Monthly Runoff Series Research Articles

Runoff Prediction in Different Forecast Periods via a Hybrid Machine Learning Model for Ganjiang River Basin, China

Accurate forecasting of monthly runoff is essential for efficient management, allocation, and utilization of water resources. To improve the prediction accuracy of monthly runoff, the long and short memory neural networks (LSTM) coupled with variational mode decomposition (VMD) and principal component analysis (PCA), namely VMD-PCA-LSTM, was developed and applied at the Waizhou station in the Ganjiang River Basin. The process begins with identifying the main forecasting factors from 130 atmospheric circulation indexes using the PCA method and extracting the stationary components from the original monthly runoff series using the VMD method. Then, the correlation coefficient method is used to determine the lag of the above factors. Lastly, the monthly runoff is simulated by combining the stationary components and key forecasting factors via the LSTM model. Results show that the VMD-PCA-LSTM model effectively addresses the issue of low prediction accuracy at high flows caused by a limited number of samples. Compared to the single LSTM and VMD-LSTM models, this comprehensive approach significantly enhances the model’s predictive accuracy, particularly during the flood season.

Monthly runoff prediction based on a coupled VMD-SSA-BiLSTM model

The accurate prediction of monthly runoff in the lower reaches of the Yellow River is crucial for the rational utilization of regional water resources, optimal allocation, and flood prevention. This study proposes a VMD-SSA-BiLSTM coupled model for monthly runoff volume prediction, which combines the advantages of Variational Modal Decomposition (VMD) for signal decomposition and preprocessing, Sparrow Search Algorithm (SSA) for BiLSTM model parameter optimization, and Bi-directional Long and Short-Term Memory Neural Network (BiLSTM) for exploiting the bi-directional linkage and advanced characteristics of the runoff process. The proposed model was applied to predict monthly runoff at GaoCun hydrological station in the lower Yellow River. The results demonstrate that the VMD-SSA-BiLSTM model outperforms both the BiLSTM model and the VMD-BiLSTM model in terms of prediction accuracy during both the training and validation periods. The Root-mean-square deviation of VMD-SSA-BiLSTM model is 30.6601, which is 242.5124 and 39.9835 lower compared to the BiLSTM model and the VMD-BiLSTM model respectively; the mean absolute percentage error is 5.6832%, which is 35.5937% and 6.3856% lower compared to the other two models, respectively; the mean absolute error was 19.8992, which decreased by 136.7288 and 25.7274 respectively; the square of the correlation coefficient (R2) is 0.93775, which increases by 0.53059 and 0.14739 respectively; the Nash–Sutcliffe efficiency coefficient was 0.9886, which increased by 0.4994 and 0.1122 respectively. In conclusion, the proposed VMD-SSA-BiLSTM model, utilizing the sparrow search algorithm and bidirectional long and short-term memory neural network, enhances the smoothness of the monthly runoff series and improves the accuracy of point predictions. This model holds promise for the effective prediction of monthly runoff in the lower Yellow River.

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.

An enhanced monthly runoff time series prediction using extreme learning machine optimized by salp swarm algorithm based on time varying filtering based empirical mode decomposition

Reliable runoff prediction plays a significant role in reservoir scheduling, water resources management, and efficient utilization of water resources. To effectively enhance the prediction accuracy of monthly runoff series, a hybrid prediction model (TVF-EMD-SSA-ELM) combining time varying filtering (TVF) based empirical mode decomposition (EMD), salp swarm algorithm (SSA) and extreme learning machine (ELM) is proposed. Firstly, the monthly runoff series is decomposed into several sub-series using TVF-EMD. Secondly, SSA is used to optimize the input weights and hidden layer biases of the selected ELM model. Finally, the prediction results are generated by summing and reconstructing each sub-series based on the SSA optimized ELM model. This hybrid model is applied to the monthly runoff prediction of Manwan hydropower, Hongjiadu hydropower, and Yingluoxia hydrological station, and compared with back propagation (BP), ELM, SSA-ELM, PSO-ELM, GSA-ELM, TVF-EMD-ELM, EMD-SSA-ELM, extreme-point symmetric mode decomposition (ESMD)-SSA-ELM, wavelet decomposition (WD)-SSA-ELM and complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN)-SSA-ELM models. The prediction performance of various models is reflected by four evaluation indicators (R, NSEC, NRMSE, MAPE). Results reveal that the prediction effect of the ELM model is better than that of BP, the optimization accuracy of SSA is better than those of particle swarm optimization (PSO) and gravitational search algorithm (GSA), and the prediction accuracy of the hybrid TVF-EMD and SSA is better than that of only TVF-EMD or SSA. TVF-EMD-SSA-ELM model has the highest prediction accuracy. When compared with the single ELM model, it’s NRMSE and MAPE at Manwan hydropower decrease by 84.4% and 72.38%, those of Hongjiadu hydropower decrease by 85.21% and 78.38%, and those of Yingluoxi hydrological station decrease by 68.42% and 39.51%, respectively. R and NSEC of the three sites are close to 1. Therefore, the proposed model provides a new method for the prediction of monthly runoff, and the results can provide a reference for the prediction of monthly runoff in the study area.

