and 7) comparable to thosereported by Kaskaoutis et al. [1] and Gerasopoulos et al.[37]. The images were obtained by the Aqua-MODIS sensor.On March 6, the dust storm originated from Libya was

transported northwards affecting Greece and was mainlydriven by strong surface and middle troposphere windsin a cyclonic pathway. This was also the case during theintense dust event on April 17, 2005 [1, 38]. The maximumintensity of the dust plume is depicted between the NorthAfrican coast and Crete; then it becomes more vague as beingdiluted by the removal processes (dry and wet deposition)(Figure 4). Significant cloud cover is also obvious overGreece on both days, and specifically on March 5; as aconsequence, the elevated dust plume may interact withcloud microphysical properties (cloud condensation nuclei,cloud albedo, water-vapor content) altering them in theprocess. It is well known that dust interacts with clouds, afterabsorbing hygroscopic material [39], and affects photolysisrates and ozone chemistry by modifying the spectrum of UVradiation [40]. The clouds in the left part of the image onMarch 6 are characteristic of a cyclonic circulation and revealthe cold front approaching Greece.

The air mass back trajectories at 4000 m, 1500 m, and500 m on the dusty days (Figure 3) verified the verticaltransport of dust, from near the surface to the middle oftroposphere; so did DREAM simulations, which showed dryand wet deposition of dust particles over Athens on thesedays.

The DREAM forecasts regarding the dry and wet depo-sition over north Africa, Mediterranean, and Europe arepresented in Figure 4. Concerning the dust deposition onFebruary 4, 2009 (Figures 4(a) and 4(b)), the dry mechanismclearly dominated on this day, since the precipitation over theregions affected by the dust plume was limited. As expected,the dry deposition covered the north African arid regions,the continental Greece and the adjoining seas, while wetdeposition was limited over areas with precipitation, as inAlbania and western Balkans. On February 6, 2009, thedry mechanism dominated again except from the northernAfrican regions, large amounts of dust dry deposition were


Met

ers

AG

L

3822

1323323

06 00 18 12 06 00 18 12 06 00 18 1202/04 02/03 02/02

600045003000

1500

Sou

rce

at 3

7.97

N

23.

78 E

February 4, 2009

(a)M

eter

s A

GL 3826 3500

30002500200015001000500

1326

326

06 00 18 12 06 00 18 12 06 00 18 1202/06 02/05 02/04

Sou

rce

at 3

7.97

N

23.

78 E

February 6, 2009

(b)

Met

ers

AG

L 3827 350030002500200015001000500

1327

327

06 00 18 12 06 00 18 12 06 00 18 1203/05 03/04 03/03

Sou

rce

at 3

7.97

N

23.

78 E

March 5, 2009

(c)

Met

ers

AG

L 3831 350030002500200015001000500

1330

330

06 00 18 12 06 00 18 12 06 00 18 1203/06 03/05 03/04

Sou

rce

at 3

7.97

N

23.

78 E

March 6, 2009

(d)

Figure 3: Air mass back trajectories at 4000 m, 1500 m, and 500 m for February 4, 2009 (a), February 6, 2009 (b), March 5, 2009 (c), andMarch 6, 2009 (d).

observed in central Mediterranean and the western part ofGreece (Figures 4(c) and 4(d)). The wet deposition was lim-ited mainly in the northwestern Greece and Albania, as wellas in a small area in southern Aegean Sea. On March 6, 2009there is a clear evidence of an intense dust event influencedthe majority of Greece. Significant dry deposition (100–500 mg/m2) and wet deposition, which on regional scaleover west Greece ranged from 500–1000 mg/m2 indicated theintensity of this SD episode (Figures 4(e) and 4(f)).

The findings resulted from DREAM products were inclose relation with those from TRMM analysis (Figure 5).Light precipitation appeared on February 4, 6, 2009 overGreece, with some exceptions over central mountainous

regions (>40 mm). A different pattern appeared on March 5,2009, when moderate-to-strong precipitations came up thecountry and mainly the western regions (>70 mm). On thenext day, March 6, 2009, the precipitations were limited onlocal scale (mountainous regions of Peloponnese), allowingthe dry deposition of the dust, which reached at maximumlevels, as recorded on the ground.

Before focusing on the PM concentrations over Athensduring the intense dust events of February 4, 6, 2009 andMarch 5-6, 2009, the temporal variation of some aerosolproperties obtained from Terra and Aqua-MODIS sensors isanalyzed. More specifically, Figure 6 depicts the daily valuesof mass concentration, α550–865 and AOD550, from Terra


15N

20N

25N

30N

35N

40N

45N

50N

55N

20W 10W 0 10E 20E 30E 40E 50E

0.5

2

5

10

100

500

1000

Dry deposition on February 4, 2009

(a)

15N

20N

25N

30N

35N

40N

45N

50N

55N

20W 10W 0 10E 20E 30E 40E 50E

0.5

2

5

10

100

500

1000

Wet deposition on February 4, 2009

(b)

15N

20N

25N

30N

35N

40N

45N

50N

55N

20W 10W 0 10E 20E 30E 40E 50E

0.5

2

5

10

100

500

1000

Dry deposition on February 6, 2009

(c)

15N

20N

25N

30N

35N

40N

45N

50N

55N

20W 10W 0 10E 20E 30E 40E 50E

0.5

2

5

10

100

500

1000

Wet deposition on February 6, 2009

(d)

15N

20N

25N

30N

35N

40N

45N

50N

55N

20W 10W 0 10E 20E 30E 40E 50E

0.5

2

5

10

100

500

1000

Dry deposition on March 6, 2009

(e)

15N

20N

25N

30N

35N

40N

45N

50N

55N

20W 10W 0 10E 20E 30E 40E 50E

0.5

2

5

10

100

500

1000

Wet deposition on March 6, 2009

(f)

Figure 4: DREAM 24 h forecast for 12:00 UTC for dry ((a), (c), and (e)) and wet ((b), (d), and (f)) dust depositions (mg/m2) on February4, 2009 ((a) and (b)), February 6, 2009 ((c) and (d)), and March 6, 2009 ((e) and (f)).

and Aqua-MODIS sensors, within the period February 1,2009 to March 10, 2009, averaged over the area 37.0◦–38.5◦ N and 23.0◦–24.5◦ E including the Greater AthensArea and its surroundings. This also allows a qualitativecomparison between the satellite retrievals and ground-based PM concentrations. During the period, the AOD550

varied widely from 0.06 to 2.67 for Terra-MODIS and from0.05 to 1.87 for Aqua. AOD550 values higher than 0.5 wereclosely related to the SD events as marked with arrows in the

figure. The remarkable increase in AOD550 values on March5-6, 2009 clearly revealed the intensity of this dust event.

