Dongting Lake Water Level Forecast and Its Relationship with the Three Gorges Dam Based on a Long Short-Term Memory Network

Worldwide, many river floodplains contain critical infrastructure that is vulnerable to extreme hydrologic events. These structures are designed based on flood frequency analysis aimed at quantifying the magnitude and recurrence of the extreme events. This research topic focuses on estimating flood vulnerability at ungauged locations based on an integrative framework consisting of a distributed rainfall-runoff model forced with long-term (37 years) reanalysis meteorological data and a hydraulic model driven by high-resolution airborne LiDAR-derived terrain elevation data. The framework is applied to a critical power infrastructure located within Connecticut's Naugatuck River Basin. The hydrologic model reanalysis is used to derive 50-, 100-, 200-, and 500-year return period flood peaks, which are then used to drive Hydrologic Engineering Center's River Analysis System (HEC-RAS) hydraulic simulations to estimate the inundation risk at the infrastructure location under different operation strategies of an upstream reservoir. This study illustrates the framework's potential for creating flood maps at ungauged locations and demonstrates the effects of different water management scenarios on the flood risk of the downstream infrastructure. 500 years return period). Distributions used in frequency analysis vary, but in the U.S.A. engineering practice relies on the use of Log-Pearson Type III, which is recommended by the U.S. Water Resource Council [3]. Information regarding magnitude and frequency of occurrence of flooding events gathered through flood frequency analysis (FFA) is instrumental in mitigating losses associated with floods, particularly when designing hydraulic structures such as reservoirs and dams.Hydrologists have developed a number of methods to conduct FFA. These methods can broadly be classified into two groups: statistical approaches and rainfall-runoff modeling. For statistical approaches, statistical analysis or hydrological regionalization are used to analyze hydrological data within a basin, or transfer hydrological information from one or more homogenous gauged catchments to a neighboring or geographically/hydrologically similar basin [4][5][6]. However, the quality of the estimation in the basin is subject to the continuous length of flow observations of the basin or neighboring gauged basins, and potential nonstationarity of the historical flood trend. In an alternate approach, observed precipitation combined with various other meteorological forcing parameters are used to drive physically-based hydrologic and flood routing models to simulate surface flows [7], which can have longer time spans but can be affected by the biased peak estimation. Surface flows are made up of both overland and channel flow routing, which can either be handled separately by two models [8,9], or combined through a coupled model [10,11]. Hydrodynamic routing models simulate surface flows based on solutions to simplified shallow water equations like the kinematic wave [12] or diffusion wave equations [9]. Solutions t...

mentioning

confidence: 99%

A Numerical Framework for Evaluating Flood Inundation Hazard under Different Dam Operation Scenarios—A Case Study in Naugatuck River

hardesty

Shen

Nikolopoulos

et al. 2018

“…(2) Establish a dataset: For each sample set, the training set (60 flood events) and validation set (15 flood events) are restructured to a supervised learning dataset by sliding window method [25]. In the supervised learning dataset, the sample size of the supervised learning training set is about 9000, and the sample size of the supervised learning validation set is about 2000.…”

Section: Model Trainingmentioning

confidence: 99%

Uncertainty Quantification in Machine Learning Modeling for Multi-Step Time Series Forecasting: Example of Recurrent Neural Networks in Discharge Simulations

