Uncertainty assessment in watershed‐scale water quality modeling and management: 1. Framework and application of generalized likelihood uncertainty estimation (GLUE) approach

Zheng, Yi; Keller, Arturo A.

doi:10.1029/2006wr005345

Cited by 40 publications

(50 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the present study, a hydrological modeling procedure was carried out using SWAT model to quantify water balance and to estimate the dynamics of hydrology. Model calibration and validation have been evaluated through sensitivity analysis (SA) and uncertainty analysis (UA) (Blasone et al 2008;Srivastava et al 2013d;Wagener and Wheater 2006;Zheng and Keller 2007). The technique of model calibration is a challenging and rigorous process, which depends on the number of input parameters, model complexity as well as iterations (Vanrolleghem et al 2003).…”

Section: Introductionmentioning

confidence: 99%

SWAT Model Calibration and Uncertainty Analysis for Streamflow Prediction in the Kunwari River Basin, India, Using Sequential Uncertainty Fitting

et al. 2015

View full text Add to dashboard Cite

The Kunwari River Basin (KRB) needs effective management of water resources for sustainable agriculture and flood hazard mitigation. The Soil and Water Assessment Tool (SWAT), a semi distributed physically based model, was chosen and set up in the KRB for hydrologic modeling. SWAT-CUP (SWAT-Calibration and Uncertainty Programs) was used for model calibration, sensitivity and uncertainty analysis, following the Sequential Uncertainty Fitting (SUFI-2) technique. The model calibration was performed for the period (1987)(1988)(1989)(1990)(1991)(1992)(1993)(1994)(1995)(1996)(1997)(1998)(1999), with initial 3 years of warm up ; then, the model was validated for the subsequent 6 years of data (2000)(2001)(2002)(2003)(2004)(2005). To assess the competence of model calibration and uncertainty, two indices, the p-factor (observations bracketed by the prediction uncertainty) and the r-factor (achievement of small uncertainty band), were taken into account. The results Environ. Process. of SWAT simulations indicated that during the calibration the p-factor and the r-factor were reported as 0.82 and 0.76, respectively, while during the validation the p-factor and the r-factor were obtained as 0.71 and 0.72, respectively. After a rigorous calibration and validation, the goodness of fit was further assessed through the use of the coefficient of determination (R 2 ) and the Nash-Sutcliffe efficiency (NS) between the observed and the final simulated values. The results indicated that R 2 and NS were 0.77 and 0.74, respectively, during the calibration. The validation also indicated a satisfactory performance with R 2 of 0.71 and NS of 0.69. The results would be useful to the hydrological community, water resources managers involved in agricultural water management and soil conservation, as well as to those involved in mitigating natural hazards such as droughts and floods.

show abstract

Section: Introductionmentioning

confidence: 99%

SWAT Model Calibration and Uncertainty Analysis for Streamflow Prediction in the Kunwari River Basin, India, Using Sequential Uncertainty Fitting

et al. 2015

View full text Add to dashboard Cite

show abstract

“…However, as more sub-processes are being incorporated, many of these physical-based models have become highly complex and involve many parameters (Beven 2006). Thus, uncertainty is inevitable in H/NPS predictions and studies have been conducted to address the sources of this uncertainty (Shen et al 2012a), through quantification (Zheng and Keller 2007), propagation (Naranjo et al 2012), and evaluation . Recently, watershed managers have gained a better understanding of existing uncertainties, but additional scientific knowledge regarding the control of uncertainty is needed to accurately predict H/NPS (Beven et al 2008).…”

Section: Introductionmentioning

confidence: 99%

Using site-specific soil samples as a substitution for improved hydrological and nonpoint source predictions

Lei

Wang

Zhong

et al. 2016

Environ Sci Pollut Res

View full text Add to dashboard Cite

Soil databases are one of the most important inputs for watershed models, and the quality of soil properties affects how well a model performs. The objectives of this study were to (1) quantify the sensitivity of model outputs to soil properties and to (2) use site-specific soil properties as a substitution for more accurate hydrological and nonpoint source (H/NPS) predictions. Soil samples were collected from a typical mountainous watershed in China, and the impacts of soil sample parameters on H/NPS predictions were quantified using the Soil and Water Assessment Tool (SWAT). The most sensitive parameters related to predicting flow, sediment, and total phosphorus (TP) mainly were the soil hydrological, the channel erosion processes, and the initial soil chemical environment, respectively. When the site-specific soil properties were used, the uncertainties (coefficient of variation) related to predicting the hydrology, sediment and TP decreased by 75∼80 %, 75∼84 %, and 46∼61 %, respectively. Based on changes in the Nash-Sutcliff coefficient, the model performance improved by 4.9 and 19.45 % for the hydrological and sediment model, accordingly. However, site-specific soil properties did not contribute to better TP predictions because of the high spatial variability of the soil P concentrations across the large watershed. Thus, although site-specific soil samples can be used to obtain more accurate H/NPS predictions, more sampling sites are required to apply this method in large watersheds.

show abstract

“…A number of UA techniques have been developed in the field of hydrological modeling (Montanari et al 2009;Zheng and Keller 2007a), among which Markov Chain Monte Carlo (MCMC) has drawn great attention in recent years (Smith and Marshall 2008;Vrugt et al 2009a, b). MCMC represents a category of formal Bayesian approaches for parameter estimation and uncertainty quantification.…”

Section: Introductionmentioning

confidence: 99%

Markov Chain Monte Carlo (MCMC) uncertainty analysis for watershed water quality modeling and management

Zheng

Feng

2015

Stoch Environ Res Risk Assess

Self Cite

View full text Add to dashboard Cite

Watershed-scale water quality (WWQ) models are now widely used to support management decisionmaking. However, significant uncertainty in the model outputs remains a largely unaddressed issue. In recent years, Markov Chain Monte Carlo (MCMC), a category of formal Bayesian approaches for uncertainty analysis (UA), has become popular in the field of hydrological modeling, but its applications to WWQ modeling have been rare. This study systematically evaluated the applicability of MCMC in assessing the uncertainty of WWQ modeling, using Differential Evolution Adaptive Metropolis (DREAM (ZS) ) and SWAT as the representative MCMC algorithm and WWQ model, respectively. The nitrate pollution in Newport Bay watershed was the case study for numerical experiments. It has been concluded that the efficiency and effectiveness of a MCMC algorithm would depend on some critical designs of the UA, including: (i) how many and which model parameters to be considered as random in the MCMC analysis; (ii) where to fix the non-random model parameters; and (iii) which criteria to stop the Markov Chain. The study results also indicate that the MCMC UA has to be management-oriented, that is, management objectives should be factored into the designs of the UA, rather than be considered after the UA.

show abstract

Uncertainty assessment in watershed‐scale water quality modeling and management: 1. Framework and application of generalized likelihood uncertainty estimation (GLUE) approach

Cited by 40 publications

References 46 publications

SWAT Model Calibration and Uncertainty Analysis for Streamflow Prediction in the Kunwari River Basin, India, Using Sequential Uncertainty Fitting

SWAT Model Calibration and Uncertainty Analysis for Streamflow Prediction in the Kunwari River Basin, India, Using Sequential Uncertainty Fitting

Using site-specific soil samples as a substitution for improved hydrological and nonpoint source predictions

Markov Chain Monte Carlo (MCMC) uncertainty analysis for watershed water quality modeling and management

Contact Info

Product

Resources

About