Evaluation of the Efficiency of Multi-Layer Perceptron Neural Network in Predicting Dust Storms at some Stations on the Persian Gulf Coast (Abadan, Ahvaz and Bushehr)

جغرافیا و مخاطرات محیطی

Volume 14, Issue 2 - Serial Number 54, June 2025, Pages 21-44   

Document Type: Research Article

DOI: 10.22067/geoeh.2025.91013.1531

Authors

Fatemeh Vatanparast Galeh Juq¹; Bromand Salahi^{* 2}

¹PhD Student of Climatology, Department of Physical Geography, Faculty of Social Sciences, University of Mohaghegh Ardabili, Ardabil, Iran

²Professor of Climatology, Department of Physical Geography, Faculty of Social Sciences, University of Mohaghegh Ardabili, Ardabil, Iran.

Abstract

Climate change, along with the subsequent occurrence of extreme events such as dust storms, not only disrupts the environment but also directly impacts human health and the natural course of life. In recent years, these events have had undesirable consequences in the agricultural sector. In this study, hourly dust data, average temperature, relative humidity, wind speed, and wind direction were collected from several stations along the Persian Gulf coast (Abadan, Ahvaz, and Bushehr) over a 37 year statistical period (1987–2023) to forecast and model dust storms. The number of days with dust storms was predicted on a seasonal scale using a multilayer perceptron neural network model and a seasonal SARIMA model. Based on the results of the multilayer perceptron neural network, the correlation coefficient between the observed and predicted values for the three synoptic stations Abadan, Ahvaz, and Bushehr was 0.43, 0.50, and 0.90, respectively, and the RMSE values were 6.96, 1.97, and 0.16, respectively. This model demonstrated lower error, higher correlation, and better predictive performance compared to the SARIMA model. Neural network forecasting for the next 16 years (2024–2040) indicated the highest probability of dust occurrence in spring and the lowest in autumn. Dust intensity was also found to be higher in Abadan than at the other stations during both the observational and future periods. Dust particle tracking using the HYSPLIT model and the backscatter method at three altitudes (200, 1000, and 1500 m) confirmed the generation of dust particles in Iraq, parts of Syria, and Saudi Arabia, as well as their movement toward western and southwestern Iran on common dates at the three stations. These results showed good agreement with dust optical depth, surface dust density, and the movement of the concentration mass as modeled by NAAPS and COAMPS. The findings of this study can contribute to the effective management of dust storm consequences and support programs aimed at combating desertification in the study areas.
Extended Abstract
Introduction
Healthy, pollution-free air is one of the most basic human needs, and among various pollutants, dust is considered one of the most significant air contaminants. Forecasting the risks associated with dust phenomena in affected areas and understanding their temporal and spatial variations are essential for risk preparedness and damage prevention. Neural networks are widely used forecasting tools and are increasingly applied in modeling dust storms.
Material and Methods
The data used in this study for forecasting dust storms include horizontal visibility measurements of less than 1 km and present weather codes indicating dust, along with seasonal time series data on average temperature, relative humidity, and wind speed and direction for the period 1987–2023 from three synoptic stations: Abadan, Ahvaz, and Bushehr.
Many methods have been developed for time series modeling and forecasting, making the selection of an appropriate method critical. In this study, a multilayer perceptron (MLP) neural network model was used to forecast the frequency of dust storms for the future period (2024–2040), and its results were compared with those of the seasonal ARIMA (SARIMA) model. To evaluate the forecasting performance of these models, Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and the correlation coefficient (R) were used as evaluation criteria.
In the next step, to provide accurate information about the timing, location, intensity, and path of dust storms, the most significant factors influencing dust concentration and transport routes were analyzed. Because the studied stations experienced joint dust events on July 29, 2009; June 8, 2010; June 3, 2011; and March 11, 2012, the Lagrangian HYSPLIT model was used to track the trajectory of dust particles at three altitudes—200, 1000, and 1500 meters—at 6-hour intervals prior to dust entry, using the backward trajectory method. The output was then compared with results from the NAAPS and COAMPS models.
Results and Discussion
The correlation analysis between dust frequency and meteorological variables such as average relative humidity, temperature, wind speed, and direction revealed a direct and statistically significant relationship at all stations. This confirmed the feasibility of modeling and forecasting future values based on these variables.
Based on evaluation metrics, the MLP model exhibited lower error rates and higher correlation than the SARIMA model across all stations and was thus identified as the optimal forecasting model. Comparison of actual and predicted dusty days during the testing phase showed strong alignment, confirming the model’s accuracy in estimating target values.
The analysis of seasonal dust variations at the studied stations during both the observation period (1987–2023) and the future period (2024–2040) revealed that all three stations experienced a higher frequency of dust storms in spring and a lower frequency in autumn—a trend consistent across both historical and projected data.
Routing analysis of dust events using the HYSPLIT model on shared dates indicated that the primary source of dust particles was a region spanning Iraq and parts of Turkey and Syria, with dust being transported to the study area. These findings were in good agreement with results from the NAAPS and COAMPS models, which tracked dust optical depth, surface dust density, and the movement of the concentration mass.
Conclusion
Dust is a highly variable phenomenon in terms of time, with its location and trajectory changing from hour to hour. The results of this study indicate that the longer the statistical period used for modeling dust storms, the more accurate the model predictions become.
From a practical standpoint, decision-makers and authorities can utilize these accurate models and forecasts to make timely, informed decisions before hazardous dust events occur. Implementing preventive measures at high-risk stations can help minimize or even avoid damage when such events take place.
 

