Modeling the Functional Relationship of Cumulative Precipitation in Eastern Iranian Stations with Spatio-Temporal Variables using a Hybrid Machine Learning Approach

Faraji, مظفر; Mirzavand, Mohammad

doi:10.22059/ije.2026.410789.1906

Modeling the Functional Relationship of Cumulative Precipitation in Eastern Iranian Stations with Spatio-Temporal Variables using a Hybrid Machine Learning Approach

Document Type : Research Article

Authors

¹ PhD Student, Department of Climatology, University of Tabriz, Tabriz, Iran

² Department of Environmental Sciences and Technologies, School of Energy Engineering and Sustainable Resources, College of Interdisciplinary Science and Technology, University of Tehran, Tehran, Iran

10.22059/ije.2026.410789.1906

Abstract

Subject: Analyzing and modeling the spatio-temporal pattern of cumulative precipitation at synoptic stations in eastern Iran using hybrid machine learning approaches.
Objective: The objective of this study is to explain and model the functional relationship between cumulative precipitation at stations in eastern Iran and spatio-temporal variables, aiming to identify dominant weather-related structures and spatial heterogeneity of precipitation through a hybrid machine learning approach.
Research Method: In this study, 24-hour cumulative precipitation data (P₍₂₄₎) from 21 synoptic stations located in eastern Iran, including the provinces of Sistan and Baluchestan, South Khorasan, Razavi Khorasan, and North Khorasan, were used for the period 1990–2020. The functional relationship between cumulative precipitation as the dependent variable and monthly temporal variations as the independent variable was evaluated using an estimated linear regression function. To analyze the spatial structure, the stations were classified based on cumulative precipitation data using hierarchical clustering according to Ward’s distance criterion.
Findings: The results of the linear regression analysis indicated that the adjusted coefficient of determination (Adjusted R²) was estimated at 0.64. This index demonstrates that the temporal predictor variable (months) was able to explain approximately 64.8% of the variance in the response variable (precipitation amount). Hierarchical cluster analysis identified three distinct climatic zones, including an arid and desert cluster, a semi-arid and transitional cluster, and a humid cluster. Among these, the Quchan station, with the highest mean cumulative precipitation (26.80 mm), emerged as a prominent precipitation hotspot. Overall, these findings indicate the presence of significant spatial variability and highlight the prominent role of temporal fluctuations in shaping precipitation patterns in eastern Iran.
Conclusions: The results demonstrate that the hybrid machine learning approach effectively models the spatio-temporal structure of precipitation in eastern Iran and clearly reveals regional weather-related heterogeneities. These findings provide a scientific basis for water resources management and planning in the arid and semi-arid regions of eastern Iran.

