Runoff Forecasting Research Articles

An enhanced monthly runoff forecasting using least squares support vector machine based on Harris hawks optimization and secondary decomposition

An enhanced monthly runoff forecasting using least squares support vector machine based on Harris hawks optimization and secondary decomposition

Investigation of the EWT–PSO–SVM Model for Runoff Forecasting in the Karst Area

As the runoff series exhibit nonlinear and nonstationary characteristics, capturing the embedded periodicity and regularity in the runoff series using a single model is challenging. To account for these runoff characteristics and enhance the forecasting precision, this research proposed a new empirical wavelet transform–particle swarm optimization–support vector machine (EWT–PSO–SVM) hybrid model based on “decomposition-forecasting-reconstruction” for runoff forecasting and investigated its effectiveness in the karst area. First, empirical wavelet transform (EWT) was employed to decompose the original runoff series into multiple subseries. Second, the support vector machine (SVM) optimized by particle swarm optimization (PSO) was applied to forecast every signal subseries. Finally, this study summarized the predictions of the subseries to reconstruct the ultimate runoff forecasting. The developed forecasting model was assessed by applying the monthly runoff series of the Chengbi River Karst Basin, and the composite rating index combined with five metrics was adopted as the performance evaluation tool. From the results of this research, it is clear that the EWT–PSO–SVM model outperforms both the PSO–SVM model and the SVM model in terms of the composite rating index, reaching 0.68. Furthermore, verifying the performance stability, the developed model was also compared with PSO–SVM and SVM models under different input data structures. The comparison demonstrated that the hybrid EWT–PSO–SVM model had a robust performance superiority and was an effective model that can be applied to karst area runoff forecasting.

A WRF/WRF-Hydro Coupled Forecasting System with Real-Time Precipitation–Runoff Updating Based on 3Dvar Data Assimilation and Deep Learning

This study established a WRF/WRF-Hydro coupled forecasting system for precipitation–runoff forecasting in the Daqing River basin in northern China. To fully enhance the forecasting skill of the coupled system, real-time updating was performed for both the WRF precipitation forecast and WRF-Hydro forecasted runoff. Three-dimensional variational (3Dvar) multi-source data assimilation was implemented using the WRF model by incorporating hourly weather radar reflectivity and conventional meteorological observations to improve the accuracy of the forecasted precipitation. A deep learning approach, i.e., long short-term memory (LSTM) networks, was adopted to improve the accuracy of the WRF-Hydro forecasted flow. The results showed that hourly data assimilation had a positive impact on the range and trends of the WRF precipitation forecasts. The quality of the WRF precipitation outputs had a significant impact on the performance of WRF-Hydro in forecasting the flow at the catchment outlet. With the runoff driven by precipitation forecasts being updated by 3Dvar data assimilation, the error of flood peak flow was decreased by 3.02–57.42%, the error of flood volume was decreased by 6.34–39.30%, and the Nash efficiency coefficient was increased by 0.15–0.52. The implementation of LSTM can effectively reduce the forecasting errors of the coupled system, particularly those of the time-to-peak and peak flow volumes.

Short-term runoff forecasting in an alpine catchment with a long short-term memory neural network

The governing hydrological processes are expected to shift under climate change in the alpine regions of Switzerland. This raises the need for more adaptive and accurate methods to estimate river flow. In high-altitude catchments influenced by snow and glaciers, short-term flow forecasting is challenging, as the exact mechanisms of transient melting processes are difficult to model mathematically and are poorly understood to this date. Machine learning methods, particularly temporally aware neural networks, have been shown to compare well and often outperform process-based hydrological models on medium and long-range forecasting. In this work, we evaluate a Long Short-Term Memory neural network (LSTM) for short-term prediction (up to three days) of hourly river flow in an alpine headwater catchment (Goms Valley, Switzerland). We compare the model with the regional standard, an existing process-based model (named MINERVE) that is used by local authorities and is calibrated on the study area. We found that the LSTM was more accurate than the process-based model on high flows and better represented the diurnal melting cycles of snow and glacier in the area of interest. It was on par with MINERVE in estimating two flood events: the LSTM captures the dynamics of a precipitation-driven flood well, while underestimating the peak discharge during an event with varying conditions between rain and snow. Finally, we analyzed feature importances and tested the transferability of the trained LSTM on a neighboring catchment showing comparable topographic and hydrological features. The accurate results obtained highlight the applicability and competitiveness of data-driven temporal machine learning models with the existing process-based model in the study area.

