Water Level Prediction Research Articles

EKLT: Kolmogorov-Arnold attention-driven LSTM with Transformer model for river water level prediction

Water level prediction is crucial for water resource management and flood warnings. Hybrid LSTM-based methods are widely used but face the following challenges: (1) The attention mechanism based on the universal approximation theorem (UAT) is the mainstream method for improving performance, but its essence is infinite approximation of functions, and the accuracy is difficult to improve; (2) LSTM struggles with modeling complex, long-term dependencies. To address these problems, we apply empirical mode decomposition (EMD) and propose input-spatial attention based on the Kolmogorov-Arnold theorem (KAT) for precise water level feature representation. Cascaded LSTM and Transformer structures capture long-term dependencies. Finally, temporal attention is embedded to achieve accurate water level prediction. Experiments show the proposed method achieves RMSE, MAE, MAPE and R2 values of 0.1870, 0.1328, 1.1228 and 0.9540 in the Liaohe River of Liaoning Province, China, and 0.3027, 0.1844, 4.6034 and 0.8659 in the Hunhe River, respectively, indicating its strong predictive performance.

A probabilistic integration of LSTM and Gaussian process regression for uncertainty-aware reservoir water level predictions

ABSTRACT Reservoir-level forecasting, while being crucial for optimal operation, is challenged by complex physical processes and changing climate conditions. Machine learning approaches offer deterministic predictions but often neglect system physics and uncertainty. This article presents a probabilistic data-driven approach combining long short-term memory (LSTM) and Gaussian process regression (GPR) to provide both point forecasts and uncertainty estimates. The hybrid model leverages LSTM’s fitting capabilities with GPR’s robust Bayesian frameworks for uncertainty estimation in non-linear problems, offering accurate predictions without extensive high-fidelity modelling, and avoiding frequent training and parameter optimization. Evaluation with real reservoir data from India shows the model’s superiority over the vanilla LSTM for both univariate and multivariate scenarios. The proposed model achieved a Nash-Sutcliffe efficiency of 0.97 to 0.98, a mean biased error of −0.5634 to −1.0314 for 10-day forecasts, and a continuous ranked probability score of 5.80 and 1.87 for the Bhakra and Pong reservoirs, respectively.

Enhancing flood forecasting and warning precision through multi-task deep learning approaches

ABSTRACT This study proposes a multi-task deep learning model for simultaneous prediction of time-series water levels and flood risk thresholds, aiming to enhance flood forecasting precision. Using AutoKeras, single-task and multi-task models were optimised to predict water levels 10–360 min ahead based on 720 min of prior data. The multi-task model consistently outperformed the single-task model across multiple evaluation metrics, including correlation coefficients, RMSE, Nash–Sutcliffe efficiency, and Kling–Gupta efficiency scores. Real-time prediction tests on actual rainfall events further validated the multi-task model's improved accuracy and applicability in operational flood forecasting. The study demonstrates significant progress in flood prediction methodologies, offering a more comprehensive approach to forecasting and categorising flood incidents.

Comparison of FLUMEN and HEC-RAS 2D Models for Flash Flood Inundation in Urban Landscape

Flood inundation maps play a crucial role in preventing flood damage. Among various numerical models used to generate these maps, the fluvial modeling engine (FLUMEN) and the hydrologic engineering center’s river analysis system (HEC-RAS 2D) are particularly effective for simulating urban floods, which are influenced by complex factors such as buildings and landscapes. This study aims to examine the differences in flood analysis results that may arise from using different numerical models. This study compares the performance of FLUMEN and HEC-RAS 2D in modeling urban flash floods, characterized by local velocity variations and complex geometries. The analysis focuses on their numerical characteristics and simulation accuracy. The simulation results show that HEC-RAS 2D outperforms FLUMEN in handling turbulence, numerical stability, and peak water level predictions. These findings provide insights into the strengths and limitations of each model for urban flood management.

