Climate Change Research

Climate Change Research

Evaluating the performance of metaheuristic algorithms for optimizing input data to model dust storms (A case study of Sistan and Baluchestan province)

Document Type : Original Article

Authors
Ramtin Tavoosi Rad 1
Sadaf Piri 2
Mohammad Ansari ghojghar 3
Hossein Azarnivand 4
1 Master Candidate, Department of Reclamation of Arid and Mountainous regions Engineering, Faculty of Natural Resources, University of Tehran, Karaj, Iran
2 Ph.D. candidate, Department of Reclamation of arid and mountainous regions Engineering, Faculty of Natural Resources, University of Tehran, Karaj, Iran
3 Assistant Professor, Department of Reclamation of Arid and Mountainous regions Engineering, Faculty of Natural Resources, University of Tehran, Karaj, Iran
4 Professor, Department of Reclamation of Arid & Mountain regions Engineering, Faculty of Natural Resources, University of Tehran, Karaj, Iran
10.30488/ccr.2025.552548.1307
Abstract
Dust storms are among the most severe climatic hazards affecting arid and semi-arid regions, particularly southeastern Iran. The Sistan and Baluchestan Province is especially vulnerable due to its proximity to desert areas, persistent drought conditions, reduced vegetation cover, and the influence of the seasonal 120-day winds. Given the complex and uncertain nature of dust storm prediction—especially in regions with limited or low-quality meteorological data—this study investigates the role of metaheuristic algorithms in improving predictive accuracy through input optimization. The main objective is to evaluate the performance of two metaheuristic algorithms, Invasive Weed Optimization (IWO) and Water Cycle Algorithm (WCA), for optimizing input variables in the hybrid Fuzzy Clustering Model Regression–Moving Average (FCMR–MA) model. The hybrid model was employed to predict the seasonal frequency of dust storm days (FDSD) across five synoptic stations in Sistan and Baluchestan over a 40-year period (1980–2020). Four predictive scenarios were examined. The baseline FCMR–MA model using only dust storm data; the same model incorporating drought indices (SPI and SPEI); and the optimized FCMR–MA models enhanced with WCA and IWO algorithms. Multiple forecasting horizons (one- to four-season lags) were applied to assess model sensitivity to past dust storm events. Model performance was evaluated using four statistical indicators: Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Nash–Sutcliffe Efficiency (NS), and the Correlation Coefficient (R). Results indicate that both metaheuristic algorithms substantially enhance the predictive accuracy of the FCMR–MA model, with the IWO algorithm consistently achieving the best performance across all stations and evaluation metrics—particularly in Zabol and Zahedan, which exhibit higher FDSD frequencies. These findings highlight the potential of metaheuristic optimization for improving dust storm forecasting in data-scarce arid environments.
Keywords
120
day winds Water Cycle Algorithm Invasive Weed Optimization FCMR
MA hybrid model