A Model for Assessing the Importance of Runoff Forecasts in Periodic Climate on Hydropower Production

Hydropower is the largest source of renewable energy in the world and currently dominates flexible electricity production capacity. However, climate variations remain major challenges for efficient production planning, especially the annual forecasting of periodically variable inflows and their effects on electricity generation. This study presents a model that assesses the impact of forecast quality on the efficiency of hydropower operations. The model uses ensemble forecasting and stepwise linear optimisation combined with receding horizon control to simulate runoff and the operation of a cascading hydropower system. In the first application, the model framework is applied to the Dalälven River basin in Sweden. The efficiency of hydropower operations is found to depend significantly on the linkage between the representative biannual hydrologic regime and the regime actually realised in a future scenario. The forecasting error decreases when considering periodic hydroclimate fluctuations, such as the dry–wet year variability evident in the runoff in the Dalälven River, which ultimately increases production efficiency by approximately 2% (at its largest), as is shown in scenarios 1 and 2. The corresponding potential hydropower production is found to vary by 80 GWh/year. The reduction in forecasting error when considering biennial periodicity corresponds to a production efficiency improvement of about 0.33% (or 13.2 GWh/year).

Runoff prediction using hydro-meteorological variables and a new hybrid ANFIS-GPR model

Abstract Precise and credible runoff forecasting is extraordinarily vital for various activities of water resources deployment and implementation. The neoteric contribution of the current article is to develop a hybrid model (ANFIS-GPR) based on adaptive neuro-fuzzy inference system (ANFIS) and Gaussian process regression (GPR) for monthly runoff forecasting in the Beiru river of China, and the optimal input schemes of the models are discussed in detail. Firstly, variables related to runoff are selected from the precipitation, soil moisture content, and evaporation as the first set of input schemes according to correlation analysis (CA). Secondly, principal component analysis (PCA) is used to eliminate the redundant information between the original input variables for forming the second set of input schemes. Finally, the runoff is predicted based on different input schemes and different models, and the prediction performance is compared comprehensively. The results show that the input schemes jointly established by CA and PCA (CA-PCA) can greatly improve the prediction accuracy. ANFIS-GPR displays the best forecasting performance among all the peer models. In the single models, the performance of GPR is better than that of ANFIS.

Monthly streamflow forecasting by machine learning methods using dynamic weather prediction model outputs over Iran

Seasonal hydrological forecasts play a critical role in water resources management. The Copernicus Climate Change Service (C3S) data store provides open access to monthly hydrological forecasts for up to six-months. This study aims to evaluate, for the first time, 1- to 3-month runoff forecasts using the European Centre for Medium-Range Weather Forecasts (ECMWF) ensembles of precipitation, runoff, and temperature in 1981–2015 period over a total of 30 s-level basins in Iran. We adopted the 5th, 50th and 95th ECMWF ensemble quantiles for each variable that represent low, medium and high probability of occurrence, respectively. Pearson correlation analysis (Pca), Recursive Feature Elimination (RFE) via random forest (RF) model, and Bayesian Networks (BN) feature selection algorithms were used in order to reduce input variable dimension and select potential predictors to be fed to the machine learning models. Multiple Linear Regression (MLR), Artificial Neural Networks (ANN), Support Vector Regression (SVR), Random Forest (RF), and eXtreme Gradient Boosting (XGBoost) machine learning models were used with Repeated K-Fold cross validation (rK-Fold CV) while model efficiency was evaluated using modified Kling-Gupta efficiency coefficient (KGE’), Nash-Sutcliffe Efficiency coefficient (NSE), and Normalized Root Mean Square Error (NRMSE). Results of this study revealed that C3S runoff ensembles have the highest impact on forecast accuracy of streamflow, followed by precipitation and temperature. Overall, model performance yield a best-to-worst ranking of ANN, XGBoost, RF, MLR, and SVR with KGE’ values of 0.70, 0.68, 0.66, 0.57, and 0.41, respectively. The predictive performance of all models decreased with lead times beyond 1-month, where ANN and XGBoost outperformed other models with KGE’ of 0.65 for 2-month lead time and 0.60 for 3-month lead time. The three superior models of XGBoost, ANN, and RF, were employed with RFE and BN FSAs most frequently across Iran’s 30 s level basins in all lead times. Almost all models in the arid central region of Iran showed the lowest performance while highest skills were achieved in the western regions of Iran. Finally, for all models and over all regions, the model performance reduced by increase in lead-time.