Despite the difference in orbiting time and the fact thatthe data pixels from the two satellites are not always thesame (due to differences in cloud cover), the Terra andAqua-MODIS sensors give similar AOD550 values, while astrong linear relation (r = 0.97) holds. Furthermore, theSD events are clearly identified from the low α550–865 valuesobtained from Aqua. In the measuring period, the α550–865


34.5N35N

35.5N36N

36.5N37N

37.5N38N

38.5N39N

39.5N40N

40.5N41N

41.5N

TRMM 3B42 daily 006 precipitations [mm/day]

(04 February 2009)

10E 20E 21E 22E 23E 24E 25E 26E 27E 28E 29E

0

9.4

18.8

28.2

37.7

47.1

February 4, 2009

(a)

34.5N35N

35.5N36N

36.5N37N

37.5N38N

38.5N39N

39.5N40N

40.5N41N

41.5N


(06 February 2009)

10E 20E 21E 22E 23E 24E 25E 26E 27E 28E 29E

0

8.5

17.1

25.6

34.2

42.7

February 6, 2009

(b)

34.5N35N

35.5N36N

36.5N37N

37.5N38N

38.5N39N

39.5N40N

40.5N41N

41.5N


(05 March 2009)

10E 20E 21E 22E 23E 24E 25E 26E 27E 28E 29E

0

16.2

32.4

48.5

64.7

80.9

March 5, 2009

(c)

34.5N35N

35.5N36N

36.5N37N

37.5N38N

38.5N39N

39.5N40N

40.5N41N

41.5N


(06 March 2009)

10E 20E 21E 22E 23E 24E 25E 26E 27E 28E 29E

0

13.3

26.5

39.8

53.1

66.4

March 6, 2009

(d)

Figure 5: TRMM 3B42 daily precipitation (mm/day) for the examined Saharan episodes days.

took values from relatively low (∼0.2-0.3), during the dustevents (February 4, 6, 2009 and March 5-6, 2009), to high(>1.0) on certain days of anthropogenic aerosols and localpollution. The α550–865 values show presence of fine andcoarsemode particles, with the fine mode to be dominant(mean value of 0.84 ± 0.41). The large variability in α550–865

values is attributed to the variety of air masses, aerosol types,and characteristics. Furthermore, on the dusty days, theaerosol mass concentration enhanced, especially on March 5-6, 2009. This also reveals the strong influence of the Saharandust outbreaks on the aerosol load in southern Europe [12,32, 41].

4. Ground-Based PM Concentrations

The daily Box and Whiskers plots for PMs recorded on 3 minbasis during the period February-March 2009 are presentedin Figure 7. The sampling station at the University Campus

of Athens, outside the urban area, could be consideredas background station with respect to PM concentrations.In general, significantly higher PM concentrations wererecorded on the dusty days against the rest days of theexamined period, while the temporal pattern for PM con-centrations was similar to that observed from Terra/AquaMODIS AOD500 over Athens (Figure 6).

More specifically, on February 4, 2009, the meandaily PM concentrations were calculated as follows: 1.9 ±1.8μg/m3for PM1, 32.3 ± 24.4μg/m3for PM2.5, 88.8 ±67.2μg/m3 for PM7, 100.1 ± 75.6μg/m3 for PM10, and107.8 ± 79.2μg/m3 for TSP. The maximum values wererecorded in the early afternoon hours, which is in agreementwith DREAM dust forecast (Figure 8(a)) indicating that themajority of aerosol particles were within the boundary layer.The PM concentrations on February 6, 2009 were found atlower levels; thus, 0.6±0.9μg/m3 for PM1, 14.4±16.0μg/m3

for PM2.5, 41.0 ± 46.9μg/m3 for PM7, 46.4 ± 52.7μg/m3 forPM10, and 49.9±55.5μg/m3for TSP. In this case, the maxima


020406080

100120140160

Mas

s co

nce

ntr

atio

n

Dust events

0

0.5

1

1.5

2

α55

0–86

5

0

0.5

1

1.5

2

2.5

Aqu

a A

OD

550

0

Date

Date

Date

Date

0.51

1.52

2.53

Terr

a A

OD

550

1/2

3/2

5/2

7/2

9/2

11/2

13/2

15/2

17/2

19/2

21/2

23/2

25/2

27/2 1/3

3/3

5/3

7/3

9/3

Figure 6: Daily time series of mass concentration, α550–865 and AOD550 from Terra and Aqua-MODIS sensors during the period February 1,2009 to March 10, 2009, averaged over the area 37.0◦–38.5◦ N and 23.0◦–24.5◦ E.

were recorded at night hours (∼21:00), in agreement withdust forecast (Figure 8(b)).