A Compound Approach for Monthly Runoff Forecasting Based on Multiscale Analysis and Deep Network with Sequential Structure

Accurate runoff forecasting is of great significance for the optimization of water resource management and regulation. Given such a challenge, a novel compound approach combining time-varying filtering-based empirical mode decomposition (TVFEMD), sample entropy (SE)-based subseries recombination, and the newly developed deep sequential structure incorporating convolutional neural network (CNN) into a gated recurrent unit network (GRU) is proposed for monthly runoff forecasting. Firstly, the runoff series is disintegrated into a collection of subseries adopting TVFEMD, considering the volatility of runoff series caused by complex environmental and human factors. The subseries recombination strategy based on SE and recombination criterion is employed to reconstruct the subseries possessing the approximate complexity. Subsequently, the newly developed deep sequential structure based on CNN and GRU (CNNGRU) is applied to predict all the preprocessed subseries. Eventually, the predicted values obtained above are aggregated to deduce the ultimate prediction results. To testify to the efficiency and effectiveness of the proposed approach, eight relevant contrastive models were applied to the monthly runoff series collected from Baishan reservoir, where the experimental results demonstrated that the evaluation metrics obtained by the proposed model achieved an average index decrease of 44.35% compared with all the contrast models.

A Hybrid Model Based on Variational Mode Decomposition and Gradient Boosting Regression Tree for Monthly Runoff Forecasting

Accurate and reliable monthly runoff forecasting is of great significance for water resource optimization and management. A neoteric hybrid model based on variational mode decomposition (VMD) and gradient boosting regression (GBRT) called VMD-GBRT was proposed and applied for monthly runoff forecasting. VMD was first employed to decompose the original monthly runoff series into several intrinsic mode functions (IMFs). The optimal number of input variables were then chosen according to the autocorrelation function (ACF) and the partial autocorrelation function (PACF). The trained GBRT model was used as a forecasting instrument to predict the testing set of each normalized subsequence. The ensemble forecasting result was finally generated by aggregating the prediction results of all subsequences. The proposed hybrid model was evaluated using an original monthly runoff series, from 1/1969 to 12/2018, measured at the Huaxian, Lintong and Xianyang hydrological stations in the Wei River Basin (WRB), China. The EEMD-GBRT, the single GBRT, and the single SVM were adopted as comparative forecast models using the same dataset. The results indicated that the VMD-GBRT model exhibited the best forecasting performance among all the peer models in terms of the coefficient of determination (R2 = 0.8840), mean absolute percentage error (MAPE = 19.7451), and normalized root-mean-square error (NRMSE = 0.3468) at Huaxian station. Furthermore, the model forecasting results applied at Lintong and Xianyang stations were consistent with those at Huaxian station. This result further verified the accuracy and stability of the VMD-GBRT model. Thus, the proposed VMD-GBRT model was effective method for forecasting non-stationary and non-linear runoff series, and can be recommended as a promising model for monthly runoff forecasting.

Water Supply Risk Analysis Based on Runoff Sequence Simulation with Change Point under Changing Environment

This study investigates the water supply risk of Panjiakou reservoir in the Luanhe River basin of China during 1956–2016 under environmental change. Since the monthly runoff series during 1956–2016 is a time series with change points, it is necessary to find a new stochastic streamflow series generation approach to preserve the statistical characteristics of the original series and to refine the reliability of water supply risk analysis. This paper improves a known stochastic streamflow simulation method of previous research to better reflect the characteristics of series with change points. And this paper simulates the monthly runoff series with change point of Panjiakou reservoir during 1956–2016 by three different methods, including Thomas–Fiering model, copula function stochastic simulation method, and copula function stochastic simulation method with the mixed distribution model. Among the three methods, the copula function stochastic simulation method with the mixed distribution model which is improved on the basis of copula function stochastic simulation method in this study performs best in simulating the observed monthly runoff series during 1956–2016, and the water supply risk indices including reliability (time-based and volume-based), resilience, vulnerability, drought risk index (DRI), and sustainability index (SUI) are evaluated for Panjiakou reservoir and analyzed by using the stochastic simulation results. By comparing with the previous studies, all indicators are between the corresponding results of 1956–1979 and 1980–2016 with stationary inflows; it can be seen that change point seriously affects the water supply risk of Panjiakou reservoir. These results make it easy to formulate water supply strategies and schemes in changing environment for water resources managers.