Improved monthly runoff time series prediction using the SOA–SVM model based on ICEEMDAN–WD decomposition

Abstract In runoff prediction, the prediction accuracy is often affected by the non-linear and non-stationary characteristics of the runoff series. In this study, a coupled forecasting model is proposed that decomposes the original runoff series by an improved complete ensemble Empirical Mode Decomposition (EMD) (ICEEMDAN) combined with a wavelet decomposition (WD) and then forecasts the monthly runoff using a support vector machine (SVM) optimized by the seagull optimization algorithm (SOA). In this method, a series of Intrinsic Mode Function (IMF) and a Residual (Res) are obtained by decomposing the original runoff series with ICEEMDAN. The WD method is used to perform quadratic decomposition of high-frequency components decomposed by the ICEEMDAN method to make the runoff series as smooth as possible. Then the decomposed components are input into the SOA-SVM model for prediction. Finally, the prediction results of each component are superimposed and reconstructed to obtain the final monthly runoff prediction results. RMSE, Mean Absolute Percentage Error (MAPE), Nash-Sutcliffe Efficiency Coefficient (NSEC), and R are selected to evaluate the prediction results and the model is compared with SOA-SVM model, EMD-SOA-SVM model and CEEMDAN-SOA-SVM model other models. The proposed model is applied to the monthly runoff forecast of the Hongjiadu and Manwan Reservoirs. When compared with other benchmarking models, the ICEEMDAN-WD-SOA-SVM model attains the smallest Root Mean Square Error (RMSE) and MAPE and the largest NSEC and R. The ICEEMDAN-WD-SOA-SVM model has the best prediction effect, the highest prediction accuracy, and the lowest prediction error.

A Stacking Ensemble Model of Various Machine Learning Models for Daily Runoff Forecasting

Improving the accuracy and stability of daily runoff prediction is crucial for effective water resource management and flood control. This study proposed a novel stacking ensemble learning model based on attention mechanism for the daily runoff prediction. The proposed model has a two-layer structure with the base model and the meta model. Three machine learning models, namely random forest (RF), adaptive boosting (AdaBoost), and extreme gradient boosting (XGB) are used as the base models. The attention mechanism is used as the meta model to integrate the output of the base model to obtain predictions. The proposed model is applied to predict the daily inflow to Fuchun River Reservoir in the Qiantang River basin. The results show that the proposed model outperforms the base models and other ensemble models in terms of prediction accuracy. Compared with the XGB and weighted averaging ensemble (WAE) models, the proposed model has a 10.22% and 8.54% increase in Nash–Sutcliffe efficiency (NSE), an 18.52% and 16.38% reduction in root mean square error (RMSE), a 28.17% and 18.66% reduction in mean absolute error (MAE), and a 4.54% and 4.19% increase in correlation coefficient (r). The proposed model significantly outperforms the base model and simple stacking model indicated by both the Friedman test and the Nemenyi test. Thus, the proposed model can produce reasonable and accurate prediction of the reservoir inflow, which is of great strategic significance and application value in formulating the rational allocation and optimal operation of water resources and improving the breadth and depth of hydrological forecasting integrated services.

