Artificial intelligence methods reliably predict crop evapotranspiration with different combinations of meteorological data for sugar beet in a semiarid area

Sevim Seda Yamaç

Agricultural Water Management, 2021, vol. 254, issue C

Abstract: Computation of crop evapotranspiration (ETc) is a necessary for irrigation planning and agricultural water management. However, ETc is complex and dynamic depending on many meteorological variables, so it is challenging to estimate accurately. Therefore, artificial intelligence could be a method to overcome these complexities. Consequently, the main objective of this study was to evaluate the performance metric of four artificial intelligence algorithms, including k-nearest neighbor (kNN), support vector machine (SVM), random forest (RF) and adaptive boosting (AB) using eight different input combinations for daily sugar beet ETc estimation. The input combinations included daily data for crop coefficient, maximum and minimum air temperature, solar radiation, wind speed, and maximum and minimum relative humidity. The values were obtained from field experiments conducted at Çumra, Konya in Central Anatolia, Turkey in 2012 and 2013 that focused on optimal water supply conditions for sugar beet production. The performance metric of four artificial intelligence methods were compared with a soil water balanced model, based on FAO methodology. All the artificial intelligence methods evaluated (kNN, SVM, RF and AB) provided satisfactory results with R2 in the range 0.799–0.992, mean squared error 0.048–1.273 mm d−1, root mean square error 0.219–1.128 mm d−1 and mean absolute error 0.156–0.832 mm d−1. The result show that the performance metrics of the artificial intelligence models get good performances even if the number of the input variables is applied less in the models. Among the input combinations, the SVM model with crop coefficient, maximum and minimum air temperature, solar radiation and wind speed (combination 5) shows better modelling accuracy than kNN, RF and AB models. Therefore, the SVM method is considered to be appropriate for estimation of daily sugar beet ETc in semiarid region.

Keywords: Adaptive boosting; k-Nearest neighbor; Penman-Monteith equation; Random forest; Support vector machine (search for similar items in EconPapers)
Date: 2021
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S037837742100233X
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:agiwat:v:254:y:2021:i:c:s037837742100233x

DOI: 10.1016/j.agwat.2021.106968

Access Statistics for this article

Agricultural Water Management is currently edited by B.E. Clothier, W. Dierickx, J. Oster and D. Wichelns

More articles in Agricultural Water Management from Elsevier
Bibliographic data for series maintained by Catherine Liu ().