Comparative analysis of kernel-based versus ANN and deep learning methods in monthly reference evapotranspiration estimation

Abstract

Abstract. Timely and accurate estimation of reference evapotranspiration (ET0) is indispensable for agricultural water management for efficient water use. This study aims to estimate the amount of ET0 with machine learning approaches by using minimum meteorological parameters in the Corum region, which has an arid and semi-arid climate and is regarded as an important agricultural centre of Turkey. In this context, monthly averages of meteorological variables, i.e. maximum and minimum temperature; sunshine duration; wind speed; and average, maximum, and minimum relative humidity, are used as inputs. Two different kernel-based methods, i.e. Gaussian process regression (GPR) and support vector regression (SVR), together with a Broyden–Fletcher–Goldfarb–Shanno artificial neural network (BFGS-ANN) and long short-term memory (LSTM) models were used to estimate ET0 amounts in 10 different combinations. The results showed that all four methods predicted ET0 amounts with acceptable accuracy and error levels. The BFGS-ANN model showed higher success (R2=0.9781) than the others. In kernel-based GPR and SVR methods, the Pearson VII function-based universal kernel was the most successful (R2=0.9771). Scenario 5, with temperatures including average temperature, maximum and minimum temperature, and sunshine duration as inputs, gave the best results. The second best scenario had only the sunshine duration as the input to the BFGS-ANN, which estimated ET0 having a correlation coefficient of 0.971 (Scenario 8). Conclusively, this study shows the better efficacy of the BFGS in ANNs for enhanced performance of the ANN model in ET0 estimation for drought-prone arid and semi-arid regions.

Evapotranspiration Trends and Interactions in Light of the Anthropogenic Footprint and the Climate Crisis: A Review

“…Zhao et al (2019) [46] developed a method for post-processing seasonal GCM outputs to predict monthly and seasonal RET. Several models on heuristic and fuzzy-logic science for estimations of PE and RET and machine learning algorithms such as combined neural networks, genetic algorithm model, linear genetic programming, fuzzy genetic, adaptive neuro-fuzzy inference system, artificial neural networks, multilayer perceptron neural network, co-active neuro-fuzzy inference system, radial basis neural network and self-organizing map neural network showed high accuracy in different climate zones [15,[47][48][49][50][51].…”

Data intelligence and hybrid metaheuristic algorithms-based estimation of reference evapotranspiration

“…The study concluded that the proposed novel data intelligent model in combination with SVM has outperformed and found best in comparison with the empirical model. Sattari et al (2021) applied five data intelligent machine learning models for estimating ET 0 using several input meteorological combinations and found that combining machine learning (hybrid) models increase predictive ability in results. Malik et al (2019) estimated ET 0 using five machine learning models using different meteorological input combinations and concluded that accuracy in results and efficiency can be increased by using hybrid machine learning models.…”

Artificial Neural Networks for the Prediction of the Reference Evapotranspiration of the Peloponnese Peninsula, Greece

“…They concluded that the model with the most inputs resulted in the most accurate predictions of ETo and that tanh was the best activation function in all cases. Sattari et al (2021) tried ten combinations of input variables (from eight to one inputs) [72]. The best performance yielded an R 2 = 0.978.…”