Model and application of annual river runoff prediction based on complementary set empirical mode decomposition combined with particle swarm optimization adaptive neuro-fuzzy system

Abstract Runoff is affected by natural and nonnatural factors in the process of formation, and the runoff series is generally nonstationary time series. How to improve the accuracy of runoff prediction has always been a difficult problem for hydrologists. The key to solve this problem is to reduce the complexity of runoff series and improve the accuracy of runoff prediction model. Based on the aforementioned ideas, this article uses the complementary set empirical mode decomposition to decompose the runoff series into multiple intrinsic components that retain time–frequency information, thus reducing the complexity of the runoff series. The particle swarm optimization (PSO) adaptive neuro-fuzzy system is used to predict each intrinsic component to improve the accuracy of runoff prediction. After that, the trained intrinsic components of the model are reconstructed into the original runoff series. The example shows that the absolute relative error of the runoff forecasting model constructed in this article is 0.039, and the determination coefficient is 0.973. This model can be applied to the annual runoff series forecasting. Comparing the prediction results of this model with empirical mode decomposition algorithm-ANFIS model and ANFIS model, complementary set empirical mode decomposition algorithm-PSO-ANFIS model shows obvious advantages.

Application of a New Hybrid Deep Learning Model That Considers Temporal and Feature Dependencies in Rainfall–Runoff Simulation

Runoff forecasting is important for water resource management. Although deep learning models have substantially improved the accuracy of runoff prediction, the temporal and feature dependencies between rainfall–runoff time series elements have not been effectively exploited. In this work, we propose a new hybrid deep learning model to predict hourly streamflow: SA-CNN-LSTM (self-attention, convolutional neural network, and long short-term memory network). The advantages of CNN and LSTM in terms of data extraction from time series data are combined with the self-attention mechanism. By considering interdependences of the rainfall–runoff sequence between timesteps and between features, the prediction performance of the model is enhanced. We explored the performance of the model in the Mazhou Basin, China; we compared its performance with the performances of LSTM, CNN, ANN (artificial neural network), RF (random forest), SA-LSTM, and SA-CNN. Our analysis demonstrated that SA-CNN-LSTM demonstrated robust prediction with different flood magnitudes and different lead times; it was particularly effective within lead times of 1–5 h. Additionally, the performance of the self-attention mechanism with LSTM and CNN alone, respectively, was improved at some lead times; however, the overall performance was unstable. In contrast, the hybrid model integrating CNN, LSTM, and the self-attention mechanism exhibited better model performance and robustness. Overall, this study considers the importance of temporal and feature dependencies in hourly runoff prediction, then proposes a hybrid deep learning model to improve the performances of conventional models in runoff prediction.

A quantile-based encoder-decoder framework for multi-step ahead runoff forecasting

Deep neural network (DNN) models have become increasingly popular in the hydrology community. However, most studies are related to (rainfall-) runoff simulation and comparatively fewer studies have focused on runoff forecasting. In this study, quantile-based (q = 0.05, 0.5, 0.95) encoder-decoder (ED) models that use long short-term memory network (LSTM) and dense network (DN) blocks were developed for three and five days ahead runoff forecasting. Through linear (LW) and non-linear (NLW) wavelet selection, hybrid models LSTM-DN, LSTM-DN-LW, LSTM-DN-NLW, ED, ED-LW, and ED-NLW were developed. For each lead time (LT = 3, 5) and value of q, different model configurations were created using different input lag lengths (IL = 15, 45, 180). The developed models were tested for runoff forecasting using three basins (with different characteristics) from the Catchment Attributes and MEteorology for Large-sample Studies (CAMELS) dataset. The models were compared using deterministic (e.g., the Kling-Gupta efficiency [KGE] metric) and probabilistic (e.g., reliability) statistical metrics. While the models showed high variability in performance across the three basins (KGE = 0.308–0.979 for the q = 0.5 models), very high accuracy (up to KGE = 0.979) was achieved for one of the basins with high snowmelt. The ED-NLW model was found to generally outperform the other models. Although the LSTM-DN model had the highest median KGE (0.434 across all configurations), the ED and ED-NLW models had higher reliability than LSTM-DN (90% and 91%, respectively, considering a 90% confidence level). Models coupled with NLW performed superior to those that used LW. All ED models had high reliability despite two of the basins achieving median KGE values of ∼ 0.390, highlighting that quantile-based models can generate reliable forecast intervals even when the KGE of the median forecast (q = 0.5) is low. An additional experiment generated synthetic precipitation forecasts with varying degrees of accuracy. The models were trained using accurate precipitation forecasts and tested using both accurate and inaccurate precipitation forecasts. While up to a 120% improvement in KGE was found when accurate precipitation forecasts were used as input to the models, using inaccurate precipitation forecasts resulted in a substantial decrease in reliability. Overall, the results of this study can serve as a benchmark for future studies developing probabilistic DNN models for runoff forecasting.