Combination of Multiple Data-Driven Models for Long-Term Monthly Runoff Predictions Based on Bayesian Model Averaging

Accurate and reliable long-term runoff forecasting is very important for water resource system planning and management. This study utilized three data-driven models to simulate and forecast the monthly runoff series of the Huangzhuang hydrological station from 1981 to 2017. To improve the accuracy and reduce the uncertainty, two model averaging techniques were applied to merge forecast results of the different models, and 90% confidence intervals were derived using Monte Carlo sampling. Several indices were used to evaluate the results of three data-driven models and two model averaging techniques. Among the many discoveries in this paper, the following stand out: (i) in general, the random forest (RF) algorithm presented nearly the same accuracy as did the artificial neural network (ANN) algorithm, and both were superior to the support vector machine (SVM) method; however, none of the models consistently provided the best result in all months; (ii) the comparison of the deterministic results indicated that Copula-Bayesian model averaging (BMA) exhibited smaller errors than did BMA, especially for the points whose uniform quantiles ranged within (0.125, 0.35) and (0.5, 0.625); and (iii) in most cases, the 90% confidence interval of the Copula-BMA scheme had higher containing ratio values, smaller average relative bandwidth values in the high-flow months, and smaller average relative deviation amplitudes than did BMA.

Regional monthly runoff forecast in southern Canada using ANN, K-means, and L-moments techniques

River runoff forecasting is necessary for numerous applications related to water use, including water supply management, power generation and flooding protection measures. In this study, a regional model using an artificial neural network (ANN) is proposed for monthly runoff forecasting, which considers stations linked to the network that belong to the same homogeneous region, and are delimited using K-means (KM-ANN) and L-moments (LM-ANN) techniques. This methodology was applied to a sample of 90 monthly runoff series in southern Canada. The results were compared to those of a traditional neural network for a given site (ANNs) using statistical indices, such as root-mean-squared error (RMSE), relative square error (RSE), mean absolute error (MAE), relative absolute error (RAE), the concordance index (d) and the coefficient of determination (r2). The LM-ANN technique produced better forecasts in 56.7% of the analysed stations, whereas the KM-ANN and ANN techniques produced better forecasts in 27.7% and 15.6% of the stations, respectively. Thus, the results indicate that the regionalisation process improved the forecasts in 84.4% of the studied cases, and the estimation uncertainty was reduced by an average of 31.8%, according to the RMSE, RSE, MAE and RAE values. Therefore, its application is recommended in Canada, where it would be useful for the Integrated Water Resources Management Program.

Variations of Runoff and Sediment Load in the Middle and Lower Reaches of the Yangtze River, China (1950-2013).

On the basis of monthly runoff series obtained in 1950–2013 and annual sediment load measured in 1956–-2013 at five key hydrological stations in the middle and lower reaches of the Yangtze River basin, this study used the Mann-Kendall methods to identify trend and abrupt changes of runoff and sediment load in relation to human activities. The results were as follows: (1) The annual and flood season runoffs showed significant decreasing trends at Yichang station, and showed slight downward trends at Hankou and Datong stations, while the abrupt changes of dry season runoff at Yichang, Hankou and Datong stations occurred in about 2007 and the change points were followed by significant increasing trends. The construction of the Three Gorges Dam, which began to operate in 2003, influenced the variations of runoff in the mainstream of Yangtze River, but the effect weakened with the distance along the downstream direction from TGD. (2) Since the 1990s, annual sediment loads at Yichang, Hankou, and Datong stations have been decreasing significantly, and after 2002, the annual sediment load at Yichang dropped below that of Hankou and Datong. The dams and deforestation/forestation contributed to the significant decreasing trend of the sediment load. In addition, the Three Gorges Dam aggravated the downward trend and caused the erosion of the riverbed and riverbanks in the middle and lower reaches. (3) The runoff and sediment load flowing from Dongting Lake into the mainstream of the Yangtze River showed significant decreasing trends at Chenglingji station after 1970s, and in contrast, slight increase in the sediment flow from Poyang Lake to the mainstream of the Yangtze River at Hukou station were detected over the post-TGD period (2003–2013). The result of the study will be an important foundation for watershed sustainable development of the Yangtze River under the human activities.

