Uncertainty Analysis in Sediment Load Modeling Using ANN and SWAT Model

Sediment load estimation is essential in many water resources projects. In this study, the capability of two different types of model including SWAT as a process-based model and ANNs as a data-driven model in simulating sediment load were evaluated. The issue of uncertainty in the simulated outputs of the two models which stems from different sources was also investigated. Calibration and uncertainty analysis of SWAT were performed using monthly observed discharge and sediment load values and through the application of SUFI-2 procedure. The issue of uncertainty in the ANN model was also accounted for by training a network several times with different initial weights and bias values as well as randomly-selected training and validation sets, each time a network trained. Trying different input variables to find the best and most efficient network structure, it was found that in the forested watershed of Kasilian, adding average daily rainfall or previous values of discharge dose not change the performance of the ANN model significantly. Comparing the results of SWAT and ANN, it was found that SWAT model has a superior performance in estimating high values of sediment load, whereas ANN model estimated low and medium values more accurately. Moreover, prediction interval for the results of ANN was narrower than that of SWAT which suggests that ANN outputs are with less uncertainty.     

Estimating Groundwater Withdrawal in Poorly Gauged Agricultural Basins

A method is presented for estimating the annual groundwater withdrawal based on water balance resulting from surface and groundwater hydrological considerations. The unknown groundwater losses of the aquifer are estimated from the groundwater level fluctuations, the specific yield and the groundwater withdrawal prior to the installation of the irrigation network. The meteorological and hydrometric data are used in the Sacramento hydrological conceptual model for the estimation of the stored groundwater volume, via minimization of the difference between the simulated and measured stream discharge. Following the installation of the network due to poorly kept field records, an initial estimation of the groundwater withdrawal is made based on the fluctuations of the groundwater level, the specific yield, and the annual precipitation. The monthly stored groundwater volume is verified against the volume obtained during the recharge of the basin (November–April). The difference between the groundwater volume and the measured discharge of the basin (May–October) is in agreement with the initial estimation of the groundwater withdrawal. This method is applied successfully in an agricultural basin on the island of Crete, Greece and its novelty lies in the fact that it can be used in basins where groundwater withdrawal is not known or data is incomplete.     

Precipitation Estimation Using Support Vector Machine with Discrete Wavelet Transform

Precipitation prediction is of dispensable importance in many hydrological applications. In this study, monthly precipitation data sets from Serbia for the period 1946–2012 were used to estimate precipitation. To fulfil this objective, three mathematical techniques named artificial neural network (ANN), genetic programming (GP) and support vector machine with wavelet transform algorithm (WT-SVM) were applied. The mean absolute error (MAE), mean absolute percentage error (MAPE), root mean square error (RMSE), Pearson correlation coefficient (r) and coefficient of determination (R²) were used to evaluate the performance of the WT-SVM, GP and ANN models. The achieved results demonstrate that the WT-SVM outperforms the GP and ANN models for estimating monthly precipitation.     

Predicting Monthly Spring Discharges Using a Simple Statistical Model

Current precipitation and past climate variability induce considerable intermonthly fluctuations in spring discharges. This study presents the DISHMET model (Discharge Hydro-Climatological Model) developed to perform historical spring reconstructions in the lack of physical assumptions. We analyzed discharge data of the Caraventa spring, located on the southern side of Mount La Montagna in Southern Italy, which has been monitored since the 1996s. The La Montagna aquifer is tectonically and litologically complex and deformed bedding controls the groundwater flow. Due to this aspect a parsimonious model should be more suitable than a complex model in spring discharge estimation. Thus, the DISHMET model incorporates monthly and annual precipitation only. The model is able to estimate sufficiently well the monthly fluctuations of groundwater discharge. DISHMET can be easily used to assess historical discharge, even when hydrological data is discontinuously available. The magnitude of this discharge is linked to the frequency and type of weather patterns transiting over the central Mediterranean area during the autumn and winter seasons. It is mainly related to the local precipitation that recharges the Mt. La Montagna aquifer. An analysis of antecedent rainfall and spring discharge reveal moderate to strong relationships.     

