Multi-objective calibration of MIKE SHE with SMAP soil moisture datasets

Li, Dayang; Liang, Zhongmin; Li, Binquan; Lei, Xiaohui; Zhou, Yan

doi:10.2166/nh.2018.110

Cited by 12 publications

(6 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Approximately 5,000 computing grids and 55 years of daily inputs make automatic calibration prohibitive for MIKE SHE as each run needs high computing time. Ten sensitive parameters of MIKE SHE (shown in Table ) were manually calibrated with the observed streamflow and the others are identified from the soil data and previous literature (Doummar et al., 2012; Li et al., 2019; McMichael et al., 2006; Rujner et al., 2018).…”

Section: Case Studymentioning

confidence: 99%

See 1 more Smart Citation

Bayesian LSTM With Stochastic Variational Inference for Estimating Model Uncertainty in Process‐Based Hydrological Models

Marshall

Liang

et al. 2021

Water Resources Research

Self Cite

View full text Add to dashboard Cite

Significant attention has recently been paid to deep learning as a method for improved catchment modeling. Compared with process‐based models, deep learning is often criticized for its lack of interpretability. One solution is to combine a process‐based hydrological model with a residual error model based on deep learning to give full scope to their respective advantages. In classical residual error models, Bayesian inference via Markov chain Monte Carlo (MCMC) is commonly used to provide an estimation of the uncertainty. However, deep neural networks tend to have excessively large numbers of parameters, making MCMC an unsuitable approach. Here, we introduce an alternative to Bayesian MCMC sampling called stochastic variational inference (SVI) which has recently been developed for Bayesian deep learning in Natural Language Processing. We implement SVI in a Long Short‐Term Memory (LSTM) network and construct residual error models in process‐based hydrological models. This approach is examined in the contrasting geographical and climatic characteristics of two catchments from China, the Tangnaihai catchment and the Shiquan catchment. Compared with the Bayesian linear regression model, the Bayesian LSTM provides better uncertainty estimates. Specifically, the proposed method improves the Continuous Ranked Probability Score (CRPS) by over 10% in both two catchments. In the Tangnaihai catchment, it provides more than 10% narrower uncertainty intervals in terms of Sharpness with slightly superior Reliability. In the Shiquan catchment, it provides comparable uncertainty intervals with better Reliability. Further, our study highlights the scalability of SVI to high‐dimensional parameter spaces in hydrological applications (e.g., distributed hydrological models, groundwater models).

show abstract

Section: Case Studymentioning

confidence: 99%

“…This makes it very flexible, whether or not there is sufficient data in the catchment. In this study, the choice of modules was the same as Li et al (2019). The effect of snowmelt was considered with a degree-day model as the spring flow is mainly caused by snowmelt (Meng et al, 2016).…”

Section: Hydrological Modelsmentioning

confidence: 99%

Bayesian LSTM With Stochastic Variational Inference for Estimating Model Uncertainty in Process‐Based Hydrological Models

Marshall

Liang

et al. 2021

Water Resources Research

Self Cite

View full text Add to dashboard Cite

show abstract

“…SBM is generally used in semiarid or arid catchments (Bao et al, 2017). MIKE SHE is one of the most widely used physically based distributed hydrologic models (Li et al, 2018b;Refsgaard, 1995;Rujner et al, 2018;Samaras et al, 2016).…”

Section: Model Comparisonmentioning

confidence: 99%

Multicriterion assessment framework of flood events simulated with the vertically mixed runoff model in semiarid catchments in the middle Yellow River

Li¹,

Liang²,

Zhou³

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

Abstract. Flood forecasting and simulation in semiarid regions are always poor, and a single criterion assessment provides limited information for decision making. Here, we propose a multicriterion assessment framework combining the absolute relative error, the flow partitioning and the confidence interval estimated by the Hydrologic Uncertainty Processor (HUP) to assess the most striking feature of an event-based flood–the peak flow. The physically based model MIKE SHE and three conceptual models (two models with a single runoff generation mechanism, the Xi’anjiang model (XAJ) and the Shanbei model (SBM), and one model with the mixed runoff generation mechanism, the vertically mixed runoff model (VMM)) are compared in terms of flood modeling performance in four semiarid catchments (Qiushui River, Qingjian River, Tuwei River and Kuye River) in the middle Yellow River. Our results show that VMM has a better flood estimation performance than the other models, and under the multicriterion assessment framework, the average acceptance of flood events accounts for 58 %, but when absolute relative error 20 % is used as the performance criterion, its figure is only 41 % in four semiarid catchments.

show abstract

“…The lack of model input is particularly serious in some developing countries [18][19][20]. For example, high resolution thematic maps of digital elevation model (DEM), soil type, land cover and geology are difficult to obtain in many regions of China, and generally, only coarse resolution (low spatial resolution) thematic maps downloaded from the Internet could be used for the application of distributed hydrological models [21,22]. Besides, although distributed hydrological models are applied, only the river discharge type observation data are available for model test [23].…”

Section: Introductionmentioning

confidence: 99%

On the Operational Flood Forecasting Practices Using Low-Quality Data Input of a Distributed Hydrological Model

Liang

Chang

et al. 2020

Sustainability

Self Cite

View full text Add to dashboard Cite

Low-quality input data (such as sparse rainfall gauges, low spatial resolution soil type and land use maps) have limited the application of physically-based distributed hydrological models in operational practices in many data-sparse regions. It is necessary to quantify the uncertainty in the deterministic forecast results of distributed models. In this paper, the TOPographic Kinematic Approximation and Integration (TOPKAPI) distributed model was used for deterministic forecasts with low-quality input data, and then the Hydrologic Uncertainty Processor (HUP) was used to provide the probabilistic forecast results for operational practices. Results showed that the deterministic forecasts by TOPKAPI performed poorly in some flood seasons, such as the years 1997, 2001 and 2008, despite which the overall accuracy of the whole study period 1996–2008 could be acceptable and generally reproduced the hydrological behaviors of the catchment (Lushi basin, China). The HUP model can not only provide probabilistic forecasts (e.g., 90% predictive uncertainty bounds), but also provides deterministic forecasts in terms of 50% percentiles. The 50% percentiles obviously improved the forecast accuracy of selected flood events at the leading time of one hour. Besides, the HUP performance decayed with the leading time increasing (6, 12 h). This work revealed that deterministic model outputs had large uncertainties in flood forecasts, and the HUP model may provide an alternative for operational flood forecasting practices in those areas with low-quality data.

show abstract

Multi-objective calibration of MIKE SHE with SMAP soil moisture datasets

Cited by 12 publications

References 31 publications

Bayesian LSTM With Stochastic Variational Inference for Estimating Model Uncertainty in Process‐Based Hydrological Models

Bayesian LSTM With Stochastic Variational Inference for Estimating Model Uncertainty in Process‐Based Hydrological Models

Multicriterion assessment framework of flood events simulated with the vertically mixed runoff model in semiarid catchments in the middle Yellow River

On the Operational Flood Forecasting Practices Using Low-Quality Data Input of a Distributed Hydrological Model

Contact Info

Product

Resources

About