Generation of Natural Runoff Monthly Series at Ungauged Sites Using a Regional Regressive Model

Many hydrologic applications require reliable estimates of runoff in river basins to face the widespread lack of data, both in time and in space. A regional method for the reconstruction of monthly runoff series is here developed and applied to Sicily (Italy). A simple modeling structure is adopted, consisting of a regression-based rainfall–runoff model with four model parameters, calibrated through a two-step procedure. Monthly runoff estimates are based on precipitation, temperature, and exploiting the autocorrelation with runoff at the previous month. Model parameters are assessed by specific regional equations as a function of easily measurable physical and climate basin descriptors. The first calibration step is aimed at the identification of a set of parameters optimizing model performances at the level of single basin. Such “optimal” sets are used at the second step, part of a regional regression analysis, to establish the regional equations for model parameters assessment as a function of basin attributes. All the gauged watersheds across the region have been analyzed, selecting 53 basins for model calibration and using the other six basins exclusively for validation. Performances, quantitatively evaluated by different statistical indexes, demonstrate relevant model ability in reproducing the observed hydrological time-series at both the monthly and coarser time resolutions. The methodology, which is easily transferable to other arid and semi-arid areas, provides a reliable tool for filling/reconstructing runoff time series at any gauged or ungauged basin of a region.

Multiscaling Analysis of Monthly Runoff Series Using Improved MF-DFA Approach

An improved multifractal detrended fluctuation analysis(MF-DFA) method is applied to analyze the long-term monthly runoff records of a hydrological station in the Yangtze River with seasonal trend eliminated, through which the long-range correlation and the multifractal characteristics have been found. The multifractal spectrum has been fitted by a generalized expression of the multiplicative cascade model, and the results show that the monthly runoff series has strong multifractal characteristics. Comparing the results for the original runoff series with those of shuffled and surrogate series, it concludes that the multifractal characteristics of the monthly runoff time series is due to the broadness of both the probability density function and long-range correlation, and the broadness of the probability density function is dominant.

Wavelet ANN Based Monthly Runoff Forecast

A wavelet artificial neural network to forecasting monthly runoff is proposed. The monthly runoff series is firstly decomposed to sub-series on different time scales, and each sub-series is modeled. The weights of the network are replaced by wavelet functions and are corrected by conjugate gradient method in the training iteration. Then the proposed network is trained with 49 years (1952-2000) actual data of one hydro power plant of Jiangxi province and is tested for target year (2001-2003). Finally, some actual results for mid and long term water inflow forecasting are obtained and which show the proposed method has a good precision for forecasting.

Analysis of Chaotic Characters for the Monthly Runoff Se-ries at Fudedian Station in Liaohe Bain

The evolution of monthly runoff is affected both by climate environment and human activities, and its characteristics play an important role in runoff prediction and simulation. In this paper, the G-P and the principal component analysis method, which are both based on the reconstruction theory of the phase space, are used to study the chaos characteristics of the monthly runoff series at Fudedian station in Liaohe basin. The results show that the monthly runoff series have a large probability of chaos.

Rainfall–runoff modeling using least squares support vector machines

Over the past decade, artificial neural networks (ANN) have been widely used in the runoff modeling studies. In spite of a number of advantages, ANN models have some drawbacks, including the possibility of getting trapped in local minima, over training, subjectivity in the determining of model parameters, initialization of the weights in each simulation randomly, and the components of its complex structure. In the past decade, a new alternative kernel‐based technique called a support vector machine (SVM) has been found to be popular in modeling studies because of its advantages over ANN. Least squares version of support vector machines (LS‐SVM) provides a computational advantage over standard support vector machines by converting quadratic optimization problem into a system of linear equations. The LS‐SVM method is preferred in this study. The main purposes of this study are to examine the applicability and capability of LS‐SVM for the prediction of runoff values of Tahtali and Gordes watersheds, which are the major surface water resources for the city of Izmir in Turkey, and to compare its performance with ANN and other traditional techniques such as autoregressive moving average and multiple linear regression models. For these purposes, meteorological data (rainfall and temperature) and lagged data of runoff were used in modeling applications. Some favorite statistical performance evaluation measures were used to assess models. The results in study indicate that the LS‐SVM and ANN methods are successful tools to model the monthly runoff series of two study regions and can give better prediction performances than conventional statistical models. Although these two methods are powerful artificial intelligence techniques, LS‐SVM makes the running time considerably faster with the same or higher accuracy. In terms of accuracy, the LS‐SVM models, which involve different normalization types, resulted in increased accuracy to that of the ANN models. Copyright © 2012 John Wiley & Sons, Ltd.