Flood Risk Quantification, Transmission, and Propagation Analysis for Flood Water Utilization of Parallel Reservoirs

Joint operation of the reservoir system can greatly improve flood water utilization and alleviate the contradiction between water supply and water demand. But yet, the flood risk brought by forecast uncertainties transfers between the reservoirs and propagates from upstream to downstream, which is crucial for flood water utilization but is rarely investigated. To address this issue, this study proposes a FLWL optimal control framework, which consists of three modules: the FLWL upper bound derivation module, the flood risk analysis module, and the optimal FLWL decision-making module. The first module is to derive the upper bound relationship of the flood limited water level (FLWL), i.e., the specific substitution relationship of storage, for the parallel reservoirs coupling the capacity-constrained pre-release method and the aggregation-decomposition method. In the second module, the method to calculate the flood risk of the reservoirs and downstream caused by rainfall and runoff forecast uncertainties based on the total probability formula is proposed. In the third module, the transmission and propagation of flood risks in the parallel reservoir system under different rainfall grades are quantified, based on which the optimal FLWLs are determined. The implication in the Qinghe-Chaihe parallel reservoir system located in Northeast China shows that FLWL in Qinghe Reservoir and Chaihe Reservoir can be raised to 129.5 m from 127 m and 108 m from 104 m, respectively. Controlling the FLWL at any point in the safety domain can improve both flood water utilization and flood control benefits. However, the flood risk nonlinearly transfers between the parallel reservoirs and nonlinearly propagates from the upstream to the downstream. Based on this relationship, the optimal FLWLs of the reservoirs under different decision preferences are recommended.

Possibilities of Using the COSMO-Ru System for Short-term Forecasting of Runoff of Russian Rivers

Possibilities of Using the COSMO-Ru System for Short-term Forecasting of Runoff of Russian Rivers

A novel feature attention mechanism for improving the accuracy and robustness of runoff forecasting

The interaction between hydrological factors is complex and the correlation effects cannot be quantitatively explained from a mechanistic perspective. The extraction of effective features from several hydrological and meteorological data and quantifying their correlation effects to make runoff prediction more accurate and stable is an urgent problem to be solved. In this study, we introduce a structural paradigm for hydrological time series to handle the entire feature space composed of some feature sequences based on deep learning methods, namely, the feature attention mechanism. This method transforms the problem of feature-target association into multiple parallel binary classification problems and assigns attention units to each specific feature in the network. The attention distribution was adjusted by updating the neural network parameters in a supervised manner to generate feature weightings. In addition, two extensions were proposed based on the initial network structure. The three methods were used in a practical study of the upper Yangtze River Basin using three evaluation metrics compared with five benchmark methods. The mean metric values were improved by up to 9.43, 10.05, and 2.65%, respectively. Meanwhile, with the characteristics of hydrological data, the rationality of the method is corroborated in hydrological feature discovery and time-series dependence extraction. Moreover, we tested the performance of the model on two constructed noise datasets to demonstrate its robustness. The results of all the above experiments show that the attention module significantly improves the learning and generalization ability, enhances noise resistance, and strengthens the robustness of the model compared with the traditional method.