The intense signature of the Saharan dust episodes onMarch 5-6, 2009 appeared in PM concentrations at theground. The mean values of PMs, on March 5, 2009 were cal-culated as follows: 0.9±0.9μg/m3 for PM1, 20.4±15.7μg/m3

for PM2.5, 60.8 ± 48.5μg/m3for PM7, 67.5 ± 54.7μg/m3

for PM10, and 71.9 ± 59.4μg/m3 for TSP, with maximaduring afternoon hours (see also Figure 9(a)). On the nextday, a further increase in PM concentrations was recorded,thus 2.6 ± 1.2μg/m3 for PM1, 59.7 ± 21.7μg/m3 for PM2.5,202.9 ± 80.5μg/m3 for PM7, 236.6 ± 100.8μg/m3 for PM10,and 265.8 ± 123.6μg/m3 for TSP; the primary maximumappeared in the morning hours, and a secondary one in thelate afternoon. This is depicted clearly in Figure 9(b), wherethe vertical profile of the dust concentration shows that thedust advected within the boundary layer. In Mediterraneancoastal urban areas, such as Athens, which are close to northAfrican arid regions, the effects of dust outbreaks play akey role in PM concentrations and further contribute to theurban air quality and human health [12, 25]. Many studies[1, 32, 37, 41, 42] have established that dramatically PM10

enhancements in Mediterranean are associated to Saharandust outbreaks.

The dust events over Greece occur either in an upperatmospheric level or in the whole atmospheric column [17,43, 44], with the latter to be more intense, directly influ-encing the PM concentrations at the surface. On the otherhand, dust transports only within the boundary layer, and itis characterized by lower AOD and PM concentrations, sincethe majority of the particles have already been depositednear the source regions [44]. According to the EU standards,since January 2005, the exceedance of the PM10 daily averagethreshold of 50 μg/m3 at any station is allowed for amaximum of 35 times per year, while the PM10 annualaverage should not exceed 40 μgm−3. In this point, it mustbe noticed that according to the 2008/50 EU directive, theexceedances of the 50 μg/m3 limit value due to Saharan dustare not included in the number of exceedances. The PM10

concentrations exceeded the daily European Union limit of50 μg/m3 for 4 days associated with dust transport, duringthe study period. It is characteristic that the SD contributionto PM10 levels is about 200 μgm−3 on March 6, comparedto the mean PM10 value in the study period. However, itshould be noted that the PM monitoring station is locatedin a suburban area and is not so characteristic for the Athensurban environment. Nevertheless, as Gobbi et al. [41] shown,the remote areas allow for a better dust monitoring than


0

1

2

3

4

5

6

7

PM

1(μ

g/m

3)

1-Fe

b3-

Feb

5-Fe

b7-

Feb

9-Fe

b11

-Feb

13-F

eb15

-Feb

17-F

eb19

-Feb

21-F

eb23

-Feb

25-F

eb27

-Feb

1-M

ar3-

Mar

5-M

ar7-

Mar

9-M

ar11

-Mar

13-M

ar15

-Mar

17-M

ar19

-Mar

21-M

ar23

-Mar

25-M

ar27

-Mar

29-M

ar31

-Mar

(a)

0

20

40

60

80

100

120

1-Fe

b3-

Feb

5-Fe

b7-

Feb

9-Fe

b11

-Feb

13-F

eb15

-Feb

17-F

eb19

-Feb

21-F

eb23

-Feb

25-F

eb27

-Feb

1-M

ar3-

Mar

5-M

ar7-

Mar

9-M

ar11

-Mar

13-M

ar15

-Mar

17-M

ar19

-Mar

21-M

ar23

-Mar

25-M

ar27

-Mar

29-M

ar31

-Mar

PM

2.5

(μg/

m3)

(b)

0

100

200

300

400

500

1-Fe

b3-

Feb

5-Fe

b7-

Feb

9-Fe

b11

-Feb

13-F

eb15

-Feb

17-F

eb19

-Feb

21-F

eb23

-Feb

25-F

eb27

-Feb

1-M

ar3-

Mar

5-M

ar7-

Mar

9-M

ar11

-Mar

13-M

ar15

-Mar

17-M

ar19

-Mar

21-M

ar23

-Mar

25-M

ar27

-Mar

29-M

ar31

-Mar

PM

7(μ

g/m

3)

(c)

0

100

200

300

400

500

600

1-Fe

b3-

Feb

5-Fe

b7-

Feb

9-Fe

b11

-Feb

13-F

eb15

-Feb

17-F

eb19

-Feb

21-F

eb23

-Feb

25-F

eb27

-Feb

1-M

ar3-

Mar

5-M

ar7-

Mar

9-M

ar11

-Mar

13-M

ar15

-Mar

17-M

ar19

-Mar

21-M

ar23

-Mar

25-M

ar27

-Mar

29-M

ar31

-Mar

PM

10(μ

g/m

3)

(d)

0

100

200

300

400

500

600

700

1-Fe

b3-

Feb

5-Fe

b7-

Feb

9-Fe

b11

-Feb

13-F

eb15

-Feb

17-F

eb19

-Feb

21-F

eb23

-Feb

25-F

eb27

-Feb

1-M

ar3-

Mar

5-M

ar7-

Mar

9-M

ar11

-Mar

13-M

ar15

-Mar

17-M

ar19

-Mar

21-M

ar23

-Mar

25-M

ar27

-Mar

29-M

ar31

-Mar

TSP

(μ

g/m

3)

MeanMean ± SDMin-max

(e)

Figure 7: Daily Box and Whiskers plots for PMs recorded on 3-minute basis during the period February-March 2009. Middle point refersto mean value, box to ±standard deviation and Whisker to min-max values.

the urban environments, where the anthropogenic emissionscontribute significantly to the background PM levels.

4.1. Synoptic Conditions. The two recorded SD episodes onFebruary 2009 were associated with alike synoptic conditions(Figure 10). More specifically, on 4 February 2009 a baromet-ric low on the surface was established with its center not toofar from the north-west coasts of France linked to a troughtowards the Adriatic Sea (Figure 10(a)). This barometricturbulence resulted in south-southwest air flow in the regionof western Greece with near gale-to-gale winds (15–20 m/s)

at 850 hPa level (Figure 11(a)) against severe gale winds (20–25 m/s) at 500 hPa level (Figure 11(c)). As far as the middleatmosphere (500 hPa) is concerned, the existence of a troughover the western Mediterranean, the Iberian Peninsula, andthe north-west coast of Africa had originated western airflow towards Greece (Figure 10(a)). The significant negativeanomaly of the 500 hPa geopotential heights from the meanvalue of 1981–2010 Climatology was very characteristic(−300 m), expanding from southern England westwards theIberian Peninsula (Figure 10(c)). This synoptic conditionrelated to high pressure system on the surface, over the regionof north-west Africa, resulted in the advection of dust from