Development and Applicability Assessment of an Hourly Water Level Prediction Model for Agricultural Reservoirs Using Automated Machine Learning

Objectives:This study aims to develop and evaluate hourly water level prediction models for agricultural reservoirs using an automated machine learning (AutoML) approach, specifically, employing TPOT.Methods:The study focuses on the Baekrok and Baekryeon reservoirs using rainfall forecast data from the Korea Meteorological Administration, along with observed rainfall and reservoir water level data. TPOT, which utilizes genetic algorithms to automate the generation of optimal machine learning pipelines, was used to build models. Additionally, Random Forest (RF) was implemented as a benchmark to evaluate TPOT’s applicability. Predictions were generated with lead times of 1, 3, 6, and 12 hours, and model accuracy was evaluated using the Nash-Sutcliffe Efficiency (NSE), coefficient of determination (R2), and root-mean-square error (RMSE).Results and Discussion:The predictions from both TPOT and RF showed high accuracy for shorter lead times, with NSE values exceeding 0.99 for the 1-hour predictions. However, predictive accuracy decreased as lead time increased, likely due to greater uncertainty in rainfall forecast data, particularly for the 12-hour predictions. Despite this trend, TPOT-derived models maintained more stable and accurate performance compared to RF models across all lead times.Conclusion:This study demonstrates the applicability of TPOT-based AutoML techniques in predicting hourly water levels in agricultural reservoirs. Future work should explore strategies to mitigate the decline in accuracy for longer lead times.

Sea Level Prediction Using Gated Recurrent Unit and Bidirectional Long Short-Term Memory Methods

Coastal areas are areas that border between land and sea. One of the main threats to this area is flooding due to rising sea levels. The coastal areas in the South Lampung Waters are geographically located directly opposite the Sunda Strait. In this area, there are various economic activities of the community and settlements that are greatly influenced by sea dynamics, including tides. This study aims to predict sea level in the South Lampung Waters using the GRU method and compare the prediction results of the model with the BiLSTM model. This study involves steps such as data collection, data pre-processing, model training, modeling, evaluation and analysis. The data used is sea level data for one year, from June 1, 2023 to May 31, 2024, which is divided into training data (80%) and test data (20%). Evaluation of the test results and predictions of the two models is carried out using MAE and RMSE. The results of the research that has been carried out, the BiLSTM method is superior to the GRU method in predicting water levels. This can be seen from the MAE and RMSE values ​​obtained, where the BiLSTM method has the lowest MAE value of 0.0102 and RMSE of 0.0218, while the GRU method has a MAE value of 0.0164 and RMSE of 0.0277. So in this study it can be concluded that the BiLSTM method is more accurate and effective than the GRU method in predicting sea level in the South Lampung Waters.

Hydro-informer: a deep learning model for accurate water level and flood predictions

AbstractThis study aims to develop an advanced deep learning model, Hydro-Informer, for accurate water level and flood predictions, emphasizing extreme event forecasting. Utilizing a comprehensive dataset from the Slovak Hydrometeorological Institute SHMI (2008–2020), which includes precipitation, water level, and discharge data, the model was trained using a ladder technique with a custom loss function to enhance focus on extreme values. The architecture integrates Recurrent and Convolutional Neural Networks (RNN, CNN), and Multi-Head Attention layers. Hydro-Informer achieved significant performance, with a Coefficient of Determination (R2) of 0.88, effectively predicting extreme water levels 12 h in advance in a river environment free from human regulation and structures. The model’s strong performance in identifying extreme events highlights its potential for enhancing flood management and disaster preparedness. By integrating with diverse data sources, the model can be used to develop a well-functioning warning system to mitigate flood impacts. This work proposes a novel architecture suitable for locations without water regulation structures.

The generalized STAR modelling with three-dimensional of spatial weight matrix in predicting the Indonesia peatland’s water level

The release rate of CO2 gas can be influenced by peatlands’ physical properties, such as water level and soil moisture, and rainfall. To anticipate the unstable condition which is when the peatland emit more carbon, we developed the Generalized Space Time Autoregressive (GSTAR) model in predicting these physical properties for the following weeks. As the innovation in modelling, the spatial weight matrix was based on three-dimensional coordinates with a modification on the height factor. The data we used are real-time data of water level on the peatlands in Pulang Pisau Regency, Central Kalimantan Province from 20 February 2021 to 18 March 2023. We then used Ordinary Kriging interpolation on the prediction results to create contour maps on different dates. There were empty data on several dates, especially from 24 March until 3 August 2022. To fill the empty data, we used linear interpolation and then we added white noise to the interpolation results. From the data, the water level has a downward trend pattern from around November to September and an upward trend pattern from October to November. Furthermore, we found that the best model for water level was GSTAR (2;0.1) with a modified matrix a=0.1\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$a=0.1$$\\end{document} and b=1.1\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$b=1.1$$\\end{document}. Based on the predicted water level, there is a risk of changes in the properties of the peatlands in several areas in Pulang Pisau Regency.

