Assessment of data intelligence algorithms in modeling daily reference evapotranspiration under input data limitation scenarios in semi-arid climatic condition

Abstract

Crop evapotranspiration is essential for planning and designing an efficient irrigation system. The present investigation assessed the capability of four machine learning algorithms, namely, XGBoost linear regression (XGBoost Linear), XGBoost Ensemble Tree, Polynomial Regression (Polynomial Regr), and Isotonic Regression (Isotonic Regr) in modeling daily reference evapotranspiration (ET0) at IARI, New Delhi. The models were developed considering full and limited dataset scenarios. The efficacy of the constructed models was assessed against the Penman–Monteith (PM56) model estimated daily ET0. Results revealed that the under full and limited dataset conditions, XGBoost Ensemble Tree gave the best results for daily ET0 modeling during the model training period. While, in the testing period under scenarios S1(Tmax) and S2 (Tmax, and Tmin), the Isotonic Regr models yielded superior results over other models. In addition, the XGBoost Ensemble Tree models outperformed others for the rest of the input data scenarios. The XGBoost Ensemble Tree algorithms reported the best values of correlation coefficient (r), mean absolute error (MAE), mean square error (MSE), root mean square error (RMSE), and mean absolute percentage error (MAPE). Thus, we recommend applying the XGBoost Ensemble Tree algorithm for precisely modeling daily ET0 in semi-arid climatic conditions.

Spatial and seasonal groundwater quality assessment for drinking suitability using index and machine learning approach

“…DWQI is a robust and widely used technique to recapitulate complex water quality parameters simply and understandably to classify the appropriateness of drinking water [ 33 ]. Support Vector Machine (SVM) is the emerging tool developed in the 1990s and effectively used among machine learning algorithms in several applications such as interpolations of DWQI values, prediction of water quality studies, classifications, and several other applications [ [37] , [38] , [39] ]. SVM performs best in geospatial interpolation especially when the data has no noise and outliers [ 40 ].…”

Predictive Modeling and Comparative Analysis of Reference Evapotranspiration with Machine Learning Algorithms

“…Various ML algorithms are employed worldwide for ET0 prediction [10], including adaptive neuro-fuzzy neural networks [11], least square-support vector machines (LS-SVM) [12], fuzzy logic [13], multiple-layer perceptron neural networks [14], relevance vector machines [15], multivariate regression splines [16], and Least Square-Support Vector Regression (LS-SVM) [17]. Multiple studies have demonstrated that ML-based models provide more accurate ET0 estimates compared to empirical methods like the Hargreaves-Samani method, Blaney-Criddle method, Thornthwaite method, Makkink method, and Penman method across various regions globally [18].…”