The Study of Desert Dust in Mashhad Metropolis Using Satellite Images and Synoptic Datasets (2009 - 2013)

Document Type : مقاله پژوهشی


Ferdowsi University of Mashhad


1. Introduction
In recent years, Desert Dust frequency has increased in the Global system. Due to lack of vegetation in the regions prone to dust, the earth’s surface air becomes warm enough in these regions, moves upward and causes a cold downward circular air flow when it collides with upper tropospheric winds. When these winds strongly hit the earth’s surface, they create dust storms (Xuan et al, 2004). Many studies from all around the world have been conducted on the source regions of this environmental phenomenon (Kai & Gao, 2007). The studies on Asian dust storms with sea sources during 2000-2002 showed 64% of storms originate from China seas and become more severe from west to east and carry dust on the land. Wang, Stein, Draxler, Rosa & Zhang (2011) studied sand and dust storms in 2008 and defined North Africa, Middle East, Mongolia, and north-west China with high frequency of dust phenomenon, using HYSPLIT Model. Shamsipour and Safar-Rad (2012), analyzed the July 2009 dust in the west of Iran. In the chosen sample, the Zagros hillsides contain the highest amount of pollutants. Also, the results indicate that the location of Naveh, upper-level divergence of 500 and forming a thermal low-pressure system on earth’s surface play the main role in flowing the dust toward Iran.

2. Study Area
Study area includes the Mashhad metropolis which is located in the northeastern part of Iran. As the second most populated metropolis in Iran and due to its vicinity to northeast deserts and strong winds of summer, Mashhad is threatened by this environmental hazard. Therefore, to be aware of this disaster, knowing about its trend and source locations are essential for the study. The studies on particulates and dust in Mashhad are based on analysis of synoptic or atmospheric circulation patterns or statistical methods using air quality monitoring equipment or weather stations. In addition, no study has clearly demonstrated the source location of dust in Mashhad. Because of the necessity of air pollutant observation and control in metropolises and dissipation or lack of weather stations, it is necessary to use other data sources to check the air quality such as remote sensing data and satellite imagery. Therefore, the aim of this study is to use HYSPLIT model outputs by analyzing MODIS satellite images in combination with synoptic methods and tracing the dust with retrograde method.
3. Material and Methods

3-1. Data
The study materials include meteorological records relating to 06 dust code, synoptic maps and MODIS satellite images in dusty days. The data of the first group was obtained from Khorasan Razavi Meteorological Organization for the period of 2009-2013 and Mashhad synoptic station at intervals of 8 hours. The second group includes gridded climate data with 2.5×2.5° longitude and latitude taken from NCEP/NCAR website. The synoptic maps of dusty days in Gradis setting in 25-55° north latitude and 35-85° east longitude were drawn and analyzed. The third group includes calibrated level-1 MODIS atmospheric product with 1 km resolution. Totally, 16 concurrent images of dusty days were obtained between March 3, 2009 and October 30, 2013 from GODDARD NASA website. The fourth group includes the average daily data of PM10 from air quality monitoring stations in 2011which were used in the same year in order to assess the validity of AOD daily data of MODIS. The fifth group is the HYSPLIT Lagrangian output model with retrograde method for tracing the dust source of Mashhad. In this research, we used HYSPLIT model to locate the path and source of dust that reach Mashhad outside the station. The model input data were extracted from NCEP/Reanalysis data. Also, the flow measurement method was isobaric lines with a 6-hour time step and altitudes of 50 and 1000m above sea level.

3-2. Methods
The present research is an analytic-descriptive case study with a quantitative method for data analysis and uses descriptive statistics and spatial analysis. Based on meteorological records, we have first analyzed temporal trend of Dust phenomenon in hourly, monthly, seasonally and annually durations. Then we documented spatial distribution of Dust on urban areas by MODIS images which is obtained from NASA website. These images were processed by two indicators: 1- Brightness Temperature Index (BTD31,32) which is suggested by Ackerman (1997) and, 2- Normalized Difference Dust Index (NDDI). BTD measures between 11-12 micrometer and MNDVI parameter were used to separate the desert arid regions and threshold temperature of 290° Kelvin in the band 32 for separating cloudy from dusty regions.