Keywords

Dust Storm; Forecasting; Multilayer Perceptron (MLP); SARIMA; HYSPLIT Trajectory Model; Persian Gulf Climate

References

Abdolshahnejad, M., Khosravi, H., Nazari Samani, A., Zehtabian, G., & Alambaigi, A. (2020). Determining the Conceptual Framework of Dust Risk Based on Evaluating Resilience (Case Study: Southwest of Iran). Strategic Research Journal of Agricultural Sciences and Natural Resources, 5(1), 33-44. [In Persian] https://doi.org/10.22047/srjasnr.2020.113050

Aboagye-Sarfo, P., Mai, Q., Sanfilippo, F. M., Preen, D. B., Stewart, L. M., & Fatovich, D. M. (2015). A comparison of multivariate and univariate time series approaches to modelling and forecasting emergency department demand in Western Australia. Journal of Biomedical Informatics, 57, 62-73. https://doi.org/10.1016/j.jbi.2015.06.022

Ahmadpour, A., Mirhashemi, S. H., & Panahi, M. (2021). Evaluation of neural network algorithms, and time-series models and SARIMA-SETAR hybrid model in Monthly wind speed prediction. Journal of Arid Biome, 10(2), 131-146. [In Persian] https://doi.org/10.29252/aridbiom.2021.15523.1828

Ahmed, N. K., Atiya, A. F., Gayar, N. E., & El-Shishiny, H. (2010). An empirical comparison of machine learning models for time series forecasting. Econometric Reviews, 29(5-6), 594-621.

Akinci, T. C. (2011). Short Term Wind Speed Forecasting with ANN in Batman, Turkey. Elektronika Ir Elektrotechnika, 107(1), 41-45. https://eejournal.ktu.lt/index.php/elt/article/view/9078

Ansari Ghojghar, M., & Pourmohammad, P. (2025). Implementation of the Tri-Hybrid RBF-GA-SARIMA Meta-Model for Dust Storm Modeling (Case Study: Sistan and Baluchestan Province). Environmental Management Hazards, 11(4), 305-322. [In Persian] https://doi.org/10.22059/jhsci.2025.387276.859

Ansari Ghojghar, M., Araghinejad, S., Bazrafshan, J., Zahraie, B., & Parsi, E. (2021). Evaluating the Performance of GRU-LSTM Hybrid Model in Predicting the Dust Storms Events (Case Study: Khuzestan Province in Southwest of Iran). Iran-Water Resources Research, 17(1), 16-32. [In Persian] https://dor.isc.ac/dor/20.1001.1.17352347.1400.17.1.2.3

Ansari Ghojghar, M., Bazrafshan, J., & Araghinejad, S. (2022). Evaluating the Efficiency of Hybrid Metamodels of Machine Learning and Box Jenkins in Order to Model Dust Storms (Case Study: Khuzestan Province). Iranian Journal of Soil and Water Research, 53(8), 1695-1714. [In Persian] https://doi.org/10.22059/ijswr.2022.346694.669335

Ansari Ghojghar, M., Pourgholam-Amiji, M., Bazrafshan, J., Liaghat, A., & Araghinejad, S. (2020). Performance Comparison of Statistical, Fuzzy and Perceptron Neural Network Models in Forecasting Dust Storms in Critical Regions in Iran. Iranian Journal of Soil and Water Research, 51(8), 2051-2063. [In Persian] https://doi.org/10.22059/ijswr.2020.302529.668607

Araghinejad, S. (2013). Data-driven modeling: using MATLAB® in water resources and environmental engineering. Springer Science & Business Media, 67. https://doi.org/10.1007/978-94-007-7506-0

