Evaluation of different machine learning methods for estimating daily reference evapotranspiration
Document Type : Research Paper
Authors
- 1 Department of Water Engineering, Faculty of Agricultural Engineering, Sari Agricultural Sciences and Natural Resources University, Sari, Iran
- 2 PhD Scholar, Department of Water Engineering, Faculty of Agricultural Engineering, Sari Agricultural Sciences and Natural Resources University, Sari, Iran
- 3 Department of Water Science and Engineering, Faculty of Agriculture, Shahrekord University, Shahrekord, Iran
Abstract
Accurate estimation of reference evapotranspiration (ET₀) is essential for effective water management in agriculture. However, ET₀ is a complex, nonlinear process influenced by many factors, and its prediction accuracy depends on the quality and combination of large, ever-growing meteorological datasets. Therefore, this study investigates the performance of the M5 tree model, the KStar algorithm, Support Vector Machines (SVM), and multiple linear regression in reducing the number of input parameters required for estimating daily reference evapotranspiration. The data used in this research include minimum and maximum temperature, average relative humidity, wind speed at two meters height, and sunshine hours, recorded at the Kuhrang station, Iran, over the period 2016–2020. The FAO Penman-Monteith model was used as the benchmark for evaluating the performance of the models. Based on data availability, various scenarios were developed for estimating ET₀ by excluding certain input variables. Evaluation metrics included Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and the correlation coefficient (R2). The results showed that the M5 tree model outperformed the other models in scenarios ( , , ) and ( , ), with correlation coefficients of R2 = 0.96 and R2 = 0.97, respectively. Further sensitivity analysis revealed that accurate estimation of evapotranspiration in this region requires temperature data, sunshine duration, and wind speed.
Keywords
Subjects
Irrigation
Alves, R. G., Maia, R. F., & Lima, F. (2023). Development of a Digital Twin for smart farming: Irrigation management system for water saving. Journal of Cleaner Production, 388, 135920, DOI: https://doi.org/10.1016/j.jclepro.2023.135920
Anifowose, F., Labadin, J., & Abdulraheem, A. (2015). Improving the prediction of petroleum reservoir characterization with a stacked generalization ensemble model of support vector machines. Applied Soft Computing, 26, 483–496. DOI: https://doi.org/10.1016/j.asoc.2014.10.017
Anonymous. (2020). Water Productivity Improvement Program for Agriculture in Chaharmahal and Bakhtiari Province. Chaharmahal and Bakhtiari Agricultural and Natural Resources Research and Education Center. 25 pages.
Ayaz, A., Rajesh, M., Singh, S. K., & Rehana, S. (2021). Estimation of reference evapotranspiration using machine learning models with limited data. AIMS Geosciences, 7, 268-290. DOI: 10.3934/geosci.2021016
Bakhtiari, B., Mohebbi-Dehaghani, A., & Qaderi, K. (2022). Comparative analysis of data-driven methods for daily reference evapotranspiration estimation of Southern Caspian Sea. Meteorological Application, 29(4), 16. DOI: https://doi.org/10.1002/met.2091
Bellido-Jim´enez, J. A., Est´evez, J., & García-Marín, A. P. (2022). A regional machine learning method to outperform temperature-based reference evapotranspiration estimations in Southern Spain. Agricultural Water Management, 274. DOI: https://doi.org/10.1016/j.agwat.2022.107955
Bellido-Jiménez, J. A., Estévez, J., & García-Marín, A. P. (2021). New machine learning approaches to improve reference evapotranspiration estimates using intra-daily temperature-based variables in a semi-arid region of Spain. Agricultural Water Management, 245, 106558. DOI: https://doi.org/10.1016/j.agwat.2020.106558
Droogers, P., & Allen, R. G. (2002). Estimating reference evapotranspiration under inaccurate data conditions. Irrigation and Drainage. Systems, 16, 33-45. DOI: 10.1023/A:1015508322413
Ekmekcioğlu, Ö., Başakın, E. E., & Özger, M. (2020). Tree-based nonlinear ensemble technique to predict energy dissipation in stepped spillways. European Journal of Environmental and Civil Engineering, 1-19. DOI: 10.1080/19648189.2020.1805024
Ganjei, S., & Nazemi, A. H. (2024). Comparing the performance of machine learning methods in modeling daily reference ET and its spatial distribution (case study: Zanjan province). Irrigation Science and Engineering, 48(1), 77-92 [In Persian].
Granata, F., Di Nunno, F., Gargano, R., & de Marinis, G. (2019). Equivalent discharge coefficient of side weirs in circular channel- a lazy machine learning approach, Water, 11(11), 2406 DOI: 10.3390/w11112406
Hosseini Vardanjani, S. M., Ganjikhorramdel, N., Khaltabadi Farahani, A., Khoshravesh, M., & Nouri Khajebelagh, R. (2024). A generalization of the M5 Tree model and artificial neural network model for estimating daily reference evapotranspiration under minimal climatic data conditions. Journal of Water and Climate Change, 15(12), 5882-5892. DOI: 10.2166/wcc.2024.500
Hosseini Vardanjani, S. M., Khoshravesh, M., Hosseini-Kakolaki, S. E., & Pourgholam-Amiji, M. (2022). Evaluation of hybrid model ANN-PSO and different data mining methods in estimating monthly evapotranspiration in two different climates. Iranian Journal of Irrigation and Drainage, 16(5), 924-936 [In Persian].
