Evaluation of five data mining algorithms in predicting monthly potential evapotranspiration (case study: Shiraz)

Mirhashemi, SeyedHassan; Bour, Ali Asghar

doi:10.22034/aes.2021.144191

Evaluation of five data mining algorithms in predicting monthly potential evapotranspiration (case study: Shiraz)

Document Type : Original research article

Authors

SeyedHassan Mirhashemi ¹

Ali Asghar Bour ²

¹ M.Sc Graduate of Water Engineering, Faculty of agriculture, University of Zanjan, Zanjan, Iran

² Master of Hydraulic Structures, graduated from Islamic Azad University, Science and Research Branch, Tehran, Iran

10.22034/aes.2021.144191

Abstract

Data mining algorithms were used in this study to predict Shiraz's monthly potential evapotranspiration. The CART (Classification and Regression Trees), M5P, K-star, M5Rules, and REP-Tree (Reduced Error Pruning Tree) algorithms were used to predict potential evapotranspiration. Meteorological data from the Shiraz weather station from 2001 to 2016 were used in this study. The CART algorithm performed better in estimating monthly averages, according to statistical indicators. The maximum amount of potential evapotranspiration was reached when the sunshine hours exceeded 9.5 hours and the wind speed exceeded 0.3 meters per second, according to the results. When there was less than 9.5 hours of sunshine and the air temperature was less than 2 °C, the potential evapotranspiration rate was the lowest. The sensitivity analysis revealed that the parameters of sunshine hours, air temperature, wind speed, and relative humidity had a positive effect on the CART algorithm's performance in estimating monthly evapotranspiration.

Highlights

This study used data mining algorithms to forecast Shiraz's monthly evapotranspiration.

Statistically, the CART algorithm predicted monthly averages better.

The maximum potential evapotranspiration was reached when the sunshine hours exceeded 9.5 and the wind speed exceeded 0.3 m/s.

The sensitivity analysis revealed that sunshine hours, air temperature, wind speed, and relative humidity all improved the CART algorithm's monthly evapotranspiration estimation performance.

Keywords

Air temperature

CART algorithm

Sunshine hours

Wind speed

Weather station

Allen, R.G., Pereira, L.S., Raes, D., Smith M., 1998. Crop evapotranspiration, Irrigation and Drainage Paper No. 56. Food and Agriculture Organization: Rome, Italy.

Asadi Zarch, M.A., Sivakumar, B., Malekinezhad, H., Sharma, A., 2017. Future aridity under conditions of global climate change. Journal of Hydrology, 554, 451-469.

Crows Corporation, T., 1999. Introduction to data mining and knowledge discovery, third ed., Postmac, MD. Available at: www.twocrows.com, (April 29, 2000)

Fooladvand, H., 2010. Monthly prediction of reference crop evapotranspiration in Fars province. Iranian Journal of Soil and Water Science, 20 (4):157-169.

Giudici, P., 2003. Applied data mining: statistical methods for business and industry. Wiley, London . pp. 364.

Itenfisu, D., Elliott, R.L., Allen R.G., Walter I.A., 2003. Comparison of reference evapotranspiration calculations as part of the ASCE standardization effort. Journal of Irrigation and Drainage Engineering, ASCE 129, 440-448

Martí, P., González-Altozano, P., López-Urrea, R., Mancha, L.A., Shiri, J., 2015. Modeling reference evapotranspiration with calculated targets. Assessment and implications. Agricultural Water Management, 149, pp.81-90.

Mirhashemi, S.H., Panahi, M., 2014a. Evaluation of CART model to estimate potential evapotranspiration and relative humidity (Case Study: Arak station). The first national conference on agriculture and sustainable natural resources. 30 January, 2014. Mehr Arvand Higher Education Institute.

Mirhashemi, S.H., Panahi, M., 2014b. Evaluation of M5P model in predicting average temperature of air and monthly potential evapotranspiration for the next month in Rasht synoptic station. The First National Conference of Water Use Optimization. 5-6 March, 2014. Gorgan University of Agricultural Sciences and Natural Resources.

Pandey, P.K., Dabral, P.P., Pandey, V., 2016. Evaluation of reference evapotranspiration methods for the northeastern region of India. International Soil and Water Conservation Research, 4, 52-63.

Samadianfard, S., Panahi, S., 2018. Estimating daily reference evapotranspiration using data mining methods of support vector regression and m5 tree model. Journal of Watershed Management Research, 9(18), 157-167, (In Persian).

Sameti, M., Ghahraman, N., Ghorbani, Kh., 2013. Application of M5 model in estimating reference crop evapotranspiration in Shiraz and Kermanshah. Journal of Water Researches in Agriculture, 27(3), 289-299. (In Persian).