Dust aerosol in the atmosphere has a significantly direct and indirect effect on climate on various spatial scales, from local and regional to global [1].

In arid and semiarid areas, dust storms occur frequently. Sand and dust storms (SDS) have caused devastating damages to properties and human health in the southwest of Asia. When surface winds are strong, large amounts of sand and dust can be lifted from bare, dry soils into the atmosphere and transported downwind affecting regions hundreds to thousands of kilometers away. Jaffe et al.

(2003) [2] have shown that extreme episodes of dust transport can adversely impact air quality in regions far from the original emission source. Zouljoodi et al. (2013) [3] have developed a research on dust events in western Iran and drought expansion over the Middle East, through this investigation they have detected firstly the dust source areas over Iraq and Syria, then it is found that the recent droughts in the external dust source areas (over Iraq and Syria) had the remarkable potential to increase the dust events in the west of Iran.

Joseph M. and Peter J. (2003) [4], have considered African droughts and dust transport to the Caribbean, so they found that, because of the great sensitivity of dust emission to climate, future changes in climate could result in large changes in emissions from African and other arid regions that, in turn, could lead to impacts on climate over large areas.

In the last few decades numerical models have been designed to reproduce the dust cycle allowing us to estimate the influence of mineral dust on the climate system [5]. Dust models are also required for short to medium range air quality forecast applications. To this end, several models have been developed with some of them providing daily forecast products, e.g. the Dust Regional Atmospheric Model (DREAM) [6], the SKIRON model [7], CHIMERE-Dust model [8], the Navy Aerosol Analysis and Prediction System (NAAPS) [9], the JMAMASINGAR dust model [10], or the ECMWF-IFS [11]. The DREAM model was further re-fined and tested in the last years. The updated version BSCDREAM8b [12- 17] is operated.

K. Haustein et al. (2012) [5] have developed a research on atmospheric dust modeling from meso to global scales with the online NMMB/BSC-Dust model—

Part 2: Experimental campaigns in Northern Africa. The new NMMB/BSC-Dust model is intended to provide short to medium-range weather and dust forecasts from regional to global scales.

In the recent years a lot of studies have been developed in Iran on the frequency of dust storm events (from statistical point of view) and the consideration of synoptic meteorological patterns which are leading to the dust storms. However we had not yet operated a reliable forecasting model of dust storms in Iran with a capability to demonstrate the dust source areas. Through this investigation we will present the derived outputs from WRF/ CHEM and HYSPLIT models which are considered a basic step to high resolution dust forecasting, recognizing of dust sources, dust storm trajectory and dust concentration rate. Meantime by using of the visibility datasets from the selected stations we illustrated the long term variability of the dust in monthly and yearly scales. The satellite images of 22-26 June 2010 have been taken into consideration to comparing of the derived out puts from the applied models.

In this work, we have used the images derived from the Meteosat-9¹ for comparing with the outputs of the applied models. This satellite (Meteosat-9) is placed in a height of 35,800 km. over the equator (lat; 00˚:00') and prime meridian (lon; 00˚:00'). This satellite is the second generation of EUMETSAT² satellites which were known already as MSG-2³ that has been operated in 2005. The Meteosat-9 covers Europe, Africa, Middle East, some parts of the south eastern America, Atlantic Ocean and western side of the India Ocean. The images of Meteosat-9 through 12 spectrum channels are produced for each 15 minutes.

In order to run the models (WRF/CHEM and HYSPLIT) the FNL datasets have been used as the inputs of the models. The FNL (final) datasets in fact are the NCEP⁴ datasets which have been reanalyzed through the NOAA⁵ related laboratories. These datasets with a spatial resolution of 1˚ × 1˚ are available from Jul 1999, for 26 atmospheric levels (100 - 1000 hPa.) with the 6-hour time intervals. The FNL datasets use the GDAS⁶ for providing the initial data. The GDAS also gathers persistently the observation based datasets worldwide through the GTS⁷ and other resources [20].

Meantime the in situ visibility/dust datasets are gathered through the information technology administration of the IR of Iran meteorological organization.

The developed studies on dust events over Iran demonstrated a high frequency of dust in the west and southwest of the country due to the external dust sources in Iraq and Syria [3], so we have chosen basically 5 stations over western and southwestern parts of the country which include: Sanandaj, Kermanshah, Ahvaz, Abadan and Boushehr synoptic stations (Figure 1). Also through the satellite images and the applied models the Middle East countries, especially Iraq and Syria, have been taken into account through this investigation.

Dust storm as one of natural hazards which affect seriously economy, health, environment, etc. in the recent decade (especially in the recent years) has been increased in the Middle East countries and particularly over west and southwest of Iran. Atmospheric circulations, land surface cover and soil moisture deficit are considered the principal factors of dust rising. Studying the mechanism of dust storm occurrence, dust trajectory, velocity and concentration are so important in dust preparedness plans.

In this research, firstly, we have considered the frequency of dust events based on the hourly visibility dataset from 5 synoptic stations in west and southwest of the country during a 30-year period from 1980 to 2010. The obtained results obviously indicate the increasing of dust events in Iran especially over southwestern parts of the country in the recent decade/years. For example, Ahvaz synoptic station in the southwest with a long term yearly mean (in 30 years) of dust frequency about 120 dusty days, illustrates an increasing trend in the recent decade with a yearly mean of 221 dusty days, and in contrast, the results of other selected stations have been placed in the first rank. In a monthly analysis, we found that the highest frequency of dust events occurre at June and July, while the cold months show the lowest dust frequency. Through the analysis of the satellite images from Meteosat-9 over 22-26 June 2010, we find that during 2 days of 23 and 24 June, the dust reach the west and southwest of Iran and the visibility is decreased. And from 25 to 26 June gradually the visibility over the west-

ern and then southwestern parts of the country has been changed to the normal status.

In the next step, we apply the dust simulation models for 22-26 June 2010. The HYSPLIT model that determines the dust sources and dispersion trajectories, illustrates that the sources of the aerosols suspended in the high levels of the atmosphere (1000 m - 3000 m) are in Syria or western parts of Iraq, whereas the particles in the height of below 100 m which include relatively coarse particles and influence seriously the visibility and health come from the sources in the centre or east parts of Iraq.

And the WRF/CHEM model has simulated the concentration/density rate of the suspended particles in the atmosphere. The results obtained from this model indicate a high accordance with the recorded visibility via the stations and the results of the satellite images analysis about dust accumulations and movements over the study area.

Briefly, the found results through this work imply that

the outputs of the applied models would be reliable to use in the dust early warning system which is considered a fundamental step of dust preparedness plan over the study area.