0

2

4

6

8

10

12

Hei

ght

(km

)

0 200 400 600 800 1000

BSC/DREAMAthens: 37.97N, 23.78E

Dust forecast at 12 UTC Wednesday, 04 February 2009

Concentration (μg/m3)

12 UTC February 4, 2009

0

2

4

6

8

10

12

Hei

ght

(km

)

0 200 400 600 800 1000


Dust forecast at 18 UTC Wednesday, 04 February 2009



0

2

4

6

8

10

12

Hei

ght

(km

)

0 200 400 600 800 1000


Dust forecast at 00 UTC Thursday, 05 February 2009



(a)

0

2

4

6

8

10

12

Hei

ght

(km

)

0 200 400 600 800 1000


Dust forecast at 12 UTC Friday, 06 February 2009



0

2

4

6

8

10

12

Hei

ght

(km

)

0 200 400 600 800 1000


Dust forecast at 18 UTC Friday, 06 February 2009



0

2

4

6

8

10

12H

eigh

t (k

m)

0 200 400 600 800 1000



Dust forecast at 00 UTC Saturday, 07 February 2009


(b)

Figure 8: DREAM 6-hour vertical profile of dust forecast for February 4, 2009 (a) and February 6, 2009 (b).

these areas (Tunis, Algeria) towards Italy and western Greece,affecting partially the eastern Greece, as well. This is shownby the lower PMs concentrations against the values measuredduring the second dust episode on March 6, 2009.

The synoptic conditions and the 500 hPa negativeanomalies appeared on February 4, 2009, shifted eastwardson February 6, 2009, without significant weather patternmodification (Figures 10(b) and 10(d)). The trough in

the middle atmosphere appeared more acute and violentstorm winds (29–32 m/s) dominated over north-west Africaweakening in the region of Greece (Figure 11(d)). At the levelof 850 hPa, moderate breeze (6–8 m/s) was established overGreece against gale winds over Tunis (Figure 11(b)).

Regarding the Saharan episode on March 5-6, 2009, thesynoptic conditions have been differentiated according to thefact that the barometric low on the surface was established


0

2

4

6

8

10

12

Hei

ght

(km

)

0 500 1000 1500 2000



Dust forecast at 12 UTC Thursday, 05 March 2009

12 UTC March 5, 2009

0

2

4

6

8

10

12

Hei

ght

(km

)

0 500 1000 1500 2000



Dust forecast at 18 UTC Thursday, 05 March 2009


0

2

4

6

8

10

12

Hei

ght

(km

)

0 500 1000 1500 2000


BSC/DREAMAthens: 37.97N, 23.78EDust forecast at 00 UTC

Friday, 06 March 2009


(a)

0

2

4

6

8

10

12

Hei

ght

(km

)

0 500 1000 1500 2000





0

2

4

6

8

10

12

Hei

ght

(km

)

0 500 1000 1500 2000





0

2

4

6

8

10

12H

eigh

t (k

m)

0 500 1000 1500 2000





(b)

Figure 9: DREAM 6-hour vertical profile of dust forecast for March 5, 2009 (a) and March 6, 2009 (b).

with its center firstly over Corsica in northern TyrrhenianSea and in the process over northern Italy (Figures 12(a)and 12(b)). This pattern combined with high pressure systemon the surface at the south-east Libya, resulted in theprevalence of south air flow towards the region of Greecewith gale winds (∼20 m/s) at 850 hPa level (Figures 13(a) and13(b)) against violent storm winds (29–32 m/s) at 500 hPalevel (Figures 13(c) and 13(d)) becoming hurricane winds

(>33 m/s) on March 6, 2009. The atmospheric circulation inthe middle atmosphere (500 hPa) showed a trough over thecentral Mediterranean and the north-west coast of Africa,which caused an advection of southern air flow towardsGreece (Figures 12(a) and 12(b)). The significant negativeanomaly of the 500 hPa geopotential heights from the meanvalue of 1981–2010 Climatology was very characteristic(−300 m), over central Mediterranean Sea. The anomaly


476

484

492500

508

516

524

532

540

552

560

568

576

584

592

600

February 4, 2009

500 hPa Geopot. (gpdm), T (C) und bodendr. (hPa)

(a)

February 6, 2009


476

484

492500

508

516

524

532

540

552

560

568

576

584

592

600

(b)

40N

30N

35N

45N

50N

55N

60N

65N

70N

75N

80N

20W

15W

10W 5W

0

5E 10E

15E

20E

25E

30E

35E

40E

45E

500 mb geopotential height (m) composite anomaly (1981–2010 climatology)

−300

−200

−100

0

100

200

300

February 4, 2009

(c)

40N

30N

35N

45N

50N

55N

60N

65N

70N

75N

80N

20W

15W

10W 5W

0

5E 10E

15E

20E

25E

30E

35E

40E

45E


−300

−200

−100

0

100

200

300

February 6, 2009

(d)

Figure 10: Spatial distribution of geopotential heights (m) for 500 hPa level along with surface barometric pressure (hPa) ((a) and (b)) andcomposite anomaly from 1981–2010 Climatology of the geopotential heights (m) at 500 hPa level ((c) and (d)) during the Saharan episodeson February 4, 2009 ((a) and (c)) and February 6, 2009 ((b) and (d)), using NCEP/NCAR reanalysis data.

shifted eastwards from south England, France and westMediterranean Sea to Italian Peninsula during the two-day atmospheric episode (Figures 12(c) and 12(d)). Thisestablished synoptic condition was the driver of strong southtransport of dust from central Libya and north-easternAlgeria towards Greece.