4. Results and Discussion
4-1. frequency of dusty days
The observation hours on July 3rd, 6th, 9th, June 12th, March 12th and 15th had the highest average dust. In contrast, January, February and December had the lowest dusty hours. July, June, March and August had the highest and January, February and December had the lowest monthly dusty days during the above-mentioned period. Also, the highest number of dusts in this period occurred in 2009.
4-2. Satellite Image Analysis
4-2-1. Validation of AOD MODIS Images and PM10 Georeferenced Data
In this analysis, about 69% of PM10 dependent variable changes are defined by the MODIS AOD 550nm independent variable which shows a fine correlation between the two validated parameters.
4-2-2. Dust Identification Index with BTD31, 32
The implementation of the index algorithm led to clarification of dust on the image bands in four calculation steps. According to the models mentioned here, the dust local algorithm in Mashhad shows that the northwest, west and south parts of this city were covered with dust. Therefore, it seems the districts 1, 8, 9, 10, 11, 12 located in these regions experience more dust compared with other districts. Also, the analyses show that compared with city center, outskirts of Mashhad are subject to more dust pollution.
4-3. Synoptic Analysis of Dust
In order to study the synoptic conditions in dusty days, the synoptic maps concurrent with images at 12 pm local time were analyzed. In dusty days, the middle level of high-pressure atmosphere with negative Omega dominates Iran which leads to dust due to lack of humidity. Considering the path and speed of the wind at the earth’s surface and domination of flows in a north-south pattern, it seems the dust moves from its focal points towards north-east of Iran and Mashhad. Generally, according to the dusty days map we can say that the main and closest source of Mashhad dusts lies to the east of the region and Afghanistan and east of Turkmenistan.

4-4. Tracing the Path of Dust
As soon as the firs dust was reported, the path of particles was examined since then back to 24 hours. The model’s output maps show 3 main paths which transfer the dust to Mashhad. In most cases with Turkmenistan and Central Asia as the source, the main path of dust was from north-east to south-west mostly during the warm seasons of the year. The second path is from north-west to south-east in cold seasons. The third path is from south-west to north-east which has a low frequency with the cold seasons domination.

5. Conclusion
According to meteorological dust records of Mashhad synoptic station, the maximum and minimum frequencies of daily dusts occur at morning and midnight hours respectively in local time in most seasons during the statistical period of the study. In monthly view, most of Mashhad dusts occur in July, June, August, and March, respectively. The minimum number of monthly records without dust in December occur during the cold period of the year. The BTD index properties was able to clarify the regions covered with dust on satellite images in Mashhad. It also provided a synoptic analysis of dusty days to identify the dust and find the way by which it is transferred. It was specified from a dusty days sample with a synoptic approach that Afghanistan and part of Turkmenistan located in the north-east of Mashhad are the main focal points of dust that reach Mashhad. The HYSPLIT model maps output overlap synoptic maps of dusty days. Generally, during the period of study, north-west, west and south of Mashhad were covered with dust. The districts in these regions experience more dust conditions compared with other Mashhad districts.