DAILY DANUBE RIVER WATER LEVEL PREDICTION USING EXTREME LEARNING MACHINE APPROACH

Anticipating water levels in vast riverbeds is crucial for preventing and mitigating floods or droughts, assessing power plant capacity, and facilitating navigation management. This study introduces an innovative water level prediction model utilizing an Extreme Learning Machine developed to solve the issues of low performance of existing forecasting methods. Development of such a system is of extreme importance when talking about the largest European river – the Danube River. Experimental findings reveal the model's satisfactory performance across various accuracy metrics, complexity considerations, and calculation speed. The prediction with the highest error rate based on MAPE criteria was for Prahovo water level prediction over a 365-day period at 2.02%, whilst the most accurate predictions were for Novi Sad and Banatska Palanka over 30 days and 180 days horizons, respectively, at 0.0550%. The highest coefficient of determination (R2) was achieved with the Novi Sad data at 0.9968, whilst the lowest was observed with the Prahovo data at 0.7353. The ELM model achieved high precision by adjusting the activation functions of the hidden layer neurons, which involved using different combinations of sigmoid and radial-basis activation functions.

Exploring Deep Learning Methods for Short-Term Tide Gauge Water Level Predictions

Accurate and timely water level predictions are essential for effective shoreline and coastal ecosystem management. As sea levels rise, the frequency and severity of coastal inundation events are increasing, causing significant societal and economic impacts. Predicting these events with sufficient lead time is essential for decision-makers to mitigate economic losses and protect coastal communities. While machine learning methods have been developed to predict water levels at specific sites, there remains a need for more generalized models that perform well across diverse locations. This study presents a robust deep learning model for predicting water levels at multiple tide gauge locations along the Gulf of Mexico, including the open coast, embayments, and ship channels, all near major ports. The selected architecture, Seq2Seq, achieves significant improvements over the existing literature. It meets the National Oceanic and Atmospheric Administration’s (NOAA) operational criterion, with the percentage of predictions within 15 cm for lead times up to 108 h at the tide gauges of Port Isabel (92.2%) and Rockport (90.4%). These results represent a significant advancement over current models typically failing to meet NOAA’s standard beyond 48 h. This highlights the potential of deep learning models to improve water level predictions, offering crucial support for coastal management and flood mitigation.

Enhancing real-time urban drainage network modeling through Crossformer algorithm and online continual learning

This study presents an innovative approach to real-time modeling of urban drainage networks, leveraging a highly accurate coupled one- and two-dimensional hydrodynamic model to generate a training dataset for node water levels. By employing global states inferred from monitoring points as model inputs, this study overcomes the limitations imposed by the scarcity of monitoring data and the challenge of capturing all node levels. The Crossformer algorithm, which simultaneously accounts for correlation at both temporal and feature scales, is applied to enhance the precision of simultaneous water level predictions across the network. Comparative analysis of different prediction patterns reveals that extending predictions based on a high-accuracy infrastructure offers more benefits than direct modifications to the algorithm's structure. In addition, this paper pioneered the application of online continuous learning concepts to update the prediction model in real time, achieving a balanced integration of measured and simulated data. Consequently, this paper establishes a complete monitoring-predicting-updating real-time simulation system for urban drainage networks.

Total water level prediction at continental scale: Coastal ocean

We demonstrate recent progress made in the simulation of total water level (TWL) at continental scale, using the coastal ocean of US East Coast/Gulf of Mexico coast as an example. A key difference between the continental-scale and small-scale modeling is that the former requires a more accurate vertical datum. Using a geoid-based datum (xGEOID20b), a satellite altimetry product, and a state-of-the-art 3D unstructured-grid model, we significantly improve the accuracy for TWL both near- and off-shore. The average root-mean-square error at all NOAA stations is 14 cm. The non-tidal signals are found to be sensitive to the representation of a large-scale current system near the boundary and extending the domain extent to accommodate this system improves these signals.