Kavoosi, M., Khasheisiuki, A., Pourrezabilondi, M., & Najafi, M. H. (2017). Evaluation of LSSVM model to estimation of evaporation and its comparison with empirical models. Environment and Water Engineering, 3(3), 235-246 [In Persian].
Karami Moghadam, M., Amini, A., & Keshavarzi, A. (2020). Intake design attributes and submerged vanes effects on sedimentation and shear stress. Water and Environment Journal. https://doi.org/10.1111/wej.12471
Leong, D. P., Teo, K. K., Rangarajan, S., Lopez-Jaramillo, P., Avezum Jr, A., & Orlandini, A. (2018). World Population Prospects 2019. Department of Economic and Social Affairs Population Dynamics. New York (NY).
Lum, M., Bateni, S. M., Shiri, J., & Keshavarzi, A. (2017). Estimation of reference evapotranspiration from climatic data. International Journal of Hydrology, 1(1), 25-30. DOI: https://doi.org/10.15406/ijh.2017.01.00005
Makwana, J. J., Tiwari, M. K., & Deora, B. S. (2023). Development and comparison of artificial intelligence models for estimating daily reference evapotranspiration from limited input variables. Smart Agricultural Technology, 3, 100115. DOI: https://doi.org/10.1016/j.atech.2022.100115
Malekahmadi, H., Mirzania, E., Khosravi, S., & Ebrahim Zadeh, A. (2022). Daily evapotranspiration modeling using regression and intelligent models (Case Study). Iranian Water Research Journal, 16(1), 49-62 [In Persian]. DOI: 10.22034/iwrj.2022.10026.2338
Mikaeili, F., Samadianfard, S., and Delirhasannia, R. (2023). Evaluating the strategy of ensemble empirical and tree-based methods in estimating reference evapotranspiration. Environment and Water Engineering, 9(2), 256-270 (In Persian). DOI: https://doi.org/10.22034/ewe.2022.345198.1789
Mirhashemi, S., Panahi, M., & Zareei, L. (2020). Evaluation of M5P algorithm for estimation of potential evapotranspiration, minimum and maximum temperature (Case study: Sari weather station). Journal of Meteorology and Atmospheric Sciences, 2(4), 287-295 [In Persian].
Piri, H. (2024). Investigating the effectiveness of data mining methods in predicting daily reference evapotranspiration (Case study: coastal strip stations in southern Iran). Ecohydrology, 11(2), 271-286. DOI: 10.22059/ije.2024.375755.1816
Sadeghian Agkandy, M., Rezaie, H., Khalili, K., & Ahmadie, F. (2024). Investigating the performance of Kstar and GPR algorithms in modeling RDI meteorological drought index (Case study: east of Urmia Lake basin). Journal of Civil and Environmental Engineering, 54.1(114), 142-151. DOI: 10.22034/ceej.2023.55292.2225. [In Persian]
Samadianfard, S., & Panahi, S. (2019). Estimating daily reference evapotranspiration using data mining methods of support vector regression and M5 model tree. Journal of Watershed Management Research, 9(18), 157-167. DOI: 10.29252/jwmr.9.18.157 [In Persian].
Sharafi, M., Abdi, E., Mohebbiyan, M., & Samadianfard, S. (2024). Prediction of daily evapotranspiration using the strategy of combining tree models with empirical Hargreaves method. Water and Soil Science, 34(2), 107-119. DOI: 10.22034/ws.2023.55840.2519
Shojaei, M. O., Mikaeili, F., & Samadianfard, S. (2022). Evaluating capabilities of gradient boosted tree and optimized random forest models in estimating daily dew point temperature. Environment and Water Engineering,8(3), 654-668. DOI: https://doi.org/10.22034/jewe.2022.313219.1662
Siasar, H., & honar, T. (2019). Application of support vector machine, CHAID and random forest models, in estimated daily Reference evapotranspiration in northern Sistan and Baluchestan province. Iranian Journal of Irrigation and Drainage, 13(2), 378-388 [In Persian].
Temesgen, B., Eching, S., Davidoff, B., & Frame, K. (2005). Comparison of some reference evapotranspiration equations for California. Journal of Irrigation and Drainage Engineering, 131(1), 73–84. DOI: https://doi.org/10.1061/(ASCE)0733-9437(2005)131:1(73)
Terzi, Ö. (2007). Data mining approach for estimation evaporation from free water surface. Journal of Applied Science, 7(4), 593-596. DOI: 10.3923/jas.2007.593.596
Valipour, M., Gholami Sefidkouhi, M. A., & Khoshravesh, M. (2017). Estimation and trend evaluation of reference evapotranspiration in a humid region. Italian Journal of Agrometeorology, 1, 19-36. DOI:10.19199/2017.1.2038-5625.019
Vapnik, V., & Lerner, A. (1963). Pattern recognition using generalized portrait method. Automation and Remote Control, 24, 774–780.
Vapnik, V. N. (2000). The Nature of statistical learning theory, the nature of statistical learning theory. Springer New York, New York, NY.
Zorati Pur, E., Neisi, L., Golabi, M., Bazaz, A., & Zoratipur, A. (2019). Simulation and comparison of potential evapotranspiration by artificial neural networks, ANFIS (Fuzzy Neural Network) and decision making M5 (Case study; synaptic station of Shiraz). Iran-Water Resources Research, 15(1), 365-371 [In Persian].
)