Arjomandi, H., Kheiralipour, K., & Amarloei, A. (2024). Prediction of dust concentration on a laboratory scale using image processing and artificial intelligence technologies. Journal of Research in Mechanics of Agricultural Machinery, 13(2), 1-9. [In Persian] https://doi.org/10.22034/jrmam.2024.14177.642

Azizi, G., Shamsipour, A., Miri, M., & Safarrad, T. (2012). Statistic and Synoptic Analysis of Dust Phenomena in West of Iran. Journal of Environmental Studies, 38(3), 123-134. [In Persian] https://doi.org/10.22059/jes.2012.29154

Box, G. E. P., Jenkins, G. M., & Reinsel, G. C. (1994). Time series Analysis: Forecasting and Control. 3rdEd, prentice Hall, Englewood Cliffs Inc, New Jersey, 598.

Cadenas, E., & Rivera, W. (2010). Wind speed forecasting in three different regions of Mexico، using a hybrid ARIMA-ANN model. Renewable Energy. 35, 2732-2738. http://dx.doi.org/10.1016/j.renene.2010.04.022

 

 

Dargahian, F., Yasrebi, B., & Khosrowshahi, M. (2022). Study the Dependence Between Climatic Factors and Dust Rise from Inrenal Hotspots in Khouzestan. Extension and Development of Watershed Management, 9(35), 11-19. [In Persian] https://www.wmji.ir/article_254170.html

Draxler, R. R. D., & Rolph, G. D. (2003). HYSPLIT (Hybrid Single-Particle Lagrangian Integrated Trajectory) Model access via NOAA ARL READY Website. http://www.arl.noaa.gov/ready/ hysplit4.html

Draxler, R., & Hes G. D. (1998). An overview of the HYSPLIT_4 modeling system for trajectories, dispersion and Deposition. Australian Meteorological Magazine, 47, 295-308.

Draxler, R., Stunder, B., Rolph, G., Stein, A., & Taylor, A. (2009). Hybrid single-particle Lagrangian integrated trajectories 4 user's guide. NOAA Tech. aMemo, ERL-ARL.

Du, P., Huang, Z., Tang, S., Dong, Q., Bi, J., Yu, X., & Gu, Q. (2023). Long-term Variation of Dust Devils in East Asia during 1959-2021. Authorea Preprints, 128(9), e2022JD038013. https://doi.org/10.1029/2022JD038013

Ebrahimi-Khusfi, Z., Taghizadeh-Mehrjardi, R., & Nafarzadegan, A. R. (2021). Accuracy, uncertainty, and interpretability assessments of ANFIS models to predict dust concentration in semi-arid regions. Environmental Science and Pollution Research, 28, 6796–6810. https://doi.org/10.1007/s11356-020-10957-z

Farzanehpey, F., Ranjbar-Fordoe, A., Khosravi, H., & Mosavi, S. H. (2024). Evaluation of dust changes and its relationship with temperature (Case study: Khuzestan province). Integrated Watershed Management, 4(1), 16-29. [In Persian] https://doi.org/10.22034/iwm.2024.2014553.1112

Fattahi Masrour, P., & Rezazadeh, M. (2022). Spatio-Temporal Distribution of Various Types of Dust Events in the Middle East during the Period 1996-2015. Journal of the Earth and Space Physics, 47(4), 231-248. https://doi.org/10.22059/jesphys.2021.321010.1007306

Fotouhi Firoozabad, F., & Malekinejad, H. (2020). Analysis and Zonation of Maximum 24-Hour Rainfall of Iran Using Wakeby Distribution and Geostatistic Technique. Desert Management, 7(14), 75-92. [In Persian] https://doi.org/10.22034/jdmal.2020.38477

Haykin, S. (1994). Neural Networks: a Comprehensive Foundation. Prentice Hall PTR. https://dl.acm.org/doi/abs/10.5555/541500

Hosseini Shah Parian, N. L., Firouzi, M. A., & Hosseini Kahnouj, S. R. (2021). Application of artificial neural network and correlation model in predicting dust phenomenon in Ahvaz metropolis. Quarterly Journal of Human and Environment, 54, 24-15. [In Persian] https://he.srbiau.ac.ir/article_16835_8bcf6205de3052d5cc91b2916bed3a3e.pdf

Hosseini, A., Alijani, B., & Waghei, Y. (2020). Predicting the Annual Dusty Days in Khorasan Razavi Province Using Spatial-Temporal Analysis. Journal of Geography and Environmental Hazards, 8(4), 103-117. [In Persian] https://doi.org/10.22067/geo.v0i0.79116