5. Discussion

Dust storms originated from Libya are usually transportednorthward throughout the atmospheric column, from nearthe surface to the middle of troposphere. Viewing thesatellite images for a long-time period (∼6 years), thesedust storms are more intense than those originated from

the northwestern Africa (mainly Algeria, Mauritania, andMali), which are the most frequent in the central/easternMediterranean and south Europe [10, 14, 18, 45]). Exceptof their intensity, other significant differences between thedust events coming over Greece from Libya and those fromnorthwestern Saharan are (1) the regional and synopticmeteorology favoring their transport, (2) the seasonality, (3)their vertical extend, and (4) their association with clouds.More specifically, the dust events affecting Greece fromsouthwestern directions, with a source in the desert regionsof northwestern Africa, also cover a great part of the centralMediterranean, before reaching Greece; as a consequence,their intensity is decreased since the larger particles weredeposited near the source. These events are more commonin summer and are mainly driven by an anticyclone over


40N

30N

35N

45N

50N

55N

60N

65N

70N

75N

80N20

W

15W

10W 5W

0

5E 10E

15E

20E

25E

30E

35E

40E

45E

850 mb vector wind (m/s) composite mean

6

9

12

15

18

21

24

February 4, 2009

(a)

40N

30N

35N

45N

50N

55N

60N

65N

70N

75N

80N

20W

15W

10W 5W

0

5E 10E

15E

20E

25E

30E

35E

40E

45E


4

6

8

10

12

14

16

18

February 6, 2009

(b)

40N

30N

35N

45N

50N

55N

60N

65N

70N

75N

80N

20W

15W

10W 5W

0

5E 10E

15E

20E

25E

30E

35E

40E

45E


February 4, 2009

10

15

20

25

30

35

(c)

40N

30N

35N

45N

50N

55N

60N

65N

70N

75N

80N20

W

15W

10W 5W

0

5E 10E

15E

20E

25E

30E

35E

40E

45E


February 6, 2009

10

15

20

25

30

35

40

(d)

Figure 11: Spatial distribution of vector wind composite mean (m/s) at 850 hPa level ((a) and (b)) and 500 hPa ((c) and (b)), during theSaharan episodes on February 4, 2009 ((a) and (c)) and February 6, 2009 ((b) and (d)), using NCEP/NCAR reanalysis data.

northwestern Africa, which transports desert particles overcentral and eastern Mediterranean with air masses followingan anticyclonic pathway; this is the case on June 29, 2002[43] and on August 20–31, 2000 [10]. The duration ofthese dust events is rather large, especially in the centralpart of the Mediterranean, where Meloni et al. [14] founddurations of 13 consecutive days occurred in two episodes,in August 1999 and from 6 to 18 July 2002. The latestwas persistent over a large part of the Mediterranean alsoaffecting Greece. These dust events are mainly detected atan elevated layer into the atmosphere [10, 46] having a clearsignal in AI values [43, 47]. For this reason, the deposition

of the smaller and lighter dust particles still suspendedon the air is not favored, enhancing the duration of thedust plumes. The stable weather and the nearly absence ofprecipitation over Mediterranean in the summer also favorthe dust-aerosol residence time. The DREAM forecasts ofthe dust events on June 29, 2002 [43] and on August 31,2000 [10] show a large vertical extent of the dust plumes,at altitudes as high as 6-7 km. Since the dust particles withinthe boundary layer are easily deposited, while those at middleand upper atmosphere are not due to the reasons mentionedabove, these dust events are mainly detected in the upperatmosphere. Similarly, Kalivitis et al. [17] found that the


March 5, 2009


476

484

492500

508

516

524

532

540

552

560

568

576

584

592

600

(a)

March 6, 2009


476

484

492500

508

516

524

532

540

552

560

568

576

584

592

600

(b)

40N

30N

35N

45N

50N

55N

60N

65N

70N

75N

80N

20W

15W

10W 5W

0

5E 10E

15E

20E

25E

30E

35E

40E

45E


−300

−200

−100

0

100

200

300

March 5, 2009

(c)

40N

30N

35N

45N

50N

55N

60N

65N

70N

75N

80N

20W

15W

10W 5W

0

5E 10E

15E

20E

25E

30E

35E

40E

45E


−250

−200

−150

−100

−50

0

50

100

150

200

250

March 5, 2009

(d)

Figure 12: Spatial distribution of geopotential heights (m) for 500 hPa level along with surface barometric pressure (hPa) ((a) and (b)) andcomposite anomaly from 1981–2010 Climatology of the geopotential heights (m) at 500 hPa level ((c) and (d)) during the Saharan episodeson March 5, 2009 ((a) and (c)) and March 6, 2009 ((b) and (d)), using NCEP/NCAR reanalysis data.

dust events over Crete in summer are mainly transportedin the upper atmosphere. Finally, dust events originatedfrom northwestern Africa in summer are associated withsunny and cloudless conditions over the largest part of theMediterranean and Greece.

In contrast, the dust events originated from Libya affectonly the eastern Mediterranean, as in the present case andon April 17, 2005. Furthermore, a different meteorologicalpattern is responsible for their exposure, mainly driven bya cyclone centered on Adriatic Sea and Italy and an extenttrough reaching the Libyan coast. The result of this atmo-spheric circulation is a northward flow associated with strongsurface and middle-troposphere winds carrying significantamount of dust over eastern Mediterranean and Greece [1].The dust plume in these cases is usually transported through-out the atmospheric column, resulting in both large AOD in

the vertical and increased surface PM concentrations. Thesedust events are more frequent in late winter (although rare)and early spring, since this period favors the depressionsabove the Mediterranean. To this respect, Kalivitis et al. [17]found that the vertical dust transport is more frequent inwinter and spring over Crete, while a similar dust transportmechanism took place on February 26, 2001, as reported byKaskaoutis et al. [43]. The duration of these dust events is 1-2days, since the depressions favoring them are quickly movedand attenuated. They are associated with extent cloud coverover eastern Mediterranean and Greece, which is generatedby the presence of the depression and the uplift of watervapor from the sea. The dust plume is mainly transportedbelow the clouds, at altitudes lower than 4 km as shownin the present cases. Additionally, such were the cases on4th April, 1988 and on March 27, 1992. Both phenomena


