Comparative assessment of reference crop evapotranspiration models and its sensitivity to meteorological variables in Peninsular Malaysia

Ng, Jing Lin; Huang, Yuk Feng; Yong, Stephen Luo Sheng; Tan, Jia W.

doi:10.1007/s00477-022-02209-y

Cited by 4 publications

(6 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For Loukkos, the ranking was: LightGBM6>XGBoost6>LightGBM3>XGBoost3 with average RMSE of 0.025 -0.027 mm.day −1 . These findings support previous studies by Fan et al [32] and Yong et al [28], where LightGBM consistently outperformed other standalone ML models with an RMSE of 0.08-0.58 mm.day −1 and 0.041-0.315 mm.day −1 , respectively. Further, there was a minor difference between…”

Section: Comparison Of Standalone and Hybrid ML Models Using Various ...supporting

confidence: 91%

“…Overal, the aforementioned studies [27,28,31,32,[35][36][37][38]53] reinforce the superiority of standalone and hybrid ML models in the RET estimation. Similary, our findings indicate that standalone ML models exhibited better performance and accuracy compared to the empirical models for estimating daily RET at Gharb and Loukkos stations, using Tmax, Tmin, RHmean, Rs, U2 as input variables.…”

Section: Comparison Of Standalone and Hybrid ML Models Using Various ...mentioning

confidence: 91%

“…Goyal et al [27] highlighted the promising findings of ML models in various climates and environments, emphasizing their ability to improve accuracy above standard empirical models. Besides, researchers have applied various ML models, such as artificial neural networks (ANN) [28,29], support vector regression (SVR) [30], M5 model tree [31], random forests (RF) [32], reduced error pruning tree (REPTree) [33], extreme gradient boosting (XGBoost) [34], light gradient boosting machine (LightGBM) [28] and decision trees (DT) [35] to estimate daily RET uing restricted meteorological data.…”

Section: Introductionmentioning

confidence: 99%

“…Their findings revealed that the GBDT and XGBoost models exhibited performance on par with the SVM and ELM models, while offering advantages in terms of simplicity, accuracy, stability, and reduced computational costs, making them recommended options for daily RET estimation. Additionally, Yong et al [28] evaluated performance of LightGBM, ANN and decision forest regression (DFR) in five Malaysian meteorological stations and reported that LightGBM and ANN have proven stable and accurate in determining daily RET. It is worth mentioning that the ML models' performance in daily RET estimation is influenced by various factors, including the selection of input climatic variables, model structure, basic parameters, and performance criteria.…”

Section: Introductionmentioning

confidence: 99%

“…Careful consideration of these factors and the correlation between individual input variables and RET is crucial for optimizing the ML models' accuracy and efficiency. Additionally, effective tuning of ML model parameters further enhances their performance and efficiency for correct estimation [27][28][29][30][31][32][33][34].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Comparative Assessment of Eight Empirical and Four Hybrid Machine Learning Models for Estimating Daily Reference Evapotranspiration in Sub-Humid and Semi-Arid Climates

Acharki,

Raza,

Alshehri

et al. 2023

Preprint

View full text Add to dashboard Cite

Improving reference evapotranspiration (RET) estimation accuracy contributes to effective water resource management, irrigation planning, and climate change assessments in agricultural systems. The widely recommended FAO-56 Penman-Monteith (PM-FAO56) model for RET estimation often faces limitations due to incomplete meteorological data availability. To address this, we evaluate the ability of eight empirical models, four machine learning (ML) models and their hybrid models to estimate daily RET in Gharb and Loukkos irrigated perimeters in Morocco. These ML and hybrid models include Random Forest (RF), M5 Pruned (M5P), eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), RF-M5P, RF-XGBoost, RF-LightGBM and XGBoost-LightGBM. Additionally, six input combinations (based on Tmax, Tmin, RHmean, Rs and U2) were designed, with PM-FAO56 model considered as a target to models. Four statistical indicators including Kling Gupta Efficiency index (KGE), Coefficient of determination (R2), Mean Squared Error (RMSE), and Root relative squared error (RRSE) were applied to assess the models’ performance, across both training and testing phases. The findings reveal that Valiantzas 2013 (VAL2013b) model outperformed the other empirical models for all station, exhibiting high KGE and R2 (0.95–0.97), low RMSE (0.32–0.35 mm.day-1) and RRSE (8.14–10.30%). Additionally, the Hargreaves and Samani 1985 (HargS1985) model performed well in Gharb's stations, while the Valiantzas 2013 (VAL2013a) model showed good results in Loukkos' stations. Besides, the ML model’s performance RET estimation was higher when Tmax, Tmin, RHmean, Rs and U2 were used as inputs (combination 6). Among the ML and hybrid models, the XGBoost-LightGBM and RF-LightGBM achieved the highest accuracy (on average RMSE 0.015–0.097 mm.day-1), closely followed by the LightGBM and XGBoost models. However, M5P model had the lowest estimation accuracy RMSE ranged from 0.022 to 0.108 mm.day-1 on average. In summary, our study highlights the potential of ML models for RET estimation in subhumid and semi-arid areas, providing vital insights for improving water resource management, helping climate change research and optimizing irrigation scheduling for optimal agricultural water usage in the region.

show abstract

Section: Comparison Of Standalone and Hybrid ML Models Using Various ...supporting

confidence: 91%

Section: Comparison Of Standalone and Hybrid ML Models Using Various ...mentioning

confidence: 91%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Comparative Assessment of Eight Empirical and Four Hybrid Machine Learning Models for Estimating Daily Reference Evapotranspiration in Sub-Humid and Semi-Arid Climates

Acharki,

Raza,

Alshehri

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

Comparative evaluation of various parameter estimation methods for extreme rainfall in Kelantan River Basin

Ng,

Huang,

Tan

et al. 2023

Theor Appl Climatol

View full text Add to dashboard Cite

Improved reference evapotranspiration methods for regional irrigation water demand estimation

Singh

Karthikeyan

2022

Agricultural Water Management

View full text Add to dashboard Cite

Comparative assessment of reference crop evapotranspiration models and its sensitivity to meteorological variables in Peninsular Malaysia

Cited by 4 publications

References 43 publications

Comparative Assessment of Eight Empirical and Four Hybrid Machine Learning Models for Estimating Daily Reference Evapotranspiration in Sub-Humid and Semi-Arid Climates

Comparative Assessment of Eight Empirical and Four Hybrid Machine Learning Models for Estimating Daily Reference Evapotranspiration in Sub-Humid and Semi-Arid Climates

Comparative evaluation of various parameter estimation methods for extreme rainfall in Kelantan River Basin

Improved reference evapotranspiration methods for regional irrigation water demand estimation

Contact Info

Product

Resources

About