Prediction of river water temperature using machine learning algorithms: a tropical river system of India

Machine learning (ML) has been increasingly adopted due to its ability to model complex and non-linearities between river water temperature (RWT) and its predictors (e.g., Air Temperature, AT). Most of these ML approaches have been applied using average AT without any detailed sensitivity analysis o...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	M. Rajesh, S. Rehana
Formato:	article
Lenguaje:	EN
Publicado:	IWA Publishing 2021
Materias:	ensemble kalman filter k-nearest neighbors random forest river water temperature sobol’ sensitivity analysis support vector regression Information technology T58.5-58.64 Environmental technology. Sanitary engineering TD1-1066
Acceso en línea:	https://doaj.org/article/8e87599fbcb04e42b095638b11ea54c9
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:8e87599fbcb04e42b095638b11ea54c9
record_format	dspace
spelling	oai:doaj.org-article:8e87599fbcb04e42b095638b11ea54c92021-11-05T17:46:54ZPrediction of river water temperature using machine learning algorithms: a tropical river system of India1464-71411465-173410.2166/hydro.2021.121https://doaj.org/article/8e87599fbcb04e42b095638b11ea54c92021-05-01T00:00:00Zhttp://jh.iwaponline.com/content/23/3/605https://doaj.org/toc/1464-7141https://doaj.org/toc/1465-1734Machine learning (ML) has been increasingly adopted due to its ability to model complex and non-linearities between river water temperature (RWT) and its predictors (e.g., Air Temperature, AT). Most of these ML approaches have been applied using average AT without any detailed sensitivity analysis of other forms of AT (e.g., maximum and minimum). The present study demonstrates how new ML approaches, such as ridge regression (RR), K-nearest neighbors (KNN) regressor, random forest (RF) regressor, and support vector regression (SVR), can be coupled with Sobol’ global sensitivity analysis (GSA) to predict accurate RWT estimates with the most appropriate form of AT. Furthermore, the proposed ML approaches have been combined with the Ensemble Kalman Filter (EnKF), a data assimilation (DA) technique to improve the predicted values based on the measured data. The proposed modelling framework's effectiveness is demonstrated with a tropical river system of India, Tunga-Bhadra River, as a case study. The SVR has been noted as the most robust ML model to predict RWT at a monthly time scale compared with daily and seasonal. The study demonstrates how ML methods can be coupled with a global sensitivity algorithm and DA techniques to generate accurate RWT predictions in river water quality modelling. HIGHLIGHTS Machine learning models coupled with global sensitivity analysis to predict RWT.; Ridge regression, KNN, random forest, SVR, along with Sobol’ sensitivity analysis were explored.; Maximum AT as the most sensitive variable in RWT prediction.; The SVR as the most robust ML model to predict RWT at monthly time scale.; Application on a tropical river system of India.;M. RajeshS. RehanaIWA Publishingarticleensemble kalman filterk-nearest neighborsrandom forestriver water temperaturesobol’ sensitivity analysissupport vector regressionInformation technologyT58.5-58.64Environmental technology. Sanitary engineeringTD1-1066ENJournal of Hydroinformatics, Vol 23, Iss 3, Pp 605-626 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	ensemble kalman filter k-nearest neighbors random forest river water temperature sobol’ sensitivity analysis support vector regression Information technology T58.5-58.64 Environmental technology. Sanitary engineering TD1-1066
spellingShingle	ensemble kalman filter k-nearest neighbors random forest river water temperature sobol’ sensitivity analysis support vector regression Information technology T58.5-58.64 Environmental technology. Sanitary engineering TD1-1066 M. Rajesh S. Rehana Prediction of river water temperature using machine learning algorithms: a tropical river system of India
description	Machine learning (ML) has been increasingly adopted due to its ability to model complex and non-linearities between river water temperature (RWT) and its predictors (e.g., Air Temperature, AT). Most of these ML approaches have been applied using average AT without any detailed sensitivity analysis of other forms of AT (e.g., maximum and minimum). The present study demonstrates how new ML approaches, such as ridge regression (RR), K-nearest neighbors (KNN) regressor, random forest (RF) regressor, and support vector regression (SVR), can be coupled with Sobol’ global sensitivity analysis (GSA) to predict accurate RWT estimates with the most appropriate form of AT. Furthermore, the proposed ML approaches have been combined with the Ensemble Kalman Filter (EnKF), a data assimilation (DA) technique to improve the predicted values based on the measured data. The proposed modelling framework's effectiveness is demonstrated with a tropical river system of India, Tunga-Bhadra River, as a case study. The SVR has been noted as the most robust ML model to predict RWT at a monthly time scale compared with daily and seasonal. The study demonstrates how ML methods can be coupled with a global sensitivity algorithm and DA techniques to generate accurate RWT predictions in river water quality modelling. HIGHLIGHTS Machine learning models coupled with global sensitivity analysis to predict RWT.; Ridge regression, KNN, random forest, SVR, along with Sobol’ sensitivity analysis were explored.; Maximum AT as the most sensitive variable in RWT prediction.; The SVR as the most robust ML model to predict RWT at monthly time scale.; Application on a tropical river system of India.;
format	article
author	M. Rajesh S. Rehana
author_facet	M. Rajesh S. Rehana
author_sort	M. Rajesh
title	Prediction of river water temperature using machine learning algorithms: a tropical river system of India
title_short	Prediction of river water temperature using machine learning algorithms: a tropical river system of India
title_full	Prediction of river water temperature using machine learning algorithms: a tropical river system of India
title_fullStr	Prediction of river water temperature using machine learning algorithms: a tropical river system of India
title_full_unstemmed	Prediction of river water temperature using machine learning algorithms: a tropical river system of India
title_sort	prediction of river water temperature using machine learning algorithms: a tropical river system of india
publisher	IWA Publishing
publishDate	2021
url	https://doaj.org/article/8e87599fbcb04e42b095638b11ea54c9
work_keys_str_mv	AT mrajesh predictionofriverwatertemperatureusingmachinelearningalgorithmsatropicalriversystemofindia AT srehana predictionofriverwatertemperatureusingmachinelearningalgorithmsatropicalriversystemofindia
_version_	1718444112481353728

Prediction of river water temperature using machine learning algorithms: a tropical river system of India

Ejemplares similares