Hyndman, R. J., & Khandakar, Y. (2008). Automatic time series forecasting: the forecast package for R. Journal of Statistical Software, 27(3), 1-22. https://doi.org/10.18637/jss.v027.i03

Jahanbakhshasl, S., Khorshiddoust, A. M., abbsighasrik, F., & abbasighasrik, Z. (2024). Precipitation, Time Series Models, Man-Kendall, Health Winters model, West Azerbaijan Province. Journal of Applied Researches in Geographical Sciences, 24(75), 98-115. [In Persian] http://dx.doi.org/10.61186/jgs.24.75.10

Karegar, M E., Bodagh Jamali, J., Ranjbar Saadat Abadi, A., Moeenoddini, M., & Goshtasb, H. (2017). Simulation and Numerical Analysis of severe dust storms Iran East. Journal of Spatial Analysis Environmental Hazards, 3(4), 101-119. [In Persian] http://jsaeh.khu.ac.ir/article-1-2658-fa.html

Keykhosravi, S. S., Nejadkoorki, F., & Amintoosi, M. (2019). Estimation of Artificial Neural Networks (MLP and RBF) Accuracy in Anticipation of the Dust of the Sabzevar Cement Factory. Journal of Research in Environmental Health, 5(1), 43-52. [In Persian] https://doi.org/10.22038/jreh.2019.38083.1277

Khosroshahi, M., Saeedifar, Z., Shahbazi, K., Zandifar, S., Lotfinasabasl, S., Gohardoust, A., … & Khodagholi, M. (2024). Investigating the trend of temporal and spatial changes of dusty days and determining the contribution of climatic elements on its spread in Iran. Iranian Journal of Range and Desert Research, 30(4), 521-541. [In Persian] https://doi.org/10.22092/ijrdr.2024.130966

Kim, J. (2019). Particulate Matter Mortality Rates and Their Modification by Spatial Synoptic Classification. International Journal of Environmental Research and Public Health, 16(11), 1904. https://doi.org/10.3390/ijerph16111904

Mahmoudi, L., & Ikegaya, N. (2023). Identifying the Distribution and Frequency of Dust Storms in Iran Based on Long-Term Observations from over 400 Weather Stations. Sustainability, 15(16), 12294. https://doi.org/10.3390/su151612294

Mahmoudi, L., Amiri Doumari, S., Safarianzengir, V., Maleki, R., Kianinejad, S., & Kianian, M. K. (2020). Monitoring and Prediction of Dust and Investigating its Environmental Impacts in the Western Half of Iran and the Eastern Borders of Turkey and Iraq, Using Remote Sensing and GIS. Journal of the Indian Society of Remote Sensing, 49(5), 713-724. https://doi.org/10.1007/s12524-020-01224-2

Mehrabi, S., Soltani, S., & Jafari, R. (2015). Analyzing the Relationship Between Dust Storm Occurrence and Climatic Parameters. Journal of Water and Soil Science, 19 (71), 69-81. [In Persian] http://dx.doi.org/10.18869/acadpub.jstnar.19.71.69

Mirzadeh, S. M., Nejadkoorki, F., Moosavi, V., & Mirhoseini, S. A. (2021). Comparison of the accuracy of the support vector regression model with two common methods of artificial neural network and adaptive neuro-fuzzy inference system in predicting the pollutant concentration of the PM10. Journal of Natural Environment, 74(1), 167-179. [In Persian] https://doi.org/10.22059/jne.2021.307289.2041

Mohammadpour Penchah, M., Memarian, M. H., & Mirrokni, S. M. (2015). Modeling and Analysis of Dust Storms of Yazd Province Using Numerical Models. Journal of Geography and Environmental Hazards, 3(4), 67-83. [In Persian] https://doi.org/10.22067/geo.v3i4.34323

Mourianizadeh, S., Khoorani, A., & Sharif, M. (2024). Modeling dust storm based on spectral dust indicators and artificial intelligence in Hormozgan province. Iranian Journal of Geophysics, 18(4), 39-57. [In Persian] https://doi.org/10.30499/ijg.2024.418114.1543

Movahedi, S., Heydari Naserabad, B., Hashemi Ana, S. K., & Ranjbar, B. (2012). Zoning of climatic regions of Khuzestan province. Quarterly Journal of Geographical Space Research, 12(40), 73-64. [In Persian]