A ConvLSTM nearshore water level prediction model with integrated attention mechanism

A ConvLSTM nearshore water level prediction model with integrated attention mechanism

Great lakes basin model based on physical flow and Data-Driven

Abstract As a group of freshwater lakes, the Great Lakes span the United States and Canada and have a great impact on the economies and environments of the regions. To meet the needs of various stakeholders, it is significant to regulate the flow through the control mechanism to achieve the optimal water level balance. Combined Physical Process and Data-Driven, this paper established a dynamic model of the Great Lakes based on the lake dependence relationship and complex climate factors. After judging the optimal water level by combining the needs of all stakeholders, the hydrodynamic model is run to verify and predict the water level, and the flow control algorithm is obtained. The results show that the prediction of optimal water level has a high accuracy rate. After the flow control, the error between the predicted and optimal water levels is greatly reduced.

Using ARIMA and ETS models for forecasting water level changes for sustainable environmental management

It is vital to provide useful hydrological forecasts for urban and agricultural water management, hydropower generation, flood protection and management, drought mitigation and alleviation, and river basin planning and management, among other things. This paper introduces a simple and flexible hydrological time series forecasting framework. Predicting water levels is crucial, given the need for sustainable environmental management. The prognosis should be reasonable to persuade individuals to take proper precautions. While many methods have been developed to predict water levels, here the effectiveness of two approaches to predicting river water levels was assessed. For this purpose, nine years of data were used, which were divided into input data (2014–2021) and validation data (2022), on water levels in the Morava e Binçës river for the Vitia station, in the form of monthly time series records, to identify the best model among those used and to identify the information necessary for water resource management and hazard control. Models Autoregressive Integrated Moving Average(ARIMA) and Error Trend and Seasonality, or Exponential Smoothing (ETS), were examined using the R package to determine the most accurate. The results indicate the applicability of both models, as evidenced by the Root Mean Square Error (RMSE) and the Mean Absolute Error (MAE). The predictive analysis based on historical water levels allowed the identification of distinct periods characterized by high and low water levels between 2022 and 2024, which is important for the area in question due to the numerous flood events occurring here.

Advancing Reservoir Water Level Predictions: Evaluating Conventional, Ensemble and Integrated Swarm Machine Learning Approaches

AbstractAccurate estimation of reservoir water level fluctuation (WLF) is crucial for effective dam operation and environmental management. In this study, seven machine learning (ML) models, including conventional, integrated swarm, and ensemble learning methods, were employed to estimate daily reservoir WLF. The models comprise multi-linear regression (MLR), shallow neural network (SNN), deep neural network (DNN), support vector regression (SVR) integrated with homonuclear molecules optimization (HMO) and particle swarm optimization (PSO) meta-heuristic algorithms, classification and regression tree (CART), and random forest (RF). These models were trained and evaluated using in situ data from three embankment dams in Algeria: the Kramis dam, the Bougous dam, and the Fontaine Gazelles dam. Performance evaluation was conducted using statistical indices, scatter plots, violin plots, and Taylor diagrams. The results revealed superior prediction accuracy for the Fontaine Gazelles dam compared to Kramis and Bougous dams. Particularly, the RF, DNN, and SVR-HMO models exhibited consistent and excellent predictive performance for WLF at the Fontaine Gazelles dam with RMSE values of 0.502 m, 0.536 m, and 0.57 m, respectively. The RF model demonstrates remarkable accuracy across all three case studies. This can be attributed to the ensemble structure of RF, as evidenced by the results. This study underscores the significance of considering factors such as seepage flow intensity in understanding WLF variability. Furthermore, the proposed ML models offer promising capabilities in WLF prediction, highlighting their potential utility in enhancing reservoir management practices and addressing the limitations of traditional regression models. Keys words. Embankment dam, Water level fluctuations, Seepage, Artificial neural network, meta-heuristic algorithm.