امیدوار، کمال. نکونام، زری، (1390)، کاربرد گلباد و گل غبار در تحلیل گرد و خاک و تعیین رژیم فصلی بادهای همراه با این پدیده مطالعه موردی شهر سبزوار. پژوهش‌های جغرافیای طبیعی، 76: 85-104.
جهانشیری، مهین، (1389)، بررسی آماری و سینوپتیکی آلودگی هوای مشهد، پایان نامه کارشناسی ارشد جغرافیای طبیعی-اقلیم شناسی، به راهنمایی علیرضا انتظاری، دانشگاه حکیم سبزواری، سبزوار.
حسین زاده، سید رضا، (1376)، بادهای 120 روزه سیستان. تحقیقات جغرافیایی، 46: 103-127.
حسین زاده، سید رضا. دوستان، رضا. حقیقت ضیابری، سیده مرضیه. حقیقت ضیابری، سیده مریم، (1392)، بررسی الگوهای همدید مؤثر بر آلودگی هوا در کلان‌شهر مشهد. جغرافیا و توسعه ناحیه‌ای. 21: 81-102.
ذوالفقاری، حسن. عابد زاده، حیدر، (1384)، تحلیل سینوپتیک سیستم های گرد و غبار در غرب ایران. مجله جغرافیا و توسعه. 6: 173-188.
شمسی‌پور، علی‌اکبر. صفر راد، طاهر، (1391)، تحلیل ماهواره ای-همدیدی پدیده گرد و غبار (گرد و غبار تیرماه 1388)، پژوهش‌های جغرافیای طبیعی، 79: 111-128.
صفری، اعظم، (1392)، شناسایی منابع غبار طوفان‌های گرد و غباری جنوب غرب ایران. پایان‌نامه کارشناسی ارشد گرایش اقلیم‌شناسی در برنامه‌ریزی محیطی. به راهنمایی عباس مفیدی، دانشگاه فردوسی مشهد. مشهد.
علوی پناه، سیدکاظم، (1382)، کاربرد سنجش از دور در علوم زمین. مؤسسه انتشارات و چاپ دانشگاه تهران. چاپ اول. ص271.
علیجانی، بهلول. نجفی‌نیک، زهرا، (1388)، بررسی سینوپتیکی اینورژن در مشهد با استفاده از تحلیل عاملی. مجله جغرافیا و توسعه ناحیه‌ای، 12: 1-12.
قربانی، محمد. فیروز زارع، علی، (1387)، مقدمه‌ای بر ارزش‌گذاری محیط‌زیست. انتشارات دانشگاه فردوسی مشهد. چاپ 1389، ص 156.
قربانی، محمد. فیروز زارع، علی، (1388)، ارزش‌گذاری ویژگی‌های مختلف آلودگی هوا در مشهد. مجله تحقیقات اقتصادی، 89: 215-242.
کاخکی مهنه، حمید، (1392)، استفاده از مدل شبکه عصبی مصنوعی در بررسی روابط بین عناصر اقلیمی و نوسانات آلودگی هوا در کلان‌شهر مشهد. پایان‌نامه کارشناسی ارشد، گرایش اقلیم شناسی در برنامه‌ریزی محیطی، به راهنمایی محمد سلیقه، دانشگاه خوارزمی، تهران.
گلریز ضیائی، زهرا. صادقی، حمید. معتمدی، محمد. ژاله رجبی، میترا، (1390)، پیش بینی، ارزش‌گذاری و ترسیم نقشه آلودگی هوای شهر مشهد؛ مطالعه موردی گاز منوکسید کربن، 2 و 3 آذر ماه. اولین کنفرانس اقتصاد شهری ایران. مشهد.
مباشری، محمدرضا. غفوری، علی، (1385)، آشکارسازی و مطالعات غبار جوی در تصاویر ماهواره‌ای با استفاده از اختلاف دمای درخشندگی، نقشه‌برداری، 84: 19-25.
مفیدی، عباس. جعفری، سجاد، (1390)، بررسی نقش گردش منطقه‌ای جو بر روی خاورمیانه در وقوع توفان‌های گرد و غبار تابستانه در جنوب غرب ایران. مطالعات جغرافیای مناطق خشک، 5: 17-46.
منصوری، نبی‌اله. علی‌نژاد، جواد، (1388)، بررسی کارایی سیستم‌های کنترل کننده ذرات معلق و تعیین میزان انتشار در کارخانجات آسفالت مشهد. فصلنامه علوم تکنولوژی محیط زیست، 40: 1-9.
Ackerman, S. A., Chung, H. (1992). Radiative effects of Airborne dust on regional energy budgets at the top of atmosphere. Journal of Applied Meteorology, 31, 223-233.
Edwards, D. P., DeCola, J., Fishman, D., Jacob, P., Bhartia.D., Burrows, J., Goldberg. M. (2006). Community input to the NRC decadal surrey from the NCAR workshop on air quality remote sensing from space: defining an optimum observing strategy. Journal of National Center for Atmospheric Research Boulder, retrieved from:
Ghrbani, M., & Firoz Zaree, A. (2009). Measurement of air pollution characteristics in Mashhad. Economic Research Journal, 89, 215-242. [in Persian]
