Assessing the Impacts of Univariate and Bivariate Flood Frequency Approaches to Flood Risk Accounting for Reservoir Operation

Abstract

Flood frequency analysis plays a fundamental role in dam planning, reservoir operation, and risk assessment. However, conventional univariate flood frequency analysis carried out by flood peak inflow or volume does not account for the dependence between flood properties. In this paper, we proposed an integrated approach to estimate reservoir risk by combining the copula-based bivariate flood frequency (peak and volume) and reservoir routing. Through investigating the chain reaction of “flood frequency—reservoir operation-flood risk”, this paper demonstrated how to simulate flood hydrographs using different frequency definitions (copula “Or” and “And” scenario), and how these definitions affect flood risks. The approach was applied to the Meishan reservoir in central China. A set of flood hydrographs with 0.01 frequency under copula “Or” and “And” definitions were constructed, respectively. Upstream and downstream flood risks incorporating reservoir operation were calculated for each scenario. Comparisons between flood risks from univariate and bivariate flood frequency analysis showed that bivariate flood frequency analysis produced less diversity in the results, and thus the results are more reliable in risk assessment. More importantly, the peak-volume combinations in a bivariate approach can be adjusted according to certain prediction accuracy, providing a flexible estimation of real-time flood risk under different prediction accuracies and safety requirements.

Complete multivariate flood frequency analysis, applied to northern Algeria

“…Through the combination of two or more hydrological variables, mainly Qp and V , provide more reliability for hydraulic structure designing, water reservoir management, flood risk assessment and support flood mitigation in the studied basins (e.g. Balistrocchi et al, 2017; Callau Poduje et al, 2014; Jiang et al, 2019; Liu et al, 2019; Reddy & Ganguli, 2012; Requena et al, 2013; Salvadori & De Michele, 2004; Zhang & Singh, 2006; Zhou et al, 2019).…”

“…Multivariate FFA constitutes a powerful tool allowing the dimensioning of suitable installations based on a more accurate flood risk assessment (e.g. Balistrocchi, Orlandini, Ranzi, & Bacchi, 2017; Callau Poduje, Belli, & Haberlandt, 2014; Jiang, Yang, & Tatano, 2019; Liu, Li, Ma, Jia, & Su, 2019; Zhou, Liu, Jin, & Hu, 2019).…”

Comparative analysis of routed flood frequency for reservoirs in parallel incorporating bivariate flood frequency and reservoir operation

“…Root‐mean‐square error (RMSE) is adopted as an indicator to quantify the proximity: Once the joint distribution is derived, some multivariate flood frequency definitions can be obtained, such as the copula “Or” (denoted as FC or ( u , v )) and “And” frequency (denoted as FC and ( u , v )). Since univariate flood frequency is described as P ( X ≥ x ), the copula “Or” frequency denotes the probability that either variable exceeds certain threshold, expressed as P ( X ≥ x ∨ Y ≥ y ), and the copula “And” frequency denotes the probability that both variables exceed certain threshold, expressed as P ( X ≥ x ∧ Y ≥ y ) (Zhou et al, 2019). FC or ( u , v ) and FC and ( u , v ) can be derived from C( u , v ): Once FC or ( u , v ) and FC and ( u , v ) are derived, peak–peak combinations with certain frequency can be obtained, and their corresponding flood volumes can be derived from the regression functions in Figure 4.…”

“…The RFFC describes flood frequency by considering the effect of reservoir routing as well as flood peak‐volume bivariate characteristics, which reflects real‐time flood risk under reservoir operation environment. Zhou, Liu, Jin, and Hu (2019) constructed a bivariate flood frequency definition that incorporating upstream and downstream flood risk on single reservoir, and comprehensively compared flood results of copula “Or” frequency and copula “And” frequency.…”

Flood Risk Analysis of Different Climatic Phenomena during Flood Season Based on Copula-Based Bayesian Network Method: A Case Study of Taihu Basin, China

“…Several factors can influence the risk of flood disasters, such as astronomy, meteorology, hydrology, topography, landforms, and human activities. Obviously, the interaction between such heterogeneous factors cannot be described clearly by using only one variable and, therefore, the study of multivariable joint distribution models is a significant facet of research on flood risk analysis [14,15].…”