Keywords

Main Subjects

Climate Change

References

Esmaeilian, M. (2005). Comprehensive SPSS guide. Naghoos Publications. (in Persian).
Esmaeilian, M., Vahdat, J., & Heydardoust, H. (2016). R software guide. University of Mohaghegh Ardabili Press. (in Persian).
Faraji, M., Rezaei Banafsheh, M., Sari Sarraf, B., & Khorshiddoust, A. (2023). Numerical simulation of climate change variations in Iran using artificial neural network algorithms. Climate Research Journal, 4(14), 43–64. (in Persian).
Sadatinejad, S. J., Soleimani Sardo, F., & Mirzavandi, M. (2024). Modeling and prediction of climatic parameters using CanESM2 model under RCP scenarios (Case study: Karaj Station). Ecohydrology Journal, 11(3), 411–426. (in Persian).
Sattari, M. T., Bagheri, R., Shirini, K., & Allahverdi Pour, P. (2024). Modeling daily and monthly precipitation of Tabriz using ensemble learning and decision tree regression models. Climate Change Research, 5(18), 31–48. https://doi.org/10.30488/ccr.2024.433394.1192
Alizadeh, O. (2022). Advances and challenges in climate modeling. Climatic Change, 170(1), 18. https://doi.org/10.1007/s10584-021-03298-4
Almazroui, M., Islam, M. N., Saeed, S., Saeed, F., & Ismail, M. (2020). Future changes in climate over the Arabian Peninsula based on CMIP6 multimodel simulations. Earth Systems and Environment, 4(4), 611–630. https://doi.org/10.1007/s41748-020-00183-5
Bi, E. G., Gachon, P., Vrac, M., & Monette, F. (2017). Which downscaled rainfall data for climate change impact studies in urban areas? Theoretical and Applied Climatology, 127(3–4), 685–699. https://doi.org/10.1007/s00704-015-1656-y
Breiman, L., & Friedman, J. H. (1985). Estimating optimal transformations for multiple regression and correlation. Journal of the American Statistical Association, 80, 580–619.
Chambers, J. M., & Hastie, T. J. (1991). Statistical Models in S. Chapman and Hall.
Das, P., Sachindra, D. A., & Chanda, K. (2022). Machine learning-based rainfall forecasting with multiple nonlinear feature selection algorithms. Water Resources Management, 36(15), 6043–6071. https://doi.org/10.1007/s11269-022-03341-8
Davenport, F. V., & Diffenbaugh, N. S. (2021). Using machine learning to analyze physical causes of climate change. Geophysical Research Letters, 48(15), e2021GL093787. https://doi.org/10.1029/2021GL093787
Farooq, R., Imteaz, M. A., & Mekanik, F. (2025). Deterministic rainfall forecasting using machine learning. Earth Science Informatics, 18(2), 224. https://doi.org/10.1007/s12145-025-01724-0
Feitoza Silva, D., Simonovic, S. P., Schardong, A., & Goldenfum, J. A. (2021). Introducing non-stationarity into IDF curves. Water, 13(8), 1008. https://doi.org/10.3390/w13081008
Hosseinzadeh-Talaei, P., Tabari, H., & Willems, P. (2020). Climate change impact on short-duration precipitation extremes. Journal of Hydrology, 590, 125249.
Hu, H., & Ayyub, B. M. (2019). Machine learning for projecting extreme precipitation. Geosciences, 9(5), 209. https://doi.org/10.3390/geosciences9050209
Jahangiri, M., Asghari, M., Niksokhan, M. H., & Nikoo, M. R. (2025). BiLSTM-Kalman framework for precipitation downscaling. Scientific Reports, 15(1), 24354. https://doi.org/10.1038/s41598-025-08264-z
Kaufman, L., & Rousseeuw, P. J. (1990). Finding Groups in Data: An Introduction to Cluster Analysis. Wiley.
Kourtis, I. M., & Tsihrintzis, V. A. (2022). Update of IDF curves under climate change. Water Supply, 22(5), 4951–4974. https://doi.org/10.2166/ws.2022.152
Lanciotti, S., Ridolfi, E., Russo, F., & Napolitano, F. (2022). IDF curves in a data-rich era. Water, 14(22), 3705. https://doi.org/10.3390/w14223705
Maity, S. S., & Maity, R. (2022). Changing pattern of IDF relationships. Water Resources Management, 36(14), 5371–5399. https://doi.org/10.1007/s11269-022-03313-y
Najafi, H., Lagerwall, G. L., Obeysekera, J., & Liu, J. (2025). Machine learning in climate downscaling: A critical review. Preprints. https://doi.org/10.20944/preprints202511.1538.v1
Neelin, J. D., et al. (2022). Precipitation extremes and water vapor. Current Climate Change Reports, 8(1), 17–33. https://doi.org/10.1007/s40641-021-00177-z
Rajagopal, S., & Steele, C. M. (2017). Hydrologic processes and prediction needs. Hydrology and Earth System Sciences, 21(1), 1–22. https://doi.org/10.5194/hess-21-1-2017
Sachindra, D. A., et al. (2018). Statistical downscaling using machine learning. Atmospheric Research, 212, 240–258. https://doi.org/10.1016/j.atmosres.2018.05.022
Struyf, A., Hubert, M., & Rousseeuw, P. J. (1997). Integrating robust clustering techniques in S-PLUS. Computational Statistics & Data Analysis, 26, 17–37.
Sun, T., et al. (2024). ML-based downscaling of GPM precipitation. Heliyon, 10(17), e36368. https://doi.org/10.1016/j.heliyon.2024.e36368
Yang, F., et al. (2024). Transformer-based precipitation downscaling. Remote Sensing, 16(22), 4292. https://doi.org/10.3390/rs16224292
Zhang, L., et al. (2023). Non-stationary runoff modeling using ML. Advances in Water Science, 34(1), 42–52. https://doi.org/10.14042/j.cnki.32.1309.2023.01.005