40N

30N

35N

45N

50N

55N

60N

65N

70N

75N

80N20

W

15W

10W 5W

0

5E 10E

15E

20E

25E

30E

35E

40E

45E


March 5, 2009

6

9

12

15

18

21

24

(a)

40N

30N

35N

45N

50N

55N

60N

65N

70N

75N

80N

20W

15W

10W 5W

0

5E 10E

15E

20E

25E

30E

35E

40E

45E


4

6

8

10

12

14

16

18

20

22

March 6, 2009

(b)

40N

30N

35N

45N

50N

55N

60N

65N

70N

75N

80N

20W

15W

10W 5W

0

5E 10E

15E

20E

25E

30E

35E

40E

45E


February 4, 2009

10

15

20

25

30

35

40

45

(c)

40N

30N

35N

45N

50N

55N

60N

65N

70N

75N

80N20

W

15W

10W 5W

0

5E 10E

15E

20E

25E

30E

35E

40E

45E


February 4, 2009

10

15

20

25

30

35

40

(d)

Figure 13: Spatial distribution of vector wind composite mean (m/s) at 850 hPa level ((a) and (b)) and 500 hPa ((c) and (d)), during theSaharan episodes on March 5, 2009 ((a) and (c)) and March 6, 2009 ((b) and (d)), using NCEP/NCAR reanalysis data.

are associated with the appearance of depressions that aregenerated in northwest Saharan, to the south of the AtlasMountains especially during spring [48].

6. Conclusions

The present work gives evidence of the synoptic conditionswhich were drivers of intense dust outbreaks over Greece,during the period February-March 2009. The performedanalysis was based upon satellite (MODIS Terra/Aqua)observations, model (DREAM) forecasts, ground-based par-ticulate matter concentrations, and NCEP/NCAR reanalysis

datasets. The synoptic conditions prevailed during theexamined SD events revealed two major synoptic patterns.The first (on February 4, 6, 2009) steered the dust plumeover Greece following anticyclonic track, influencing firstlythe west and central Mediterranean and Italy. Late springand summer are favorable seasons for this type of advection,mainly appeared in the upper layers of troposphere (∼4000 m). In these cases, the dust transport does not affectsignificantly the PM concentrations on the ground, whilethere is not extensive cloud cover over the region, with anexception of the development of icicle clouds. Regardingthe second pattern (on March 5-6, 2009), the dust plume


was transported from Libya northwards, and associated withstrong winds following cyclonic track, while extensive cloudcover over the region was observed as a direct result ofdepression activity. The duration of this type of episodes issmall, one or two days, and these SD are characterized bygreat intensity. The advection of dust concerns mainly thelower tropospheric layers and significantly affects PM con-centrations on the ground. The appearance of such episodesis experienced mainly during winter, against low frequencyduring spring as well. Finally, the overall agreement inthe temporal variation between the AOD550 derived fromMODIS and ground-based PM concentrations, over Athens,both on dusty and nondusty days, suggests that the majorityof aerosol particles are within the boundary layer.

Acknowledgments

The present work was funded by SYNERGASIA 2009PROGRAMME. This Programme is cofunded by the Eu-ropean Regional Development Fund and National Re-sources. (Project code: 09ΣYN-31-711). The author grate-fully acknowledges data and/or images from the BSC-DREAM8b (Dust REgional Atmospheric Model) model,operated by the Barcelona Supercomputing Center (http://www.bsc.es/projects/earthscience/DREAM/), the NOAA AirResources Laboratory (ARL) for the provision of the HYS-PLIT transport and dispersion model and/or READY website(http://www.arl.noaa.gov/ready.php) used in this publicationand the NCEP/NCAR reanalysis project scientific teams.

References

[1] D. G. Kaskaoutis, H. D. Kambezidis, P. T. Nastos, and P. G.Kosmopoulos, “Study on an intense dust storm over Greece,”Atmospheric Environment, vol. 42, no. 29, pp. 6884–6896,2008.

[2] Intergovernmental Panel on Climate Change (IPCC), “Sum-mary for policymakers,” in Climate Change 2007: The PhysicalScience Basis: Contribution of Working Group I to the FourthAssessment Report of the Intergovernmental Panel on ClimateChange, S. Solomon, D. Qin, M. Manning et al., Eds.,Cambridge University Press, Cambridge, UK, 2007.

[3] T. Claquin, M. Schulz, Y. Balkanski, and O. Boucher, “Uncer-tainties in assessing radiative forcing by mineral dust,” TellusB, vol. 50, no. 5, pp. 491–505, 1998.

[4] S. Kinne and R. Pueschel, “Aerosol radiative forcing for Asiancontinental outflow,” Atmospheric Environment, vol. 35, no.30, pp. 5019–5028, 2001.

[5] F. Patadia, E. S. Yang, and S. A. Christopher, “Does dust changethe clear sky top of atmosphere shortwave flux over highsurface reflectance regions?” Geophysical Research Letters, vol.36, no. 15, Article ID L15825, 5 pages, 2009.

[6] J. M. Prospero, P. Ginoux, O. Torres, S. E. Nicholson, andT. Gill, “Environmental characterization of global sources ofatmospheric soil dust identified with the nimbus 7 total ozonemapping spectrometer (TOMS) absorbing aerosol product,”Reviews of Geophysics, vol. 40, no. 1, article 1002, 31 pages,2002.

[7] S. Engelstaedter, I. Tegen, and R. Washington, “North Africandust emissions and transport,” Earth-Science Reviews, vol. 79,no. 1-2, pp. 73–100, 2006.

[8] C. Moulin, C. E. Lambert, U. Dayan et al., “Satellite cli-matology of African dust transport in the Mediterraneanatmosphere,” Journal of Geophysical Research, vol. 103, no. 11,pp. 13137–13144, 1998.

[9] J. Barkan, P. Alpert, H. Kutiel, and P. Kishcha, “Synoptics ofdust transportation days from Africa toward Italy and centralEurope,” Journal of Geophysical Research, vol. 110, no. 7, ArticleID D07208, pp. 1–14, 2005.