Runoff prediction of lower Yellow River based on CEEMDAN–LSSVM–GM(1,1) model

Accurate medium and long-term runoff forecasts play a vital role in guiding the rational exploitation of water resources and improving the overall efficiency of water resources use. Machine learning is becoming a common trend in time series forecasting research. Least squares support vector machine (LSSVM) and grey model (GM(1,1)) have received much attention in predicting rainfall and runoff in the last two years. “Decomposition-forecasting” has become one of the most important methods for forecasting time series data. Complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) decomposition method has powerful advantages in dealing with nonlinear data. Least squares support vector machine (LSSVM) has strong nonlinear fitting ability and good robustness. Gray model (GM(1,1)) can solve the problems of little historical data and low serial integrity and reliability. Based on their respective advantages, a combined CEEMDAN–LSSVM–GM(1,1) model was developed and applied to the runoff prediction of the lower Yellow River. To verify the reliability of the model, the prediction results were compared with the single LSSVM model, the CEEMDAN–LSSVM model and the CEEMDAN–support vector machines (SVM)–GM(1,1). The results show that the combined CEEMDAN–LSSVM–GM(1,1) model has a high accuracy and the prediction results are better than other models, which provides an effective prediction method for regional medium and long-term runoff prediction and has good application prospects.

Runoff Forecast Model Based on an EEMD-ANN and Meteorological Factors Using a Multicore Parallel Algorithm

Runoff Forecast Model Based on an EEMD-ANN and Meteorological Factors Using a Multicore Parallel Algorithm

An attention-based LSTM model for long-term runoff forecasting and factor recognition

With advances in artificial intelligence, machine learning-based models such as long short-term memory (LSTM) models have shown much promise in forecasting long-term runoff by mapping pathways between large-scale climate patterns and catchment runoff responses without considering physical processes. The recognition of key factors plays a vital role and thus affects the performance of the model. However, there is no conclusion on which recognition algorithm is the most suitable. To address this issue, an LSTM model combined with two attention mechanisms both in the input and hidden layers, namely AT-LSTM, is proposed for long-term runoff forecasting at Yichang and Pingshan stations in China. The added attention mechanisms automatically assign weights to 130 climate phenomenon indexes, avoiding the use of subjectively set recognition algorithms. Results show that the AT-LSTM model outperforms the Pearson’s correlation based LSTM model in terms of four evaluation metrics for monthly runoff forecasting. Further, the set indirect runoff prediction method verifies that the AT-LSTM model also performs effectively in precipitation and potential evapotranspiration forecasting, and the indirect runoff prediction is inferior to the AT-LSTM model to establish a direct link between climate factors and runoff. Finally, four key factors related to runoff are identified by the attention mechanism and their impacts on runoff are analyzed on intra- and inter-annual scales. The proposed AT-LSTM model can effectively improve the accuracy of long-term forecasting and identify the dynamic influence of input factors.

Optimal dispatching rules of hydropower reservoir in flood season considering flood resources utilization: A case study of Three Gorges Reservoir in China

Making full use of flood resources on the basis of ensuring flood control safety plays a positive role in improving the comprehensive benefit of reservoirs during flood season. To improve the utilization rate of flood resources of hydropower reservoirs, taking the Three Gorges Reservoir (TGR) as an example, an optimal forecast dispatching model for flood resources utilization (FRU) in flood season is established first based on the safety requirements of flood control, combining the power demand and the safe operating conditions of the power grid. Then, considering the uncertainty of runoff forecast, the stochastic simulation method of inflow forecast based on the gaussian mixture model (GMM) is used to assess the flood control risk of FRU dispatching. Finally, the optimal dispatching rules of TGR in flood season considering FRU are derived by using the gradient boosting regression tree (GRBT). The experimental results show that the optimal dispatching rules can effectively improve the utilization rate of flood resources and increase the generation benefit of TGR during flood season without an additional increase in flood control risk. Compared with routine dispatching, the hydropower generation of the TGR by rule dispatching considering inflow forecast information of 3, 5, and 7 days can be increased by 3.38%, 5.57%, and 6.93% respectively. This shows that the dispatching rules derived are very reliable and promising, which can provide decision support for reservoirs operation in flood season.

Runoff Forecasting using Convolutional Neural Networks and optimized Bi-directional Long Short-term Memory

Runoff Forecasting using Convolutional Neural Networks and optimized Bi-directional Long Short-term Memory