Assessing GPR and SVM Performance with Uncertainty Analysis for Drought Prediction in Iran's Diverse Climate Regions

Yahyazadeh, Mehrnaz; Bazrafshan, Ommolbanin; Moradi, Navazollah; Zamani, Hossein; Biniaz, Mehdi

doi:10.22077/jdcr.2025.9952.1164

Assessing GPR and SVM Performance with Uncertainty Analysis for Drought Prediction in Iran's Diverse Climate Regions

Document Type : Original Article

Authors

Mehrnaz Yahyazadeh ¹

Ommolbanin Bazrafshan ¹

Navazollah Moradi ²

Hossein zamani ³

Mehdi Biniaz ¹

¹ Department of Natural Resources Engineering, Faculty of Agricultural and Natural Resources Engineering, University of Hormozgan, Bandarabbas, Iran.

² Department of Natural Resources Engineering, Faculty of Agriculture and Natural Resources, University of Hormozgan, BandarAbbas, Iran.

³ Department of Mathematics and Statistics, Faculty of Science, University of Hormozgan, Bandarabbas, Iran.

https://doi.org/10.22077/jdcr.2025.9952.1164

Abstract

This study presents a comprehensive evaluation of Support Vector Machine (SVM) and Gaussian Process Regression (GPR) models for drought prediction across Iran's diverse climate zones using the Standardized Precipitation-Evapotranspiration Index (SPEI). The research integrates teleconnection indices, satellite data, and machine learning to address limitations of traditional drought forecasting methods. Results demonstrate the superior performance of GPR with Laplace kernel, achieving higher accuracy (R²: 0.91-0.75 in training, 0.85-0.37 in testing) and better uncertainty quantification (UA: 1.12-2.33, PICP: 1.0) compared to SVM-RBF. This practical improvement translates to a 10-15% increase in the explained variance of drought intensity, a critical distinction for activating different levels of emergency response. The Laplace kernel's flexibility in modeling abrupt climatic variations and GPR's probabilistic framework provide more reliable drought forecasts, particularly in extreme climates. Random Forest analysis revealed distinct climatic drivers, with temperature and evapotranspiration dominating arid regions, while oceanic oscillations (ENSO, WHWP) controlled humid zones. The UNEEC method provided robust uncertainty assessment, showing GPR's consistent performance across different climate classifications. While SVM-RBF remained competitive in moderate climates, its accuracy declined in complex conditions. The findings highlight GPR's advantages for precision drought forecasting in operational early warning systems, where reliable probabilistic forecasts can optimize reservoir management and agricultural advisory services, while acknowledging SVM's computational efficiency for large-scale monitoring applications.

Keywords

Drought Prediction

Machine Learning

Uncertainty Quantification

Teleconnection Patterns

UNEEC Method

Subjects

Drought

Azhdari Moghadam, M., Khosravi, M., Hosseinpour Niknam. H., Jaffari Nadoushan, E. (2011). Prediction of drought using FuzzyNeural network model, climatic indices, precipitationand drought index. Kournal of Geography and Development, 10(26): 61-72. [In Persian]. https://doi.org/10.22111/gdij.2012.443.

Abbasi, A., Khalili, K., Behmanesh, J., & Shirzad, A. (2020). Application of support vector machine and bayesian network for agricultural drought prediction. Watershed Engineering and Management, 12(1), 107-124.https://doi.org/10.22092/ijwmse.2019.122619.1519.

Alizadeh, A., Erfanian, M., & Ansari, H. (2011). Assessment of teleconnection patterns affecting rainfall and temperature (case study: Mashhad synoptic station). [In Persian]. http://dx.doi.org/10.29252/jwmr.9.17.280.

Bazrafshan, J. (2017). Effect of air temperature on historical trend of long-term droughts in different climates of Iran. Water Resources Management, 31(14), 4683-4698.https://doi.org/10.1007/s11269-017-1773-8.

Breiman, L. (2001). Random forests. Machine learning, 45(1), 5-32. https://doi.org/10.1023/A:1010933404324.

Bazrafshan, O. (2013). Hydrologic drought forecasting using teleconnection and intelligent models (Case study: Karkheh watershed). Ph.D. Thesis in Watershed Science and Engineering, University of Tehran, Iran. [In Persian].

Bazrafshan, J. (2017). Effect of air temperature on historical trend of long-term droughts in different climates of Iran. Water Resources Management, 31(14), 4683-4698. http://dx.doi.org/10.1007/s11269-017-1773-8.

