Which machine learning algorithm is best suited for estimating reference evapotranspiration in humid subtropical climate?

Timely and reliable estimates of reference evapotranspiration (ET0) are imperative for robust water resources planning and management. Applying machine learning (ML) algorithms for estimating ET0 has been evolving, and their applicability in different sectors is still a compelling field of research....

Descripción completa

Detalles Bibliográficos
Autores principales: Deb, P., Kumar, V., Urfels, A., Lautze, Jonathan F., Kamboj, B. R., Sharma, J. R., Yadav, S.
Formato: Journal Article
Lenguaje:Inglés
Publicado: Wiley 2025
Materias:
Acceso en línea:https://hdl.handle.net/10568/175610
_version_ 1855523919428059136
author Deb, P.
Kumar, V.
Urfels, A.
Lautze, Jonathan F.
Kamboj, B. R.
Sharma, J. R.
Yadav, S.
author_browse Deb, P.
Kamboj, B. R.
Kumar, V.
Lautze, Jonathan F.
Sharma, J. R.
Urfels, A.
Yadav, S.
author_facet Deb, P.
Kumar, V.
Urfels, A.
Lautze, Jonathan F.
Kamboj, B. R.
Sharma, J. R.
Yadav, S.
author_sort Deb, P.
collection Repository of Agricultural Research Outputs (CGSpace)
description Timely and reliable estimates of reference evapotranspiration (ET0) are imperative for robust water resources planning and management. Applying machine learning (ML) algorithms for estimating ET0 has been evolving, and their applicability in different sectors is still a compelling field of research. In this study, four Gaussian process regression (GPR) algorithms—polynomial kernel (PK), polynomial universal function kernel (PUK), normalized poly kernel (NPK), and radial basis function (RBF)—were compared against widely used random forest (RF) and a simpler locally weighted linear regression (LWLR) algorithm at a humid subtropical region in India. The sensitivity analysis of the input variables was followed by application of the best combination of variables in algorithm testing and training for generating ET0. The results were then compared against the Penman–Monteith method at both daily and monthly time steps. The results indicated that ET0 is least sensitive to wind speed at 2 m height. Additionally, at a daily time step, RF, followed by PUK, generated the best results during both training and testing phases. In contrast, at a monthly time step, using multiple model evaluation matrices, PUK followed by RF performed best. These results demonstrate the application of the ML algorithms is subjected to user-required time steps. Although this study focused on Northwest India, the findings are relevant to all humid subtropical regions across the world.
format Journal Article
id CGSpace175610
institution CGIAR Consortium
language Inglés
publishDate 2025
publishDateRange 2025
publishDateSort 2025
publisher Wiley
publisherStr Wiley
record_format dspace
spelling CGSpace1756102025-10-14T15:09:09Z Which machine learning algorithm is best suited for estimating reference evapotranspiration in humid subtropical climate? Deb, P. Kumar, V. Urfels, A. Lautze, Jonathan F. Kamboj, B. R. Sharma, J. R. Yadav, S. machine learning algorithms evapotranspiration subtropical climate groundwater models Timely and reliable estimates of reference evapotranspiration (ET0) are imperative for robust water resources planning and management. Applying machine learning (ML) algorithms for estimating ET0 has been evolving, and their applicability in different sectors is still a compelling field of research. In this study, four Gaussian process regression (GPR) algorithms—polynomial kernel (PK), polynomial universal function kernel (PUK), normalized poly kernel (NPK), and radial basis function (RBF)—were compared against widely used random forest (RF) and a simpler locally weighted linear regression (LWLR) algorithm at a humid subtropical region in India. The sensitivity analysis of the input variables was followed by application of the best combination of variables in algorithm testing and training for generating ET0. The results were then compared against the Penman–Monteith method at both daily and monthly time steps. The results indicated that ET0 is least sensitive to wind speed at 2 m height. Additionally, at a daily time step, RF, followed by PUK, generated the best results during both training and testing phases. In contrast, at a monthly time step, using multiple model evaluation matrices, PUK followed by RF performed best. These results demonstrate the application of the ML algorithms is subjected to user-required time steps. Although this study focused on Northwest India, the findings are relevant to all humid subtropical regions across the world. 2025-01 2025-07-11T08:40:17Z 2025-07-11T08:40:17Z Journal Article https://hdl.handle.net/10568/175610 en Limited Access Wiley Deb, P.; Kumar, V.; Urfels, A.; Lautze, Jonathan; Kamboj, B. R.; Sharma, J. R.; Yadav, S. 2025. Which machine learning algorithm is best suited for estimating reference evapotranspiration in humid subtropical climate? CLEAN – Soil, Air, Water, 53(1):e202300441. doi: https://doi.org/10.1002/clen.202300441
spellingShingle machine learning
algorithms
evapotranspiration
subtropical climate
groundwater
models
Deb, P.
Kumar, V.
Urfels, A.
Lautze, Jonathan F.
Kamboj, B. R.
Sharma, J. R.
Yadav, S.
Which machine learning algorithm is best suited for estimating reference evapotranspiration in humid subtropical climate?
title Which machine learning algorithm is best suited for estimating reference evapotranspiration in humid subtropical climate?
title_full Which machine learning algorithm is best suited for estimating reference evapotranspiration in humid subtropical climate?
title_fullStr Which machine learning algorithm is best suited for estimating reference evapotranspiration in humid subtropical climate?
title_full_unstemmed Which machine learning algorithm is best suited for estimating reference evapotranspiration in humid subtropical climate?
title_short Which machine learning algorithm is best suited for estimating reference evapotranspiration in humid subtropical climate?
title_sort which machine learning algorithm is best suited for estimating reference evapotranspiration in humid subtropical climate
topic machine learning
algorithms
evapotranspiration
subtropical climate
groundwater
models
url https://hdl.handle.net/10568/175610
work_keys_str_mv AT debp whichmachinelearningalgorithmisbestsuitedforestimatingreferenceevapotranspirationinhumidsubtropicalclimate
AT kumarv whichmachinelearningalgorithmisbestsuitedforestimatingreferenceevapotranspirationinhumidsubtropicalclimate
AT urfelsa whichmachinelearningalgorithmisbestsuitedforestimatingreferenceevapotranspirationinhumidsubtropicalclimate
AT lautzejonathanf whichmachinelearningalgorithmisbestsuitedforestimatingreferenceevapotranspirationinhumidsubtropicalclimate
AT kambojbr whichmachinelearningalgorithmisbestsuitedforestimatingreferenceevapotranspirationinhumidsubtropicalclimate
AT sharmajr whichmachinelearningalgorithmisbestsuitedforestimatingreferenceevapotranspirationinhumidsubtropicalclimate
AT yadavs whichmachinelearningalgorithmisbestsuitedforestimatingreferenceevapotranspirationinhumidsubtropicalclimate