Hyperparameter optimization of regional hydrological LSTMs by random search: A case study from Basque Country, Spain

This paper introduces a novel approach for hyperparameter optimization of long short-term memory networks (LSTMs) to achieve highly accurate hourly streamflow and water level predictions in the realm of regional rainfall-runoff modeling. Leveraging simultaneous systematic hyperparameter optimization of 10 distinct hyperparameters by Random Search, the study achieves high accuracy in terms of predictions across 40 humid flashy catchments in Basque Country, north of Spain. By carefully designing the search space and incorporating domain expertise, the approach quickly converges to optimal and highly accurate network configurations with both efficiency and efficacy. LSTMs ingested precipitation, temperature, and potential evapotranspiration as inputs to predict 2 targets of streamflow and water level, in an hourly timestep. On the test set, the optimized LSTM networks accurately predicted streamflow and water level with Nash-Sutcliffe (NSE) and Kling-Gupta (KGE) efficiencies as high as 0.97, in one of the catchments. Across all 40 studied catchments, the overall average NSE and KGE values for streamflow were 0.89 and 0.87, respectively; water level exhibited average NSE and KGE scores of 0.91 and 0.92.Moreover, statistical analysis reveals significant differences in the performance of the 2 distinct optimized network architectures in different hydrological catchments, underscoring the importance of deliberate network configuration selection post-random search. This selection process is vital for achieving higher performance in as many catchments as possible. The findings highlight opportunities for enhancing the “learning maturity” of regional hydrological deep learning LSTM networks. This research provides valuable insights for researchers and practitioners involved in optimizing regional hydrological deep learning models for a variety of applications and on new datasets.

Invasive-Weed-Optimization-Based Extreme Learning Machine for Prediction of Lake Water Level Using Major Atmospheric–Oceanic Climate Scenarios

Fresh water lakes are vulnerable assets that need to be protected against manmade/natural challenges like climate change and anthropogenesis activities. This study addresses the predictability of the lake water level changes based on the knowledge acquired directly from the climate data. Two fresh water lakes named Lake Iznik and Uluabat, located in Turkey, are addressed. Time series of the lake water levels during October 1990–September 2019 at a monthly scale, along with the corresponding anomalies of 24 Large-Scale Atmospheric–Oceanic Oscillations (LSAOOs) from around the globe, are used in the analysis. The relationship between variables and the structure of the models are initially acquired based on the significance of the dependence between climate indices and lake water levels with consideration of the significance of the Spearman rank-order coefficient. Then, the time series are divided into training (80%) and testing (20%) sets. The Extreme Learning Method (ELM), enhanced with the genetic algorithm (ELM-GA) and Invasive Weed Optimization (ELM-IWO), is then used in the predictive models. Based on the results, Lake Uluabat showed a stronger teleconnection with LSAOOs, while the ELM-GA for Lake Iznik and ELM-IWA for Lake Uluabat depicted the best performance in the prediction of lake water levels. Comparison of the enhanced ELM-IWO to the corresponding ELM-GA illustrates that the ELM-IWO reveals more acceptable results owing to its flexible nature.

Analysis of the influence of the daily regulation of power stations on navigable flow conditions at river confluences using the LSTM model

The daily operations of large hydropower stations on rivers induce frequent variations in downstream water levels and flow velocities, resulting in unsteady and complex hydraulic characteristics at the confluence of the main stream and its tributaries, which adversely affect navigation safety. The continuity and momentum equations, along with the RNG k-ϵ turbulence model and the VOF model, were employed to simulate the river flow process at the confluence under unsteady flow conditions. The simulated results for water levels and velocity vector fields are in good agreement with experimental data. Variations in hydraulic characteristics, including water level, flow velocity, and longitudinal gradient at the confluence of the main stream and its tributaries, were analysed. The results indicated that the water level at the confluence decreases when the tributary merges with the main stream. As the confluence ratio increases, fluctuations in the water level at the confluence decrease. However, an increase in the staggered period results in greater fluctuations in the water level within this region. Moreover, a crescent-shaped high-speed flow region is formed at the confluence of the tributary and the main stream. As the unsteady flow discharge of the main stream increases, the relative area of this region correspondingly enlarges, reaching its maximum at peak discharge, and subsequently gradually diminishes as the discharge decreases. Based on these simulation data, a Long Short-Term Memory (LSTM) model was developed to effectively predict water levels and flow velocities, providing a more convenient and accurate method for obtaining real-time information on confluence areas than traditional mathematical and statistical approaches. This study provides novel insights into predicting flow characteristics in confluence areas, thereby offering a basis for formulating navigation safety plans.

Water Level Predictions at Both Entrances of a Sea Strait by Using Machine Learning

Water Level Predictions at Both Entrances of a Sea Strait by Using Machine Learning