Bazrafshan, O., Zamani, H., & Shekari, M. (2020). A copula‐based index for drought analysis in arid and semi‐arid regions of Iran. Natural Resource Modeling, 33(1), e12237. https://doi.org/10.1111/nrm.12237.

Bishop, C. M., & Nasrabadi, N. M. (2006). Pattern recognition and machine learning (Vol. 4, No. 4, p. 738). New York: springer.

Chalimourda, A., Schölkopf, B., & Smola, A. J. (2004). Experimentally optimal ν in support vector regression for different noise models and parameter settings. Neural Networks, 17(1), 127-141. https://doi.org/10.1016/S0893-6080(03)00209-0.

Choubin, B., Malekian, A., Samadi, S., Khalighi‐Sigaroodi, S., & Sajedi‐Hosseini, F. (2017). An ensemble forecast of semi‐arid rainfall using large‐scale climate predictors. Meteorological Applications, 24(3), 376-386. https://doi.org/10.1002/met.1635.

Cao, J., Zhang, Z., Tao, F., Chen, Y., Luo, X., & Xie, J. (2023). Forecasting global crop yields based on El Nino Southern Oscillation early signals. Agricultural Systems, 205, 103564. https://doi.org/10.1016/j.agsy.2022.103564.

Dogulu, N., López López, P., Solomatine, D. P., Weerts, A. H., & Shrestha, D. L. (2015). Estimation of predictive hydrologic uncertainty using the quantile regression and UNEEC methods and their comparison on contrasting catchments. Hydrology and Earth System Sciences, 19(7), 3181-3201. https://doi.org/10.5194/hess-19-3181-2015, 2015.

Elbeltagi, A., Kumari, N., Dharpure, J. K., Mokhtar, A., Alsafadi, K., Kumar, M., ... & Kuriqi, A. (2021). Prediction of combined terrestrial evapotranspiration index (CTEI) over large river basin based on machine learning approaches. Water, 13(4), 547. https://doi.org/10.3390/w13040547.

Emadodin, I., Reinsch, T., & Taube, F. (2019). Drought and desertification in Iran. Hydrology, 6(3), 66. https://doi.org/10.3390/hydrology6030066.

Fallah-Ghalhari, G., Shakeri, F., & Dadashi-Roudbari, A. (2019). Impacts of climate changes on the maximum and minimum temperature in Iran. Theoretical and Applied Climatology, 138(3), 1539-1562. http://doi.org/10.1007/s00704-019-02906-9.

Farzanegan, M. R., Feizi, M., & Gholipour, H. F. (2021). Drought and property prices: Empirical evidence from provinces of Iran. Economics of Disasters and Climate Change, 5(2), 203-221. https://doi.org/10.1007/s41885-020-00081-0.

Ghasemi, P., Karbasi, M., Nouri, A. Z., Tabrizi, M. S., & Azamathulla, H. M. (2021). Application of Gaussian process regression to forecast multi-step ahead SPEI drought index. Alexandria Engineering Journal, 60(6), 5375-5392. https://doi.org/10.1016/j.aej.2021.04.022.

Ghazipour, F., & Mahjouri, N. (2022). A multi-model data fusion methodology for seasonal drought forecasting under uncertainty: Application of Bayesian maximum entropy. Journal of Environmental Management, 304, 114245. https://doi.org/10.1016/j.jenvman.2021.114245.

Haas, M., Mittnik, S., & Paolella, M. S. (2006). Modelling and predicting market risk with Laplace–Gaussian mixture distributions. Applied Financial Economics, 16(15), 1145-1162. https://doi.org/10.1080/09603100500438817.

Hager, J., Hu, G., Hewage, K., & Sadiq, R. (2019). Performance of low-impact development best management practices: a critical review. Environmental Reviews, 27(1), 17-42. https://doi.org/10.1139/er-2018-0048

Jafarpour, M., Adib, A., Lotfirad, M., & Kisi, Ö. (2023). Spatial evaluation of climate change-induced drought characteristics in different climates based on De Martonne Aridity Index in Iran. Applied Water Science, 13(6), 133.https://doi.org/10.1007/s13201-023-01939-w.

Javanshiri, Z., Babaeian, I., & Pakdaman, M. (2023). Influence of large-scale climate signals on the precipitation variability over Iran. Stochastic Environmental Research and Risk Assessment, 37(5), 1745-1762. https://doi.org/10.21203/rs.3.rs-1982295/v1.

