An Uncertainty Partition Approach for Inferring Interactive Hydrologic Risks

Fan, Yurui; Huang, Kai; Huang, Guohe; Li, Yongping; Wang, Rui

doi:10.5194/hess-2019-434

Cited by 4 publications

(3 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Factorial designs are the cornerstone of industrial experimentation and used extensively in industrial research and development for process improvement, among which the multilevel factorial design is a powerful statistical technique to study the effects of several independent variables (factors) with multiple levels on a dependent variable (response) (Wang and Huang, 2015;Fan et al, 2020aFan et al, , 2020bFan et al, , 2021. A 3 k factorial design is proposed for screening the effects of parameters in a copula model on its performance for evaluating flooding risk.…”

Section: Interactive and Sensitivity Analysis For Parameters Uncertai...mentioning

confidence: 99%

Parameter Uncertainty and Sensitivity Evaluation of Copula-Based Multivariate Hydroclimatic Risk Assessment

Huang¹,

Fan²

2021

J ENVIRON INFORM LET

View full text Add to dashboard Cite

Extensive uncertainties exist in hydroclimatic risk analysis. Especially in multivariate hydrologic risk inferences, uncertainties in individual hydroclimatic extremes such as floods and their dependence structure may lead to bias and uncertainty in future hydrologic risk predictions. In this study, a parameter uncertainty and sensitivity evaluation (PUSE) framework is proposed to quantify parameter uncertainties and then reveal their contributions to the multivariate hydroclimatic risk predictions. The predictive risks are finally generated by "integrating" the values over the posterior distributions of the parameters. The proposed approach was applied for bivariate risk analysis of compound floods at the Xiangxi River to characterize the concurrence probabilities of flood peaks and volumes. The results demonstrate that the proposed approach can quantify uncertainties in a copula-based multivariate risk analysis and characterize effects and contributions of parameters in marginal and dependence structures on the multivariate hydroclimatic risk predictions. In terms of the bivariate risk for flood peak and volume at the Xiangxi River, uncertainties in model parameters would lead to noticeable uncertainties even for moderate floods. The performances of the copula model for flood peak-volume at Xiangxi River are mainly affected by the uncertainties in location parameters of the two individual flood variables. Also, parameter uncertainty in the dependence structure (i.e., copula) would also poses explicit impacts on performance of the copula-based risk analyses model. These uncertainties would result into higher bivariate predictive risks than the values obtained by "optimal/deterministic" predictions. This indicates that uncertainties are required to be considered to provide reliable multivariate hydroclimatic risk predictions.

show abstract

Section: Interactive and Sensitivity Analysis For Parameters Uncertai...mentioning

confidence: 99%

Parameter Uncertainty and Sensitivity Evaluation of Copula-Based Multivariate Hydroclimatic Risk Assessment

Huang¹,

Fan²

2021

J ENVIRON INFORM LET

View full text Add to dashboard Cite

show abstract

“…Precipitation varies from year to year and over decades and changes in quantity, frequency and intensity with a direct influence of global warming (Trenberth, 2011). The impact of extreme precipitation events on the populations and economies in affected areas has drawn widespread attention (Fan et al, 2020;He et al, 2021). The unusual behavior of summer precipitation often causes local flooding during summer.…”

Section: Introductionmentioning

confidence: 99%

Extreme Summer Precipitation Events in China and Their Changes during 1982 ~ 2019

Wang¹,

Duan²,

Li³

et al. 2021

J ENVIRON INFORM LET

View full text Add to dashboard Cite

This work analyzed the spatial and temporal variations of the extreme precipitation over China during the period 1982 to 2019, based on GPCC data. Meanwhile, the spatially clustering characteristic of China is evaluated based on the trends of summer precipitation through K-means. This is the first to clustering based on the trend of summer precipitation. The results indicated that the spatial distribution of MK trends for the summer precipitation indices over China during 1982 ~ 2019 shows increasing and decreasing trends in different regions. The extreme precipitation has been found mainly in Pearl River Basin, Southeastern River Basin, Huaihe River Basin, and the lower reaches of the Yangtze River. The risk of floods in SERB (the Southeastern River Basin), especially those caused by short-term heavy precipitation, may increase in recent decades. Four clusters are identified through K-means, which can be named as stable (59%), extreme wetter (5%), dryer (19%), and wetter (17%) zones. Combining the spatial distribution of the multi-year average of precipitation indicators and four clusters, the 'wet to wetter and dry to dryer' is found in China summer precipitation.

show abstract

“…With the aim of assessing the flow discharge at a given location, they can take into account individual gauge uncertainties in the form of probabilistic distributions. More importantly, they allow the decomposition of uncertainty components related to the model of the gauging curve ("structural error"), parameter estimation ("parametric error") and potentially systematic and non‐systematic errors in the water level series ("propagation error") (Boutkhamouine, Roux, & Pérès, 2017, Boutkhamouine, Roux, Pérès, & Vervoort, 2018, Jiang et al, , 2019, Fan, 2019; Leandro, 2019, Dittes, Špačková, & Straub, 2019).…”

Section: Introductionmentioning

confidence: 99%

Data‐driven model for river flood forecasting based on a Bayesian network approach

Boutkhamouine

Roux

Pérès

2020

Contingencies & Crisis Mgmt

View full text Add to dashboard Cite

Uncertainty analysis of hydrological models often requires a large number of model runs, which can be time consuming and computationally intensive. In order to reduce the number of runs required for uncertainty prediction, Bayesian networks (BNs) are used to graphically represent conditional probability dependence between the set of variables characterizing a flood event. Bayesian networks (BNs) are relevant due to their capacity to handle uncertainty, combine statistical data and expertise and introduce evidences in real‐time flood forecasting. In the present study, a runoff–runoff model is considered. The discharge at a gauging station located is estimated at the outlet of a basin catchment based on discharge measurements at the gauging stations upstream. The BN model shows good performances in estimating the discharges at the basin outlet. Another application of the BN model is to be used as a reverse method. Knowing discharges values at the outlet of the basin, we can propagate back these values through the model to estimate discharges at upstream stations. This turns out to be a practical method to fill the missing data in streamflow records which are critical to the sustainable management of water and the development of hydrological models.

show abstract

An Uncertainty Partition Approach for Inferring Interactive Hydrologic Risks

Cited by 4 publications

References 28 publications

Parameter Uncertainty and Sensitivity Evaluation of Copula-Based Multivariate Hydroclimatic Risk Assessment

Parameter Uncertainty and Sensitivity Evaluation of Copula-Based Multivariate Hydroclimatic Risk Assessment

Extreme Summer Precipitation Events in China and Their Changes during 1982 ~ 2019

Data‐driven model for river flood forecasting based on a Bayesian network approach

Contact Info

Product

Resources

About