[10] A. Papayannis, D. Balis, V. Amiridis et al., “Measurementsof Saharan dust aerosols over the Eastern Mediterraneanusing elastic backscatter-raman lidar, spectrophotometric andsatellite observations in the frame of the EARLINET project,”Atmospheric Chemistry and Physics, vol. 5, no. 8, pp. 2065–2079, 2005.

[11] D. Antoine and D. Nobileau, “Recent increase of Saharan dusttransport over the Mediterranean sea, as revealed from oceancolor satellite (SeaWiFS) observations,” Journal of GeophysicalResearch, vol. 111, no. 12, Article ID D12214, 19 pages, 2006.

[12] S. Rodriguez, X. Querol, A. Alastuey, G. Kallos, and O.Kakaliagou, “Saharan dust contributions to PM10 and TSPlevels in Southern and Eastern Spain,” Atmospheric Environ-ment, vol. 35, no. 14, pp. 2433–2447, 2001.

[13] A. Fotiadi, N. Hatzianastassiou, E. Drakakis et al., “Aerosolphysical and optical properties in the Eastern Mediterraneanbasin, Crete, from aerosol robotic network data,” AtmosphericChemistry and Physics, vol. 6, no. 12, pp. 5399–5413, 2006.

[14] D. Meloni, A. di Sarra, G. Biavati et al., “Seasonal behavior ofSaharan dust events at the Mediterranean island of Lampedusain the period 1999–2005,” Atmospheric Environment, vol. 41,no. 14, pp. 3041–3056, 2007.

[15] G. Kallos, A. Papadopoulos, P. Katsafados, and S. Nickovic,“Transatlantic Saharan dust transport: model simulation andresults,” Journal of Geophysical Research, vol. 111, no. 9, ArticleID D09204, 11 pages, 2006.

[16] G. Kallos, C. Spyrou, N. Papantoniou et al., “Analysis of theparticulate matter exceedances in Greece, Period 2001–2004,”Final Report, The Ministry of Environment City Planning andPublic Work, 2007.

[17] N. Kalivitis, E. Gerasopoulos, M. Vrekoussis et al., “Dusttransport over the Eastern Mediterranean derived from totalozone mapping spectrometer, aerosol robotic network, andsurface measurements,” Journal of Geophysical Research, vol.112, no. 3, Article ID D03202, 9 pages, 2007.

[18] D. Meloni, A. di Sarra, F. Monteleone, G. Pace, S. Piacentino,and D. M. Sferlazzo, “Seasonal transport patterns of intenseSaharan dust events at the Mediterranean island of Lampe-dusa,” Atmospheric Research, vol. 88, no. 2, pp. 134–148, 2008.

[19] J. Dunion and C. Velden, “The impact of the Saharan air layeron atlantic tropical cyclone activity,” Bulletin of the AmericanMeteorological Society, vol. 85, no. 3, pp. 353–365, 2004.

[20] K. V. S. Badarinath, S. K. Kharol, V. K. Prasad, D. G.Kaskaoutis, and H. D. Kambezidis, “Variation in aerosolproperties over Hyderabad, India during intense cyclonicconditions,” International Journal of Remote Sensing, vol. 29,no. 15, pp. 4575–4597, 2008.

[21] K. V. S. Badarinath, S. K. Kharol, A. R. Sharma, V.Ramaswamy, D. G. Kaskaoutis, and H. D. Kambezidis, “Inves-tigations of an intense aerosol loading during 2007 cycloneSIDR—a study using satellite data and ground measurementsover Indian region,” Atmospheric Environment, vol. 43, no. 24,pp. 3708–3716, 2009.

[22] C. Perez, S. Nickovic, J. M. Baldasano, M. Sicard, F. Rocaden-bosch, and V. E. Cachorro, “A long Saharan dust event overthe Western: lidar, sun photometer observations, and regional


dust modeling,” Journal of Geophysical Research, vol. 111, no.15, Article ID D15214, 16 pages, 2006.

[23] I. Carmona and P. Alpert, “Synoptic classification of moderateresolution imaging spectroradiometer aerosols over Israel,”Journal of Geophysical Research, vol. 114, no. 7, Article IDD07208, 2009.

[24] K. N. Grigoropoulos, P. T. Nastos, and G. Feredinos, “Spatialdistribution of PM1 and PM10 during Saharan dust episodesin Athens, Greece,” Advances in Science Research, vol. 3, pp. 59–62, 2009.

[25] P. T. Nastos, A. G. Paliatsos, M. B. Anthracopoulos, E. S. Roma,and K. N. Priftis, “Outdoor particulate matter and childhoodasthma admissions in Athens, Greece: a time-series study,”Environmental Health, vol. 9, no. 1, article 45, 2010.

[26] P. T. Nastos, N. A. Kampanis, K. N. Giaouzaki, and A.Matzarakis, “Environmental impacts on human health duringa Saharan dust episode at Crete Island, Greece,” Meteorologis-che Zeitschrift, vol. 20, no. 5, pp. 517–529, 2011.

[27] E. Samoli, P. T. Nastos, A. G. Paliatsos, K. Katsouyanni, and K.N. Priftis, “Acute effects of air pollution on pediatric asthmaexacerbation: evidence of association and effect modification,”Environmental Research, vol. 111, no. 3, pp. 418–424, 2011.

[28] A. R. Rashki, C. J. de W. Rautenbach, P. G. Eriksson, D. G.Kaskaoutis, and P. Gupta, “Temporalchanges of particulateconcentration in the ambient air over the city of Zahedan,Iran,” Air Quality, Atmosphere & Health. In press.

[29] L. A. Remer, Y. J. Kaufman, D. Tanre et al., “The MODISaerosol algorithm, products, and validation,” Journal of theAtmospheric Sciences, vol. 62, no. 4, pp. 947–973, 2005.

[30] R. C. Levy, L. A. Remer, and O. Dubovik, “Global aerosoloptical properties and application to moderate resolutionimaging spectroradiometer aerosol retrieval over land,” Jour-nal of Geophysical Research, vol. 112, no. 13, Article ID D13210,2007.