Song

Ding

Liu

et al. 2020

As a revolutionary tool leading to substantial changes across many areas, Machine Learning (ML) techniques have obtained growing attention in the field of hydrology due to their potentials to forecast time series. Moreover, a subfield of ML, Deep Learning (DL) is more concerned with datasets, algorithms and layered structures. Despite numerous applications of novel ML/DL techniques in discharge simulation, the uncertainty involved in ML/DL modeling has not drawn much attention, although it is an important issue. In this study, a framework is proposed to quantify uncertainty contributions of the sample set, ML approach, ML architecture and their interactions to multi-step time-series forecasting based on the analysis of variance (ANOVA) theory. Then a discharge simulation, using Recurrent Neural Networks (RNNs), is taken as an example. Long Short-Term Memory (LSTM) network, a state-of-the-art DL approach, was selected due to its outstanding performance in time-series forecasting, and compared with simple RNN. Besides, novel discharge forecasting architecture is designed by combining the expertise of hydrology and stacked DL structure, and compared with conventional design. Taking hourly discharge simulations of Anhe (China) catchment as a case study, we constructed five sample sets, chose two RNN approaches and designed two ML architectures. The results indicate that none of the investigated uncertainty sources are negligible and the influence of uncertainty sources varies with lead-times and discharges. LSTM demonstrates its superiority in discharge simulations, and the ML architecture is as important as the ML approach. In addition, some of the uncertainty is attributable to interactions rather than individual modeling components. The proposed framework can both reveal uncertainty quantification in ML/DL modeling and provide references for ML approach evaluation and architecture design in discharge simulations. It indicates uncertainty quantification is an indispensable task for a successful application of ML/DL.

“…Considering the representativeness of the peak discharge, we replaced one flood event of the validation set and two flood events of the test set to the training set (see Figure 4). The data from 75 flood events is restructured to a supervised learning dataset by a sliding window method [25]. The sample size of the training set, validation set and test set is 6595 (45 flood events), 2334 (15 flood events), and 2109 (15 flood events) for a 1 h lead-time, respectively.…”

Section: Training Processmentioning

confidence: 99%

Flash Flood Forecasting Based on Long Short-Term Memory Networks

Song

Ding

et al. 2019

Flash floods occur frequently and distribute widely in mountainous areas because of complex geographic and geomorphic conditions and various climate types. Effective flash flood forecasting with useful lead times remains a challenge due to its high burstiness and short response time. Recently, machine learning has led to substantial changes across many areas of study. In hydrology, the advent of novel machine learning methods has started to encourage novel applications or substantially improve old ones. This study aims to establish a discharge forecasting model based on Long Short-Term Memory (LSTM) networks for flash flood forecasting in mountainous catchments. The proposed LSTM flood forecasting (LSTM-FF) model is composed of T multivariate single-step LSTM networks and takes spatial and temporal dynamics information of observed and forecast rainfall and early discharge as inputs. The case study in Anhe revealed that the proposed models can effectively predict flash floods, especially the qualified rates (the ratio of the number of qualified events to the total number of flood events) of large flood events are above 94.7% at 1-5 h lead time and range from 84.2% to 89.5% at 6-10 h lead-time. For the large flood simulation, the small flood events can help the LSTM-FF model to explore a better rainfall-runoff relationship. The impact analysis of weights in the LSTM network structures shows that the discharge input plays a more obvious role in the 1-h LSTM network and the effect decreases with the lead-time. Meanwhile, in the adjacent lead-time, the LSTM networks explored a similar relationship between input and output. The study provides a new approach for flash flood forecasting and the highly accurate forecast contributes to prepare for and mitigate disasters.Water 2020, 12, 109 2 of 15 concluded no model could make reliable flash flood forecasts in spite of the plausible results of physically-based distributed hydrological models. Many studies showed that distributed hydrological models have advantages over lumped hydrological models and data-driven models, but they are computationally inefficient and need high-resolution sophisticated input data (e.g., DEM, land-use and soil maps, and soil characteristics) [5]. Hence, their applicability is limited in mountainous catchments. The expected benefits of using high-resolution distributed models might be masked by the increasing uncertainties at small scales [6]. Lumped hydrological models for flash flood forecasting are limited by their coarse resolution and inadequate description of rainfall spatial distribution, which has a great impact on the catchment response [2]. In addition, physically-based hydrological models depend heavily on their boundary conditions, which are often poorly defined [7]. It is difficult to describe flash flood generation and propagation by a deterministic approach due to the complexity of its processes. Numerous studies indicate the gap of physically-based hydrological models in short-term flood prediction [8].With the advancements in sys...