2015
DOI: 10.1016/j.jhydrol.2015.06.059
|View full text |Cite
|
Sign up to set email alerts
|

Using Bayesian model averaging to estimate terrestrial evapotranspiration in China

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1

Citation Types

1
36
0

Year Published

2016
2016
2024
2024

Publication Types

Select...
7

Relationship

1
6

Authors

Journals

citations
Cited by 64 publications
(37 citation statements)
references
References 91 publications
(104 reference statements)
1
36
0
Order By: Relevance
“…Recently, Ershadi et al (2014) showed that even the simple averaging (SA) method performed better than any individual model in estimating ET across the 20 selected FLUXNET sites. Also, the Bayesian model averaging (BMA) approach has been used to merge a range of satellite-based models for regional/global ET estimations (Vinukollu et al, 2011;Mueller et al, 2011;Yao et al, 2014;Chen et al, 2015). The results indicated that the BMA method can generally outperforms the best individual model, and provides a useful tool for generating a long-term regional/global terrestrial ET product (Yao et al, 2014;Chen et al, 2015).…”
Section: Introductionmentioning
confidence: 99%
See 1 more Smart Citation
“…Recently, Ershadi et al (2014) showed that even the simple averaging (SA) method performed better than any individual model in estimating ET across the 20 selected FLUXNET sites. Also, the Bayesian model averaging (BMA) approach has been used to merge a range of satellite-based models for regional/global ET estimations (Vinukollu et al, 2011;Mueller et al, 2011;Yao et al, 2014;Chen et al, 2015). The results indicated that the BMA method can generally outperforms the best individual model, and provides a useful tool for generating a long-term regional/global terrestrial ET product (Yao et al, 2014;Chen et al, 2015).…”
Section: Introductionmentioning
confidence: 99%
“…Also, the Bayesian model averaging (BMA) approach has been used to merge a range of satellite-based models for regional/global ET estimations (Vinukollu et al, 2011;Mueller et al, 2011;Yao et al, 2014;Chen et al, 2015). The results indicated that the BMA method can generally outperforms the best individual model, and provides a useful tool for generating a long-term regional/global terrestrial ET product (Yao et al, 2014;Chen et al, 2015). However, the success of the BMA method depends on the skill and performance of the individual members of the ensemble (Vrugt and Robinson, 2007).…”
Section: Introductionmentioning
confidence: 99%
“…Consequently, we focus here on comparing the improved versions (DSRC-R, DSRC-D, and DSRC-RD) of the DSRC method with the original version. Performance was evaluated using several measures, including the Nash-Sutcliffe Efficiency (NSE) coefficient (Nash & Sutcliffe, 1970;Schaefli & Gupta, 2007;Yapo et al, 1996), and more comprehensively, the Kling-Gupta efficiency (KGE; Gupta et al, 2009) and its components: the correlation coefficient r, standard deviation ratio α, and mean ratio β, for which mathematical definitions are provided in Appendix C. Ideally, when there are no simulation errors, we would have NSE = 1 and KGE = 1, these being achievable only when r = 1, α = 1, and β = 1 (Chen et al, 2015;Gupta et al, 2009).…”
Section: Synthetic Data Studymentioning
confidence: 99%
“…In this method, consensus predictions are derived from multiple competing predictors, and MSPD is generated to exploit the strength of every prediction. BMA method assigns weights to each member based on its predictive performance [27,42,43]. It provides a more convincing portrayal of the predictive uncertainty that considers both within-model and between-model variances and also outperformed other merging techniques by producing reliable and precise results [40,42,43].…”
mentioning
confidence: 99%
“…BMA method assigns weights to each member based on its predictive performance [27,42,43]. It provides a more convincing portrayal of the predictive uncertainty that considers both within-model and between-model variances and also outperformed other merging techniques by producing reliable and precise results [40,42,43]. This method has also been successfully applied in hydrological modeling and streamflow simulations [44][45][46].…”
mentioning
confidence: 99%