Desert Dust Mapping and Identification Using MODIS Level 1 and AOD- AI Indices in South West of Iran

Authors

‎10.22052/deej.2021.10.33.39

Abstract

Extended Abstract
Introduction: Dust storm is a major weather event that plays an important role in the Earth’s atmosphere and ocean surfaces. Dust storms affect marine phytoplankton, soil physical and chemical characteristics in entrainment and deposition areas, climate and radiative forcing, economic loss, and human health. They mostly originate from Introduction: Dust storm is a major weather event that plays an important role in the Earth’s atmosphere and ocean surfaces. Dust storms affect marine phytoplankton, soil physical and chemical characteristics in entrainment and deposition areas, climate and radiative forcing, economic loss, and human health. They mostly originate from plains and playas across North Africa, the Middle East and Asia. One of the main sources of dust is relatively recent flood sediments deposited since the late tertiary. Asia, with approximately 60% of the population of the world, is an important source of dust that impacts the climate on a global scale. A dust storm is defined by the World Meteorological Organization (WMO) protocol as where ‘strong winds lift large quantities of dust particles, reducing visibility to between 200 and 1000 m’. Therefore, the aim of present study was to investigate the potential of desert dust mapping algorithms using satellite images in the west and western parts of Iran.

Materials and methods: The study area has hot and arid climate; therefore, dust storms are usually occurred especially during dry seasons in this region including Iran, Iraq, Syria, Jordan, Kuwait, Saudi Arabia, Bahrain, Qatar, UAE, Oman, and Yemen. Although, some of dust in the study area comes from the Sahara in the Africa. In recent years, the most environmental problem in Iran is the dust crisis in the western half of the country in mountainous regions of Zagros. Zagros Mountains occupy a broad extent of western Iran, covering an area roughly 1300 km by 200 km. The mean annual rainfall is from 400 to 800 mm and mostly in the winter and spring. The average annual temperature ranges between 9°C and 25°C. The most common ecosystems in the region are the forest and semi-steppe areas. Forests with an area of 5 million hectares cover about 40% of Iran’s forests and are the widest forest regions of the country. This region with its semi-arid climate is generally dominated by broad-leaved trees with the dominant species of Quercus brantii that covers more than 50% of Zagros’ forest area. Dust storms from western neighbouring countries such as Iraq have significantly increased in recent years and much of Iran is affected. For example, 20 and 52 dusty days in 2008 with visibility less than 1000 m occurred in Somar and Abadan meteorological stations in Kermanshah and Khozestan provinces, which are the nearest stations to Iraq. Around 31% of Iraq’s total land is desert, which is the main source of soil-derived mineral dust in the region. There are three important sources of dust storms in Iraq: one is centred over Baghdad; the second is centred west of Basrah; and the third is the Southern Desert. Thus, dust cases are common in central and southern Iraq. The major hotspots for dust generation are aeolian deposits south of Baghdad, Karbula, Najaf, Nasiriya, Basrah, as well as Kuwait, and also alluvial deposits of Tigris and Euphrates Rivers from sandy clayey silt (72%) to clayey sandy silt (28%). In addition to the high potential of Iraq’s desert deposits for dust production, recent changes in the region such as war in Iraq, dam construction in Iraq and neighbouring countries, and intense drought conditions have increased the frequency of dust events. Saudi Arabia is rich in fine sediments from dry riverbeds and lakebeds and sand seas, and is another important dust source in the region. Therefore, for mapping this environmental crisis, MODIS level 1 B satellite images were acquired in winter and summer seasons and processed with MCT tool. Then, the BTD, Ackerman, Miller and TDI algorithms were applied to the images and dust regions were mapped with use of appropriate thresholds. The accuracy of the outputs maps were assessed against natural color composites and dust products including Aqua AOD and Aura AI.

Results: Results showed that a single method for identifying different dust plumes cannot be used due to differences in mineralogy and weather conditions of the events and each algorithm needs different threshold based on the event type and characteristics. The algorithms worked best on dense dust, but they performed differently in cloudy regions and over bright desert surfaces. Most of the algorithms examined here misidentified thick cloud cover as dust. Despite the published dust/no-dust thresholds for the methods tested here; results indicated that it was not possible to use a single dust/no-dust threshold for any of the algorithms applied to the studied events. Therefore, it seems that for each dust event an event-specific threshold is needed.

Discussion and conclusion: Comparison of the studied algorithms showed that all of them produced almost similar results and, among them, the TDI index had relatively better performance over dust sources and showed its usefulness as an effective approach for dust detection and mapping in the region. It appears that the combination of these simple algorithms is the best way to overcome the limitations of different dust detection methods. By combining several algorithms used in this study, the performance of dust detection and mapping may improve. The finding indicated that free and available dust products are sufficient for assessment and monitoring dust storms in Iran if the maps less than 10 km are not required.
plains and playas across North Africa, the Middle East and Asia. One of the main sources of dust is relatively recent flood sediments deposited since the late tertiary. Asia, with approximately 60% of the population of the world, is an important source of dust that impacts the climate on a global scale. A dust storm is defined by the World Meteorological Organization (WMO) protocol as where ‘strong winds lift large quantities of dust particles, reducing visibility to between 200 and 1000 m’. Therefore, the aim of present study was to investigate the potential of desert dust mapping algorithms using satellite images in the west and western parts of Iran.
Materials and methods: The study area has hot and arid climate; therefore, dust storms are usually occurred especially during dry seasons in this region including Iran, Iraq, Syria, Jordan, Kuwait, Saudi Arabia, Bahrain, Qatar, UAE, Oman, and Yemen. Although, some of dust in the study area comes from the Sahara in the Africa. In recent years, the most environmental problem in Iran is the dust crisis in the western half of the country in mountainous regions of Zagros. Zagros Mountains occupy a broad extent of western Iran, covering an area roughly 1300 km by 200 km. The mean annual rainfall is from 400 to 800 mm and mostly in the winter and spring. The average annual temperature ranges between 9°C and 25°C. The most common ecosystems in the region are the forest and semi-steppe areas. Forests with an area of 5 million hectares cover about 40% of Iran’s forests and are the widest forest regions of the country. This region with its semi-arid climate is generally dominated by broad-leaved trees with the dominant species of Quercus brantii that covers more than 50% of Zagros’ forest area. Dust storms from western neighbouring countries such as Iraq have significantly increased in recent years and much of Iran is affected. For example, 20 and 52 dusty days in 2008 with visibility less than 1000 m occurred in Somar and Abadan meteorological stations in Kermanshah and Khozestan provinces, which are the nearest stations to Iraq. Around 31% of Iraq’s total land is desert, which is the main source of soil-derived mineral dust in the region. There are three important sources of dust storms in Iraq: one is centred over Baghdad; the second is centred west of Basrah; and the third is the Southern Desert. Thus, dust cases are common in central and southern Iraq. The major hotspots for dust generation are aeolian deposits south of Baghdad, Karbula, Najaf, Nasiriya, Basrah, as well as Kuwait, and also alluvial deposits of Tigris and Euphrates Rivers from sandy clayey silt (72%) to clayey sandy silt (28%). In addition to the high potential of Iraq’s desert deposits for dust production, recent changes in the region such as war in Iraq, dam construction in Iraq and neighbouring countries, and intense drought conditions have increased the frequency of dust events. Saudi Arabia is rich in fine sediments from dry riverbeds and lakebeds and sand seas, and is another important dust source in the region. Therefore, for mapping this environmental crisis, MODIS level 1 B satellite images were acquired in winter and summer seasons and processed with MCT tool. Then, the BTD, Ackerman, Miller and TDI algorithms were applied to the images and dust regions were mapped with use of appropriate thresholds. The accuracy of the outputs maps were assessed against natural color composites and dust products including Aqua AOD and Aura AI.
Results: Results showed that a single method for identifying different dust plumes can not be used due to differences in mineralogy and weather conditions of the events and each algorithm needs different threshold based on the event type and characteristics. The algorithms worked best on dense dust, but they performed differently in cloudy regions and over bright desert surfaces. Most of the algorithms examined here misidentified thick cloud cover as dust. Despite the published dust/no-dust thresholds for the methods tested here; results indicated that it was not possible to use a single dust/no-dust threshold for any of the algorithms applied to the studied events. Therefore, it seems that for each dust event an event-specific threshold is needed.
Discussion and conclusion: Comparison of the studied algorithms showed that all of them produced almost similar results and, among them, the TDI index had relatively better performance over dust sources and showed its usefulness as an effective approach for dust detection and mapping in the region. It appears that the combination of these simple algorithms is the best way to overcome the limitations of different dust detection methods. By combining several algorithms used in this study, the performance of dust detection and mapping may improve. The finding indicated that free and available dust products are sufficient for assessment and monitoring dust storms in Iran if the maps less than 10 km are not required.

