Intercomparing the robustness of machine learning models in simulation and forecasting of streamflow

The intercomparison of streamflow simulation and the prediction of discharge using various renowned machine learning techniques were performed. The daily streamflow discharge model was developed for 35 observation stations located in a large-scale river basin named Cauvery. Various hydrological indi...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Parthiban Loganathan, Amit Baburao Mahindrakar
Formato: article
Lenguaje:EN
Publicado: IWA Publishing 2021
Materias:
Acceso en línea:https://doaj.org/article/15d1237f55504149ba753cb71a61f845
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:15d1237f55504149ba753cb71a61f845
record_format dspace
spelling oai:doaj.org-article:15d1237f55504149ba753cb71a61f8452021-11-05T19:02:02ZIntercomparing the robustness of machine learning models in simulation and forecasting of streamflow2040-22442408-935410.2166/wcc.2020.365https://doaj.org/article/15d1237f55504149ba753cb71a61f8452021-08-01T00:00:00Zhttp://jwcc.iwaponline.com/content/12/5/1824https://doaj.org/toc/2040-2244https://doaj.org/toc/2408-9354The intercomparison of streamflow simulation and the prediction of discharge using various renowned machine learning techniques were performed. The daily streamflow discharge model was developed for 35 observation stations located in a large-scale river basin named Cauvery. Various hydrological indices were calculated for observed and predicted discharges for comparing and evaluating the replicability of local hydrological conditions. The model variance and bias observed from the proposed extreme gradient boosting decision tree model were less than 15%, which is compared with other machine learning techniques considered in this study. The model Nash–Sutcliffe efficiency and coefficient of determination values are above 0.7 for both the training and testing phases which demonstrate the effectiveness of model performance. The comparison of monthly observed and model-predicted discharges during the validation period illustrates the model's ability in representing the peaks and fall in high-, medium-, and low-flow zones. The assessment and comparison of hydrological indices between observed and predicted discharges illustrate the model's ability in representing the baseflow, high-spell, and low-spell statistics. Simulating streamflow and predicting discharge are essential for water resource planning and management, especially in large-scale river basins. The proposed machine learning technique demonstrates significant improvement in model efficiency by dropping variance and bias which, in turn, improves the replicability of local-scale hydrology. HIGHLIGHTS The credibility of machine learning models in representing the regional-scale hydrology is performed.; Evaluation to prioritize model selection for river basin management.; Season-based approach in evaluating model performance in local hydrology.; Hydrological indices were inter-compared for high-, medium-, and low-flow zones.; Outcome delivers valuable suggestions to decision-makers in the planning of future water resources.;Parthiban LoganathanAmit Baburao MahindrakarIWA Publishingarticlecauvery river basinclimate changehydrological modelmachine learningstreamflowEnvironmental technology. Sanitary engineeringTD1-1066Environmental sciencesGE1-350ENJournal of Water and Climate Change, Vol 12, Iss 5, Pp 1824-1837 (2021)
institution DOAJ
collection DOAJ
language EN
topic cauvery river basin
climate change
hydrological model
machine learning
streamflow
Environmental technology. Sanitary engineering
TD1-1066
Environmental sciences
GE1-350
spellingShingle cauvery river basin
climate change
hydrological model
machine learning
streamflow
Environmental technology. Sanitary engineering
TD1-1066
Environmental sciences
GE1-350
Parthiban Loganathan
Amit Baburao Mahindrakar
Intercomparing the robustness of machine learning models in simulation and forecasting of streamflow
description The intercomparison of streamflow simulation and the prediction of discharge using various renowned machine learning techniques were performed. The daily streamflow discharge model was developed for 35 observation stations located in a large-scale river basin named Cauvery. Various hydrological indices were calculated for observed and predicted discharges for comparing and evaluating the replicability of local hydrological conditions. The model variance and bias observed from the proposed extreme gradient boosting decision tree model were less than 15%, which is compared with other machine learning techniques considered in this study. The model Nash–Sutcliffe efficiency and coefficient of determination values are above 0.7 for both the training and testing phases which demonstrate the effectiveness of model performance. The comparison of monthly observed and model-predicted discharges during the validation period illustrates the model's ability in representing the peaks and fall in high-, medium-, and low-flow zones. The assessment and comparison of hydrological indices between observed and predicted discharges illustrate the model's ability in representing the baseflow, high-spell, and low-spell statistics. Simulating streamflow and predicting discharge are essential for water resource planning and management, especially in large-scale river basins. The proposed machine learning technique demonstrates significant improvement in model efficiency by dropping variance and bias which, in turn, improves the replicability of local-scale hydrology. HIGHLIGHTS The credibility of machine learning models in representing the regional-scale hydrology is performed.; Evaluation to prioritize model selection for river basin management.; Season-based approach in evaluating model performance in local hydrology.; Hydrological indices were inter-compared for high-, medium-, and low-flow zones.; Outcome delivers valuable suggestions to decision-makers in the planning of future water resources.;
format article
author Parthiban Loganathan
Amit Baburao Mahindrakar
author_facet Parthiban Loganathan
Amit Baburao Mahindrakar
author_sort Parthiban Loganathan
title Intercomparing the robustness of machine learning models in simulation and forecasting of streamflow
title_short Intercomparing the robustness of machine learning models in simulation and forecasting of streamflow
title_full Intercomparing the robustness of machine learning models in simulation and forecasting of streamflow
title_fullStr Intercomparing the robustness of machine learning models in simulation and forecasting of streamflow
title_full_unstemmed Intercomparing the robustness of machine learning models in simulation and forecasting of streamflow
title_sort intercomparing the robustness of machine learning models in simulation and forecasting of streamflow
publisher IWA Publishing
publishDate 2021
url https://doaj.org/article/15d1237f55504149ba753cb71a61f845
work_keys_str_mv AT parthibanloganathan intercomparingtherobustnessofmachinelearningmodelsinsimulationandforecastingofstreamflow
AT amitbaburaomahindrakar intercomparingtherobustnessofmachinelearningmodelsinsimulationandforecastingofstreamflow
_version_ 1718444021795258368