[31] S. Nickovic, G. Kallos, A. Papadopoulos, and O. Kakaliagou, “Amodel for prediction of desert dust cycle in the atmosphere,”Journal of Geophysical Research, vol. 106, no. 16, pp. 18113–18130, 2001.

[32] M. Viana, P. Salvador, B. Artano et al., “Assessing theperformance of methods to detect and quantify African dust inairborne particulates,” Environmental Science and Technology,vol. 44, no. 23, pp. 8814–8820, 2010.

[33] K. V. S. Badarinath, S. K. Kharol, D. G. Kaskaoutis, A. R.Sharma, V. Ramaswamy, and H. D. Kambezidis, “Long-rangetransport of dust aerosols over the Arabian sea and Indianregion—a case study using satellite data and ground-basedmeasurements,” Global and Planetary Change, vol. 72, no. 3,pp. 164–181, 2010.

[34] R. R. Draxler and G. D. Rolph, “HYSPLIT—Hybrid SingleParticle Lagrangian Integrated Trajectory Model,” NOAAAir Resources Laboratory, Silver Spring, Md, USA, 2012,http://ready.arl.noaa.gov/HYSPLIT.php.

[35] G. D. Rolph, “Real-time Environmental Applications andDisplay sYstem (READY),” NOAA Air Resources Laboratory,Silver Spring, Md, USA, 2012, http://ready.arl.noaa.gov/.

[36] E. Kalnay, M. Kanamitsu, R. Kistler et al., “The NCEP/NCAR40-year reanalysis project,” Bulletin of the American Meteoro-logical Society, vol. 77, no. 3, pp. 437–471, 1996.

[37] E. Gerasopoulos, G. Kouvarakis, P. Babasakalis, M. Vrekoussis,J. P. Putaud, and N. Mihalopoulos, “Origin and variabilityof particulate matter (PM10) mass concentrations over theEastern Mediterranean,” Atmospheric Environment, vol. 40, no.25, pp. 4679–4690, 2006.

[38] M. Astitha, G. Kallos, and P. Katsafados, “Air pollution mod-eling in the Mediterranean region: analysis and forecasting ofepisodes,” Atmospheric Research, vol. 89, no. 4, pp. 358–364,2008.

[39] Z. Levin, E. Ganor, and V. Gladstein, “The effects of desertparticles coated with sulfate on rain formation in the EasternMediterranean,” Journal of Applied Meteorology, vol. 35, no. 9,pp. 1511–1523, 1996.

[40] C. S. Zerefos, K. A. Kourtidis, D. Melas et al., “Photochemicalactivity and solar ultraviolet radiation (PAUR) modulationfactors: an overview of the project,” Journal of GeophysicalResearch, vol. 107, no. 18, article 8134, 15 pages, 2002.

[41] G. P. Gobbi, F. Barnaba, and L. Ammannato, “Estimating theimpact of Saharan dust on the year 2001 PM10 record ofRome, Italy,” Atmospheric Environment, vol. 41, no. 2, pp. 261–275, 2007.

[42] K. Bouchlaghem, B. Nsom, N. Latrache, and H. H. Kacem,“Impact of Saharan dust on PM10 concentration in theMediterranean Tunisian coasts,” Atmospheric Research, vol. 92,no. 4, pp. 531–539, 2009.

[43] D. G. Kaskaoutis, P. T. Nastos, P. G. Kosmopoulos, and H.D. Kambezidis, “The combined use of satellite data, air-masstrajectories and model applications for monitoring of thedust transport over Athens, Greece,” International Journal ofRemote Sensing, vol. 31, no. 19, pp. 5089–5109, 2010.

[44] D. G. Kaskaoutis, P. G. Kosmopoulos, P. T. Nastos, H. D.Kambezidis, M. Sharma, and W. Mehdi, “Transport pathwaysof Saharan dust over Athens, Greece as detected by MODISand TOMS,” Geomatics, Natural Hazards and Risk, vol. 3, no.1, pp. 35–54, 2012.

[45] C. M. Coen, E. Weingartner, D. Schaub et al., “Saharan dustevents at the jungfraujoch: detection by wavelength depen-dence of the single scattering albedo and first climatologyanalysis,” Atmospheric Chemistry and Physics, vol. 4, no. 11-12,pp. 2465–2480, 2004.

[46] A. M. Tafuro, F. Barnaba, F. De Tomasi, M. R. Perrone, andG. P. Gobbi, “Saharan dust particle properties over the centralMediterranean,” Atmospheric Research, vol. 81, no. 1, pp. 67–93, 2006.

[47] P. Alpert, P. Kishcha, A. Shtivelman, S. O. Krichak, and J. H.Joseph, “Vertical distribution of Saharan dust based on 2.5-year model predictions,” Atmospheric Research, vol. 70, no. 2,pp. 109–130, 2004.

[48] N. G. Prezerakos, A. G. Paliatsos, and K. V. Koukouletsos,“Diagnosis of the relationship between dust storms over theSaharan desert and dust deposit or coloured rain in theSouth Balkans,” Advances in Meteorology, vol. 2010, Article ID760546, 14 pages, 2010.

Submit your manuscripts athttp://www.hindawi.com

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

ClimatologyJournal of

EcologyInternational Journal of


EarthquakesJournal of


Hindawi Publishing Corporationhttp://www.hindawi.com

Applied &EnvironmentalSoil Science

Volume 2014

Mining


Journal of

Hindawi Publishing Corporation http://www.hindawi.com Volume 2014

International Journal of

Geophysics

OceanographyInternational Journal of


Journal of Computational Environmental SciencesHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Journal ofPetroleum Engineering


GeochemistryHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Journal of

Atmospheric SciencesInternational Journal of


OceanographyHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Advances in


MineralogyInternational Journal of


MeteorologyAdvances in

The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014

Paleontology JournalHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

ScientificaHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014


Geological ResearchJournal of


Geology Advances in

Documents

MeteorologicalPatternsAssociatedwithIntense …downloads.hindawi.com/journals/amete/2012/828301.pdf · 2019. 7. 31. · the regional and synoptic weather forecast models, also able