Karimi, G., Gandomkar, A., & Abbasi, A. (2024). Correlation of Teleconnection Patterns with Temperature Series and Productivity of Barberry in Ghaenat Basin. Journal of Climate Research, 1403(58), 123-140. https://doi.org/10.22034/jcr.2024.200832.

Khalili, A., Bazrafshan, J., & Cheraghalizadeh, M. (2022). A Comparative study on climate maps of Iran in extended de Martonne classification and application of the method for world climate zoning. Journal of Agricultural Meteorology, 10(1), 3-16. httl:// 10.22125/agmj.2022.156309.

Kushnir, Y., Seager, R., Ting, M., Naik, N., & Nakamura, J. (2010). Mechanisms of tropical Atlantic SST influence on North American precipitation variability. Journal of Climate, 23(21), 5610-5628. https://doi.org/10.1175/2010JCLI3172.1.

Malik, A., Tikhamarine, Y., Sammen, S. S., Abba, S. I., & Shahid, S. (2021). Prediction of meteorological drought by using hybrid support vector regression optimized with HHO versus PSO algorithms. Environmental Science and Pollution Research, 28(29), 39139-39158. http://doi.org/10.1007/s11356-021-13445-0.

Modarres, R., Sarhadi, A., & Burn, D. H. (2016). Changes of extreme drought and flood events in Iran. Global and Planetary Change, 144, 67-81. https://doi.org/10.1016/j.gloplacha.2016.07.008.

Nouri, M., & Homaee, M. (2020). Drought trend, frequency and extremity across a wide range of climates over Iran. Meteorological Applications, 27(2), e1899. https://doi.org/10.1002/met.1899.

Oruc, S., Hinis, M. A., & Tugrul, T. (2024). Evaluating performances of LSTM, SVM, GPR, and RF for drought prediction in Norway: a wavelet decomposition approach on regional forecasting. Water, 16(23), 3465. https://doi.org/10.3390/w16233465.

Pande, C. B., Sidek, L. M., Varade, A. M., Elkhrachy, I., Radwan, N., Tolche, A. D., & Elbeltagi, A. (2024). Forecasting of meteorological drought using ensemble and machine learning models. Environmental Sciences Europe, 36(1), 160. https://doi.org/10.1186/s12302-024-00975-w.

Piri, S.& Ansari ghojghar, M. (2025). Development of intelligent machine learning and Jenkins box models based on the full ensemble mode decomposition method for meteorological drought modeling (Case study: Khuzestan Province)., Journal of Climate Change Research. 6 (22),81-96. [In Persian]. https://doi.org/10.30488/ccr.2025.516420.1276.

Rahmati, O., Falah, F., Dayal, K. S., Deo, R. C., Mohammadi, F., Biggs, T., ... & Bui, D. T. (2020). Machine learning approaches for spatial modeling of agricultural droughts in the south-east region of Queensland Australia. Science of the total environment, 699, 134230. https://doi.org/10.1016/j.scitotenv.2019.134230.

Shi, J. Q., & Choi, T. (2011). Gaussian process regression analysis for functional data. CRC press. http://dx.doi.org/10.1201/b11038.

Smola, A., & Schölkopf, B. (2004). Experimentally optimal ν in support vector regression for different noise models and parameter settings. Neural Networks, 17(1), 127-141. https://doi.org/10.1016/S0893-6080(03)00209-0.

Vicente-Serrano, S. M., Beguería, S., & López-Moreno, J. I. (2010). A multiscalar drought index sensitive to global warming: the standardized precipitation evapotranspiration index. Journal of climate, 23(7), 1696-1718. https://doi.org/10.1175/2009JCLI2909.1.

Williams, C. K., & Rasmussen, C. E. (2006). Gaussian processes for machine learning, 2(3), p 4. Cambridge, MA: MIT press. https://doi.org/10.7551/mitpress/3206.001.0001.

Wallace, J. M., & Gutzler, D. S. (1981). Teleconnections in the geopotential height field during the Northern Hemisphere winter. Monthly weather review, 109(4), 784-812. https://doi.org/10.1175/1520-0493(1981)109%3C0784:TITGHF%3E2.0.CO;2.

Zamani, H., Pakdaman, Z., Shakari, M., Bazrafshan, O. and Jamshidi, S. (2025). Enhancing drought monitoring with a multivariate hydrometeorological index and machine learning-based prediction in the south of Iran. Environmental Science and Pollution Research, 32(9), pp.5605-5627. https://doi.org/10.1007/s11356-025-36049-4.