Nabavi, S. S., Moradi, H., & Shrifikia, M. (2019). Evaluation of dust storm temporal distribution and the relation of the effective factors with the frequency of occurrence in Khuzestan Province from 2000 to 2015. Scientific- Research Quarterly of Geographical Data (SEPEHR), 28(111), 191-203. [In Persian] https://doi.org/10.22131/sepehr.2019.37518

Negahban, S., Ganjaeian, H., Ghisarian, S. S., & Ebrahimi, A. (2024). Identifying the centers of dust and analyzing the factors influencing its occurrence Based on Remote Sensing Data (Case Study: Southwest Iran). Journal of Geography and Environmental Hazards, 13(4), 386-405. [In Persian] https://doi.org/0.22067/geoeh.2024.89088.1504

Nejadkoorki, F., & Baroutian, S. (2012). Forecasting Extreme PM10 Concentrations Using Artificial Neural Networks. International Journal of Environmental Research, 6(1), 277-284. [In Persian] https://doi.org/10.22059/ijer.2011.493

Pourgholam Amiji, M., Ansari Ghojghar, M., Bazrafshan, J., Liaghat, A., & Araghinejad, S. (2020). Comparing the Performance of SARIMA and Holt-Winters Time Series Models With Artificial Intelligence Methods in Dust Storms Forecasting (Case Study: Sistan and Baluchestan Province). Physical Geography Research, 52(4), 567-587. [In Persian] https://doi.org/10.22059/jphgr.2021.303847.1007524

Pourgholam-Amiji, M., Ansari Ghojghar, M., & Ahmadaali, K. (2021). Prediction of Dust Storms in Khuzestan Province Using Artificial Neural Networks. Nivar, 45(114-115), 56-75. [In Persian] https://doi.org/10.30467/nivar.2021.303747.1200

Pourmaafi Esfahani, E., Almodaresi, A., Mousaei Sanjerehei, M., & Hghparast, H. (2023). Evaluation of dust emission using artificial neural network model of Kashan city. Environmental Sciences, 21(2), 69-80. [In Persian] https://doi.org/10.48308/envs.2023.1177

Rahi, G. R., Bahraini, F., KhosroShahi, M., & Biabani, L. (2022). The Effect of Drought on Dust Storm Frequency (Case study: Bushehr Province). Journal of Water and Soil Conservation, 29(1), 31-51. [In Persian] https://doi.org/10.22069/jwsc.2022.19677.3511

Safarian Zengir, V., Zenali, B., Jafari Hasi Kennedy, Y., & Jafarzadeh, L. (2018). Investigation of dust and evaluation of its prediction in Ardebil province using ANFIS model. Journal of Spatial Analysis Environmental Hazards, 5(2), 107-124. [In Persian] http://jsaeh.khu.ac.ir/article-1-2771-fa.html

Saligheh, M., & Kakhaki Mehneh, H. (2015). Investigating the Relationship among Climate Factors and Air Pollution Fluctuation (Case Study of the City of Mashhad). Journal of Geography and Environmental Hazards, 4(2), 77-94. [In Persian] https://doi.org/10.22067/geo.v4i2.31769

Shaiba, H., Alaashoub, N. S., & Alzahrani, A. A. (2018). Applying machine learning methods for predicting sand storms. Paper presented at the Proceedings of the 1st International Conference on Computer Applications & Information Security (ICCAIS). Riyadh, Saudi Arabia. http://dx.doi.org/10.1109/CAIS.2018.8441998

Sobhani, B., & Safarian Zengir, V. (2019). Analysis and prediction of Dust phenomenon in the southwest of Iran. Journal of Natural Environmental Hazards, 8(22), 179-198. [In Persian] https://doi.org/10.22111/jneh.2019.28148.1481

Taghavi, F., Owlad, E., Safarrad, T., & Irannejad, P. (2013). Identifying and monitoring dust storm in the western part of Iran using remote sensing techniques. Journal of the Earth and Space Physics, 39(3), 83-96. [In Persian] https://doi.org/10.22059/jesphys.2013.35600

Xiao, F., Wong, M. S., Lee, K. H., Campbell, J. R., & Shea, Y. K. (2015). Retrieval of dust storm aerosols using an integrated neural network model. Computers & Geosciences, 85, 104-114. http://dx.doi.org/10.1016/j.cageo.2015.02.016

Zhu, Y. M., Lu, X. X., & Zhou, Y. (2007). Suspended sediment flux modeling with artificial neural network: An example of the longchuanjiang river in the upper yangtze catchment. Geomorphology, 84(4), 111-125. https://doi.org/10.1016/j.geomorph.2006.07.010