Keywords


  1. Ackerman, S. A., 1989. Using the radiative temperature difference at 3.7 and 11 μm to tract dust outbreaks. Journal of Remote Sensing of Environment 27(2), 129-133‌.
  2. Al-Askary, H., Gautam, R., Singh, R. and Kafatos, M., 2006. Dust storms detection over the Indo-Gangetic basin using multi sensor data. Journal of Advances in Space Research 4, 728-733‌.
  3. Alidadi, S., 2014. Dust prediction, trajector and mapping using WRF and HYSPLIT models and satellite images. M.Sc. Thesis, Isfahan University of Technology. 171 pp.
  4. Baddock, C., Jonna, E. and Bullard, G., 2009. Dust source identification using MODIS: A comparison of techniques applied to the Lake Eyre Basin, Australia. Journal of Remote Sensing Environment, 1511-1528‌.
  5. Bahrami, H.A.Jalali, M.Darvishi Boloorani, A. and Azizi, R., 2013. Spatio-temporal modeling of dust storm events in Khuzestan province. Iranian Journal of Remote Sensing & GIS, 5 (2), 95 – 114.
  6. Broomandi, P., Karaca, F., Guney, M., Fathian, A., Geng, X. and Kim, J.R., 2021. Destinations frequently impacted by dust storms originating from southwest Iran. Atmospheric Research 248, 105264.
  7. Chakravarty, K., Vincent, V., Vellore, R., Srivastava, A.K., Rastogi, A. and Soni, V.K., 2021. Revisiting Gandhi in northern India: A case study of severe dust-storm over the urban megacity of New Delhi. Urban Climate 37, 100825.
  8. Ebrahimi Khusfi, Z., Roustaei, F., Ebrahimi Khusfi, M. and Naghavi, S., 2020. Investigation of the relationship between dust storm index, climatic parameters, and normalized difference vegetation index using the ridge regression method in arid regions of Central Iran. Arid Land Research and Management 34, 239-263.
  9. Ebrahimi-Khusfi, Z., Mirakbari, M. and Soleimani-Sardo, M., 2021. Aridity Index Variations and Dust Events in Iran from 1990 to 2018. Annals of the American Association of Geographers, 1-18.
  10. Ebrahimi Khusfi, Z., Khosroshahi, M., Roustaei, F. and Mirakbari, M., 2020. Spatial and seasonal variations of sand-dust events and their relation to atmospheric conditions and vegetation cover in semi-arid regions of central Iran. Geoderma 365, 114225.
  11. Esmaili, O., Tajrishy, M. and Daneshkar, P., 2006. "Result of the 50 years ground-based measurement in comparison with satellite remote sensing of two prominent dust emission sources located in Iraq.", proceedings of SPIE Europe conference on remote sensing of clouds and the Atmosphere XI. Stockholm, Sweden.
  12. Goudie, A. S. and Middleton, N., 2006. Desert dust in the global system, Springer, Berlin‌
  13. Habib, A., Chen, B., Khalid, B., Tan, S., Che, H., Mahmood, T., Shi, G. and Butt, M.T., 2019. Estimation and inter-comparison of dust aerosols based on MODIS, MISR and AERONET retrievals over Asian desert regions. Journal of Environmental Sciences 76, 154-166.
  14. Hao, X. and Qu, J., 2007. Saharan dust storm detection using moderate resolution imaging spectroradiometer thermal infrared bands. Journal of Applied Remote Sensing 1, doi: 10.1117/ 1.2740039.
  15. Hoseinabadi, S., Yaghoobzadeh, M. andForozanmehr, M., 2020. Detecting dust and analyzing its effect on Modis satellite photos: A case study of the city of Zabol. Journal of Geographical Research in Desert Regions, 8 (1), 167-186.
