File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Using Bayesian model averaging to estimate terrestrial evapotranspiration in China

TitleUsing Bayesian model averaging to estimate terrestrial evapotranspiration in China
Authors
KeywordsBayesian model averaging
China
Evapotranspiration
Remote sensing
Water balance
Issue Date2015
Citation
Journal of Hydrology, 2015, v. 528, p. 537-549 How to Cite?
AbstractEvapotranspiration (ET) is critical to terrestrial ecosystems as it links the water, carbon, and surface energy exchanges. Numerous ET models were developed for the ET estimations, but there are large model uncertainties. In this study, a Bayesian Model Averaging (BMA) method was used to merge eight satellite-based models, including five empirical and three process-based models, for improving the accuracy of ET estimates. At twenty-three eddy covariance flux towers, we examined the model performance on all possible combinations of eight models and found that an ensemble with four models (BMA_Best) showed the best model performance. The BMA_Best method can outperform the best of eight models, and the Kling-Gupta efficiency (KGE) value increased by 4% compared with the model with the highest KGE, and decreased RMSE by 4%. Although the correlation coefficient of BMA_Best is less than the best single model, the bias of BMA_Best is the smallest compared with the eight models. Moreover, based on the water balance principle over the river basin scale, the validation indicated the BMA_Best estimates can explain 86% variations. In general, the results showed BMA estimates will be very useful for future studies to characterize the regional water availability over long-time series.
Persistent Identifierhttp://hdl.handle.net/10722/321637
ISSN
2023 Impact Factor: 5.9
2023 SCImago Journal Rankings: 1.764
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorChen, Yang-
dc.contributor.authorYuan, Wenping-
dc.contributor.authorXia, Jiangzhou-
dc.contributor.authorFisher, Joshua B.-
dc.contributor.authorDong, Wenjie-
dc.contributor.authorZhang, Xiaotong-
dc.contributor.authorLiang, Shunlin-
dc.contributor.authorYe, Aizhong-
dc.contributor.authorCai, Wenwen-
dc.contributor.authorFeng, Jinming-
dc.date.accessioned2022-11-03T02:20:24Z-
dc.date.available2022-11-03T02:20:24Z-
dc.date.issued2015-
dc.identifier.citationJournal of Hydrology, 2015, v. 528, p. 537-549-
dc.identifier.issn0022-1694-
dc.identifier.urihttp://hdl.handle.net/10722/321637-
dc.description.abstractEvapotranspiration (ET) is critical to terrestrial ecosystems as it links the water, carbon, and surface energy exchanges. Numerous ET models were developed for the ET estimations, but there are large model uncertainties. In this study, a Bayesian Model Averaging (BMA) method was used to merge eight satellite-based models, including five empirical and three process-based models, for improving the accuracy of ET estimates. At twenty-three eddy covariance flux towers, we examined the model performance on all possible combinations of eight models and found that an ensemble with four models (BMA_Best) showed the best model performance. The BMA_Best method can outperform the best of eight models, and the Kling-Gupta efficiency (KGE) value increased by 4% compared with the model with the highest KGE, and decreased RMSE by 4%. Although the correlation coefficient of BMA_Best is less than the best single model, the bias of BMA_Best is the smallest compared with the eight models. Moreover, based on the water balance principle over the river basin scale, the validation indicated the BMA_Best estimates can explain 86% variations. In general, the results showed BMA estimates will be very useful for future studies to characterize the regional water availability over long-time series.-
dc.languageeng-
dc.relation.ispartofJournal of Hydrology-
dc.subjectBayesian model averaging-
dc.subjectChina-
dc.subjectEvapotranspiration-
dc.subjectRemote sensing-
dc.subjectWater balance-
dc.titleUsing Bayesian model averaging to estimate terrestrial evapotranspiration in China-
dc.typeArticle-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1016/j.jhydrol.2015.06.059-
dc.identifier.scopuseid_2-s2.0-84936850744-
dc.identifier.volume528-
dc.identifier.spage537-
dc.identifier.epage549-
dc.identifier.isiWOS:000358968200044-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats