Using a Bayesian Probabilistic Forecasting Model to Analyze the Uncertainty in Real-Time Dynamic Control of the Flood Limiting Water Level for Reservoir Operation

Abstract

Water scarcity and rapid economic growth have increased the pressure on water resources and environment in Northern China, causing decreased groundwater tables, ecosystem degradation, and direct economic losses due to insufficient water supply. The authors applied the water value method, a variant of stochastic dynamic programming, to optimize water resources management in the Ziya River basin. Natural runoff from the upper basin was estimated with a rainfall-runoff model autocalibrated using in situ measured discharge. The runoff serial correlation was described by a Markov chain and used as input for the optimization model. This model was used to assess the economic impacts of ecosystem minimum flow constraints, limited groundwater pumping, and the middle route of the South-North Water Transfer Project (SNWTP). A regional climate shift has exacerbated water scarcity and increased water values, resulting in stricter water management. The results show that the SNWTP reduces the impacts of water scarcity and impacts optimal water management in the basin. The presented modeling framework provides an objective basis for the development of tools to avoid overpumping groundwater resources at minimum costs.

Reservoir operation by staging due to climate variability

“…In regard to flood season divisions, Jiang et al (2014) took both the flood regimes and the timings of flood occurrence into account to derive the most suitable method to structure the divisions of the flood season with an ensemble method structure (EMS), and obtained the final division of the flood season in the Chengbihe Reservoir as follows: early flood season (1 April-4 June), main flood season (5 June-16 August), and late flood season (17 August-31 October). Compared with the results of Liu et al(2015a), who used the probability change-point method, the flood seasons in the Longtan Reservoir, located in Tian-e county and near the Chengbihe Reservoir, could be divided into a pre-flood season (1 April-7 June), a main flood season (8 June-1 September) and a post-flood season (2 September-31 October). Thus, the onset date of the main flood season (20 May) during the period 1991-2015 in this study has been brought forward by more than the result of Jiang et al (2014) and Liu (2015), with the maximum difference being 19 d (20 May-8 June).…”

Nordic contributions to stochastic methods in hydrology

“…the shuffled complex evolution (SCE) algorithm) can be applied for reservoir operation (Le Ngo et al 2007Ngo et al , 2008. A Bayesian probabilistic forecasting model was used to assess the uncertainty of the flood limiting water level control for reservoir operation in realtime dynamic (Liu et al 2015). A compound model has been applied to optimize the reservoir rule curve; for this purpose, a simulation was coupled with a genetic algorithm and inner linear programming (Taghian et al 2014).…”

Optimizing the Flood Limit Water Level of Reservoirs in Sediment-Laden Rivers under Changing Water and Sediment Conditions: A Case Study of the Xiaolangdi Reservoir

“…The staged static FLWL is based on seasonal changes in flood patterns in the basin and adopts different FLWLs at different stages to utilize part of the original storage of flood control at a specific time to generate profits, thereby improving the flood resource utilization efficiency without lowering the flood control standard [15][16][17]. The staged dynamic FLWL refers to real-time dynamic adjustment in the FLWL control threshold of the reservoirs based on the forecast information of weather and hydrology [18][19][20][21], which further improves the flood resources utilization efficiency in the flood period based on staged static FLWL.…”