  16. Kheirandish, Z., Jamali, J. and Rayegani, B., 2018. Identification of the best algorithm for dust detection using MODIS data. Journal of Natural Environment Hazards 7 (15), 207- 220.
  17. Kumar, A., 2020. Spatio-temporal variations in satellite based aerosol optical depths & aerosol index over Indian subcontinent: Impact of urbanization and climate change. Urban Climate 32,100598
  18. Li, J., Wong, M.S., Lee, K.H., Nichol, J. and Chan, P.W., 2021. Review of dust storm detection algorithms for multispectral satellite sensors. Atmospheric Research 250,105398.
  19. Li, Q., Wang, Q., Wang, W. j., He, L. M. and Wang, C. Z., 2006. The application of the operational sand storm monitoring based on Terra/MODIS. Journal of Remote Sensing for Land & Resources 1, 412-430‌.
  20. Miller, S., 2003. A consolidated technique for enhancing desert dust storms with MODIS. Journal of Geophysical research letters 30(20), doi/abs/10.1029/2003GL018279.
  21. Mirakbari, M. and Ebrahimi Khusfi, Z., 2020. Investigation of spatial and temporal changes in atmospheric aerosol using aerosol optical depth in Southeastern Iran. Journal of RS and GIS for Natural Resources 11 (3), 1-18.
  22. Namdari, S., 2020. Temporal-spatial analysis of aerosols trend in the zone of influence Urmia aerosols by processing of satellite imageries in 2000-2015 (Case study: east Azerbaijan and west Azerbaijan). Journal of Geography and Planning 24 (72):427 – 446.
  23. Prospero, J. M., Ginoux, P., Torres, O., Nicholson, S. E. and Gill, T. E., 2002. Environmental characterization of global sources of atmospheric soil dust identified with the Nimbus 7 Total Ozone Mapping Spectrometer (TOMS) absorbing aerosol product. Journal of Reviews of geophysics 40(1), 221-231.
  24. Rashki, A., Middleton, N.J. and Goudie, A.S., 2021. Dust storms in Iran – distribution, causes, frequencies and impacts. Aeolian Research 48,100655.
  25. Saeedifar, Z., Khosroshahi, M., Gohardust, A., Ebrahimi khusfi, Z., Lotfi nasab asl, S. and Dargahian, F., 2020. Investigation of the origin and spatial distribution of high dust concentrations and its synoptical analysis in Gavkhooni basin. Journal of RS and GIS for Natural Resources 11 (4), 47-64.
  26. Shirazi, M., Akhavan Ghalibaf, M., Matinfar, H. R. and Nakhkesh, M., 2019. Comparison of MODIS and OLI image downscaling methods for industrial dust detection. Iranian Journal of Range and Desert Research 26 (3), 570 – 586.
  27. Varga, G., 2012. Spatio-temporal distribution of dust storms – a global coverage using NASA Toms aerosol measurements. Hungarian geographical bulletin 61(4), 275-298‌.
  28. You, Y., Zhao, T., Xie, Y., Zheng, Y., Zhu, J., Xia, J., Cao, L., Wang, C., Che, H., Liao, Y., Duan, J., Zhou, J. and Zhou, X., 2020. Variation of the aerosol optical properties and validation of MODIS AOD products over the eastern edge of the Tibetan Plateau based on ground-based remote sensing in 2017. Atmospheric Environment 223, 117257.
  29. Yue, H., He, C., Zhao, Y., Ma, Q. and Zhang, Q., 2017. The brightness temperature adjusted dust index: An improved approach to detect dust storms using MODIS imagery. International Journal of Applied Earth Observation and Geoinformation 57,166-176.
  30. Wang, Y., Yuan, Q., Li, T., Shen, H., Zheng, L. and Zhang, L., 2019. Large-scale MODIS AOD products recovery: Spatial-temporal hybrid fusion considering aerosol variation mitigation. ISPRS Journal of Photogrammetry and Remote Sensing 157, 1-12.