Using wavelet transform to improve generalization capability of feed forward neural networks in monthly runoff prediction

In the study presented, a hybrid model is proposed for monthly runoff prediction by using wavelet transform and feed forward neural networks. Discrete wavelet transform (DWT) and Levenberg-Marquardt optimization algorithm based feed forward neural networks (FFNN) are considered for the modeling study. The study region covers the basins of Medar River which is located at the Aegean region of Turkey. Meteorological data, which represent the study region, were decomposed into wavelet sub-time series by DWT. Ineffective sub-time series were eliminated by using Mallow Cp coefficient based all possible regression method to prevent collinearity. Then, effective sub-time series components constituted the new inputs of FFNN. Some favorite evaluation measures, that is, determination coefficient (R2), adjusted determination coefficient (Adj.R2), Nash-Sutcliffe efficiency coefficient (NS), root mean squared error (RMSE), weighted mean absolute percentage error (WMAPE), were employed to assess modeling performances. The results determined in study indicate that the DWT based FFNN models (DWT-FFNN) are successful tools to model the monthly runoff series and can give good prediction performances than conventional methods. Key words: Wavelet transform, feed forward neural networks, Levenberg-Marquardt algorithm, monthly runoff prediction.

Analysis of the Characteristics of Runoff in Manasi River Basin in the Past 50 Years

Abstract Runoff generated in the upper Manasi River Basin is water resource supplied for the middle and lower reaches. Based on the monthly runoff series (1957-2006) of the two hydrological gauging stations, Kensiwate, Hongshanzui and precipitation series at 4 precipitation stations of the upper Manasi River Basin, multiple indexes and a series of methods and wavelet methods are adopted to analyze the inter-annual, annual and periodic variation characteristics of the streamflows in the watershed for the last 50 years. The results show that: showed that: (1) Distribution of annual runoff was uneven, and mainly focuses on June and july and August. The inter-annual variations were great, with significant difference between the wet year and dry year; (2) annual runoff changes had obvious stages and the tendency of annual runoff increasing was significant; and (3) annual runoff changes had a main period of about30 years and 15 years, with an abrupt point happening in 1995. The increasing of precipitation and Climate warming are the key factors influencing the runoff.

Analysis the Hydrological Situation of the Influx Runoff Series for Poyang Lake

According to historical observed data for the runoff into the Poyang Lake, the analysis of runoff into the Poyang Lake about 50 years was conducted. The results showed that the five rivers runoff for the influx into the Poyang Lake, were decreased since 2002, and the runoff decline phenomenon like this has appeared in the 60's and 80's. About the runoff annual distribution, the uneven distribution was detected. The sum of the monthly runoff from April to July account for 46%∼69% about the entire year, and in the dry season, the sum of the monthly runoff from October to next year January only account for 12%∼23%. About the each monthly runoff series of the five rivers from 1956 to 2008, it was observed that during the flood season the dispersion coefficient of the series is small, and in the non-flood season the dispersion coefficient of the series is bigger. It illuminated that the monthly runoff series were changer in the non-flood than the flood season and the increased human activities may be the major factor.

Using the R/ S method to determine the periodicity of time series

R/ S method is widely used to estimate long-range dependence of a time series, but few papers do research on how to use the R/ S method to determine the periodicity. In this paper, the log ( h ) - log ( ( R / S ) h ) figures and the log ( h ) - V h figures are further studied by lots of numeral simulations, which shows that the two figures of a periodic time series both have obviously similar structures. Based on these structures, a new method, similar figure method (SFM), is established to estimate whether a time series has periodicity and determine the length of the periodicity. SFM is tested with a disturbed nonlinear time series, an actual monthly runoff series and a random series. The results show that SFM is effective. This method is an extension to the R/ S analysis.

Dynamics of River Flow Regimes Viewed through Attractors

Analysing complicated systems we must never forget that the extraordinary increase in the rate of progress in our understanding can tell us nothing about the state of our knowledge. H.Metzger (1994) The hydrological system is extremely complex. To get insight into its behaviour and possible future states it is important to assess not only its individual components but also consider their interaction. Studying the hydrological system in its context allows focusing on the inherent patterns of its behaviour and search for external rules of its functioning. Changing climate can alter the seasonality of river flow, increasing the frequency of some patterns and decreasing that of the other, which can be adequately described in probabilistic terms. Dimensionality reduction facilitates to discern the patterns present in high dimensional systems. The approach suggested herein allows studying the dynamics of river flow seasonality patterns in the context of the present hydrological system in terms of the probability density functions of the weight coefficients in the reduced phase space, using Principal Components for dimensionality reduction. The approach is presented on the example of a large sample of monthly runoff series for Scandinavia and France, which gives a possibility to identify some common features in the dynamics of seasonality of river flow across a vast region of Europe.