Jahanshiri, M. (2010). Statistical and synoptic investigations of air pollution of Mashhad Unpublished master's thesis, University of Hakim Sabzevari, Sabzevar. [in Persian]
Kai, Z., & GAO, H. (2007). The characteristics of Asian-dust storms during 2000-2002: From the source to the sea. Journal of Atmospheric Environment,41, 9136-9145.
Karimi, N., Moridnejad,A., Golian, S., M.V., Samani, J., Karimi, D., & Javadi, S., (2012). Comparison of dust source identification techniques over land in the Middle East region using MODIS data. Canadian Journal of Remote Sensing, 38(5), 586-899.
Kaufaman, Y. J. (1997). Passive remote sensing of tropospheric aerosol and atmospheric correction for the aerosol effect. Journal of Geophysical Research, 102, 16815-16830.
Kaufman, Y. J., & Tanre, D. (1998). Remote sensing of tropospheric aerosols from EOS Modis overland. Journal of Geophysical Research, 102, 17051-17067.
Mansoori, N., & Alinejad, J. (2009). The Study of efficiency of suspension particles controller systems and determination of distribution amounts in Mashhad asphalt companies. Environmental Technology Sciences, 40, 1-9.
Miller, S. D. (2003). A consolidated technique for enhancing desert dust storms with MODIS. Journal of Geophysical Research Letters, 30(20), 2071-2074.
Mobasheri, M. R., & Ghafori, A. (2006). Detection of atmospheric dusts in satellite images using Brightness Temperature Difference. Survey, 84, 19-25. [in Persian]
Omidvar, K., & Nokonam, Z. (2011). Application of wind rose and dust rose in dust analysis and determination of seasonal wind regime related with this phenomenon (Case study: Sabzevar city). Geographical Research, 76, 85-104. [in Persian]
Safari, A. (2013). Recognition of dust storm sources in southwestern of Iran. Unpublished master's thesis, Ferdowsi University of Mashhad, Mashhad. [in Persian]
Shunlin, L., Zong, B., & Hongliang, F. (2006). Improved estimation of aerosol optical depth from MODIS imagery over land surface. Journal of Remote Sensing Of Environment, 104, 416-425.
Tom, X. P., Zhao, Ackerman, S, Guo, W. (2010). Dust and smoke detection for multi-channel imagers, Journal of Remote Sensing, 2, 2347-2368.
Wang, S., Yang, D., Jing, J., Xu, Q., Yang, Yu. (1995). Study on the formative causes and countermeasures of the catastrophic sandstorm occurred in Northwest China. Journal of Desert Research, 15(1), 19-30.
Wang, Y., Stein. A., Draxler, R., Rosa, D., Zhang, X. (2011). Global sand and dust storms in 2008: Observation and HYSPLIT model verification. Journal of Atmospheric Environment, 45, 6368-6381.
Wang. E., Qiming. L., Na. Y., Xianhua. W., & Jinji. M. (2013). Air quality evaluation on an urban scale based on MODIS satellite images. Journal of Atmospheric Research, 132-134, 22-34.
Xia, L. (2010) . Sea surface temperature extraction by modis remote sensing data in Lianyungang sea area. Paper presented at the Second IITA International Conference on Geoscience and Remote Sensing. Qingdao, 118-120.
Xuan J., Sokolik IN., Hao J., Guo F., Mao H., & Yang G. (2004). Identification and characterization of atmospheric mineral dust in East Asia. Journal of Atmospheric Environments, 38(36), 6239-6252.