Subjects

1.    Alshammari, RK,. Alrwais, O,. & Aksoy, M. (2022). Machine learning Applications to dust storms: A Meta-Analysis. Taiwan Association for Aerosol Research, ISSN:2071-1409. https://doi.org/10.4209/aaqr.220183.
2.    Almofti, S,. Shaban, A,. Ali, R, & Foente, J.(2023). Overview of Metaheuristic Algorithms. Polaris Global Journal Of Scholarly Research and Trends, 2(2), 10-32. https://doi.org/10.58429/pgjsrt.v2n2a144
3.    Bezdek, J.C., Ehrlich, R. and Full, W. (1984) FCM: The Fuzzy C-Means Clustering Algorithm. Comput. Geosci.10, 191-203. https://doi.org/10.1016/0098-3004(84)90020-7
4.    Dehshiri, S,. & Firouzabadi, B.(2023). A multi-objective framework to select numerical options in air quality prediction models: A case study on dust storm modeling. Science Total Environment. doi: 10.1016/j.scitotenv.2022.160681
5.    Ghosh, P,. Sadhu, P,. Basak, P,. & Sanal, A.(2020). Ain Shams Engineering Journal, 11, 1139-1147. https://doi.org/10.1016/j.asej.2020.01.017
6.    Gholami, H,. & Mohammadifar, A.(2022). Novel deep learning hybrid models (CNN-GRU and DLDL-RF) for the susceptibility classification of dust sources in the Middle East: a global source. Scientific Reports, 11;12(1), 19342. doi: 10.1038/s41598-022-24036-5
7.    Guidolin. (2018). Autoregressive Moving Average (ARMA) Models and their Practical Applications. University lecture.
8.    Hanafi, M,. Bahreininejad, A,. & Uddin, N.(2021). Optimization of shell and tube heat exchanger using the water cycle algorithm. IOP Confrence: Materials Science and Engineering. doi:10.1088/1757-899X/1173/1/012005
9.    Hui,. Tang et al.(2023). Modelling the 2021 East Asia super dust storm using FLEXPART and FLEXDUST and its comparison with reanalyses and observations. Frontiers in Environmental Science. DOI 10.3389/fenvs.2022.1013875
10.              Ibrahim, A., Anayi, F., Packianather, M., & Alomari, O. A. (2022). New Hybrid Invasive Weed Optimization and Machine Learning Approach for Fault Detection. Energies, 15(4), 1488. https://doi.org/10.3390/en15041488
11.              Ivanovski, Z,. & Milenovski, A.(2018). Time Series Forecasting Using a Moving Average Model for Extrapolation of Number of Tourist. UTMS Journal of Economics 9 (2), 121–132. https://hdl.handle.net/10419/281836
12.              Javadnia, E,. & Zahedi, N. (2022). Investigation Impact of Massive Dust Storm on Aerosol Optical, Physical, Radiative Properties over Southwest Iran. Earth Observation and Geomatics Engineering, 6(1), 135-149. DOI:  10.22059/eoge.2022.347630.1123
13.              Javadian, M. Behrangi, A. & Sorooshian, A. (2019). Impact of drought on dust storms: Case study over southest Iran. Environmenral Research Letter, 14(12), 29-40. DOI: 10.1088/1748-9326/ab574e
14.              Kakkar, S,. Maity, T,. Ahuja, R,. Walde, P,. Saket, R.K,. Khan, B,. & Padmanaban, S.(2021). Design and Control of Grid-Connected PWM Rectifiers by Optimizing Fractional Order PI Controller Using Water Cycle Algorithm. IEEE POWER & ENERGY SOCIETY SECTION, 9, 125941-125954. DOI:10.1109/ACCESS.2021.3110431
15.              Ma, C., Gao, Z., Yang, J., Cheng, L., & Zhao, T. (2022). Calibration of Adjustment Coefficient of the Viscous Boundary in Particle Discrete Element Method Based on Water Cycle Algorithm. Water, 14(3), 439. https://doi.org/10.3390/w14030439
16.              McKee, T.B., N.J. Doesken and J. Kleist. (1993). The relationship of drought frequency and duration to time scale. In: Proceedings of the Eighth Conference on Applied Climatology, Anaheim, California,. Boston, American Meteorological Society, 179–184.
17.              Mehrabi, Sh,. Soltani, S,. & Jafari, R. (2015). Analyzing  the relationship between duststorms occurence and climatic parameters (case study:Khozestan province). Journal of Agricultural Sciences and Technologies and Natural Resources, Water and Soil Sciences, 19(71). DOI:10.18869/acadpub.jstnar.19.71.69
18.              Mobarhan, E,. & Khaleghi, A.(2024). Analyzing the Trend of Changes in the Dust Storm Index (DSI) and Its Relationship with the Meteorological Drought in the Arid Climate (Case Study: Semnan Province). Environmental Sciences, 22(2), 289-304. DOI: 10.48308/envs.2024.1369
19.              Motlagh, F,. Bardsiri, V,. & Bardsiri, A.(2023). Human-Whale cooperation optimization (HWO) algorithm: A metaheuristic algorithm for solve optimization problems. Int. J. Nonlinear Anal. Appl, 14(1), 2279-2300. http://dx.doi.org/10.22075/ijnaa.2022.27197.3529
20.              Omid Rahmati, M. Panahi, Seid Saeid Ghiasi, R. Deo et al. Hybridized neural fuzzy ensembles for dust source modeling and prediction.(2020). Atmospheric Environment. DOI:10.1016/j.atmosenv.2020.117320
21.              Rahmati, O et al.(2020). Hybridized neural fuzzy ensembles for dust source modeling and prediction. Atmospheric Environment, 224(1), https://doi.org/10.1016/ j.atmosenv.2020.117320
22.              Razavi-Termeh, S., Sadeghi-Niaraki, A., Naqvi, R.A. & Choi, Soo Mi.(2023). Dust detection and susceptibility mapping by aiding satellite imagery time series and integration of ensemble machine learning with evolutionary algorithms. Environmental Pollution, 15:335:122241.doi: 10.1016/j.envpol.2023.122241.
23.              Salehpour Jama, A., Karimpour Reihanb, M., Mohseni Saravi, M., Bazrafshand, J.& Khalighi Sigaroudi, SH. (2017). Investigation of climate change effect on drought characteristics in the future period using the HadCM3 model (Case study: Khoy station, northwest of Iran). Desert Journal, 22(1), 43-50. DOI: 10.22059/jdesert.2017.62294
24.              Zeng, W.(2024). Application of AR, MA, and ARMA Models in Financial Time Series Analysis. ICFTBA Workshop. DOI: 10.54254/2754-1169/141/2024.GA18862