基于WD-COA-LSTM模型的月降水量预测

为进一步提高月降水量预测精度,提出了基于小波分解（WD）和郊狼优化（COA）算法的长短期记忆神经网络（LSTM）降水量预测模型（WD-COA-LSTM）。首先用小波分解对时间序列进行预处理,消除序列的非平稳性,得到1个低频序列和3个高频序列;然后通过郊狼优化算法对神经网络（LSTM）模型进行参数优化;最后将各子序列预测值叠加得到月降水量预测值。将提出的模型应用于洛阳市栾川县白土镇和洛宁县故县镇两个雨量站的月降水量预测中,并与LSTM、COA-LSTM、WD-LSTM模型预测结果进行对比。结果表明：提出的WD-COA-LSTM模型的预测精度最高,说明小波分解和郊狼优化算法能有效加强LSTM模型预测的精度和泛化能力,为月降水量的预测提供了一种新的途径。     

基于Matlab神经网络的流域年径流量预测

阐述了运用人工神经网络模型对流域年径流量径流序列做出预报,表明人工神经网络模型在水文预报中具有一定的优势。通过BP神经网络算法得到了适合该神经网络模型的训练算法。以渔峡口站年径流量实测序列为研究对象,在数值试验的基础上建立了年径流序列预报的人工神经网络预报模型结构,提高了该模型的预报准确性。     

Reference Evapotranspiration Prediction Using Neural Networks and Optimum Time Lags

The reference evapotranspiration (ET₀) plays a significant role especially in agricultural water management and water resources planning for irrigation. It can be calculated using different empirical equations and forecasted by applying various artificial intelligence techniques. The simulation result of a machine learning technique is a function of its structure and model inputs. The purpose of this study is to investigate the effect of using the optimum set of time lags for model inputs on the prediction accuracy of monthly ET₀ using an artificial neural network (ANN). For this, the weather data time-series i.e. minimum and maximum air temperatures, vapour pressure, sunshine hours, and wind speed were collected from six meteorological stations in Serbia for the period 1980–2010. Three ANN models were applied to monthly ET₀ time-series to study the impacts of using the optimum time lags for input time-series on the performance of ANN model. Achieved results of goodness–of–fit statistics approved the results obtained by scatterplots of testing sets - using more time lags that are selected based on their correlation to the dataset is more efficient for monthly ET₀ prediction. It was realized that all the developed models showed the best performances at Loznica and Vranje stations and the worst performances at Nis station. Simultaneous assessment of the impact of using a different number of time lags and the set of time lags that show a stronger correlation to the dataset for input time-series, on the performance of ANN model in monthly ET₀ prediction in Serbia is the novelty of this study.

     

The Research of Monthly Discharge Predictor-corrector Model Based on Wavelet Decomposition

Based on wavelet analysis theory, a wavelet predictor-corrector model is developed for the simulation and prediction of monthly discharge time series. In this model, the non-stationary time series of monthly discharge is decomposed into an approximated time series and several stationary detail time series according to the principle of wavelet decomposition. Each one of the decomposed time series is predicted, respectively, through the ARMA model for stationary time series. Then the correction procedure is conducted for the sum of the prediction results. Taking the monthly discharge at Yichang station of Yangtse River as an example, the monthly discharge is simulated by using ARMA model, seasonal ARIMA model, BP artificial neural network model and the wavelet predictor-corrector model proposed in this article, respectively. And the effect of decomposition scale for the wavelet predictor-corrector model is also discussed. It is shown that the wavelet predictor-corrector model has higher prediction accuracy than the some other models and the decomposition scale has no obvious effect on the prediction for monthly discharge time series in the example.     

Drought Forecasting using Markov Chain Model and Artificial Neural Networks

Water resources management is a complex task. It requires accurate prediction of inflow to reservoirs for the optimal management of surface resources, especially in arid and semi-arid regions. It is in particular complicated by droughts. Markov chain models have provided valuable information on drought or moisture conditions. A complementary method, however, is required that can both evaluate the accuracy of the Markov chain models for predicted drought conditions, and forecast the values for ensuing months. To that end, this study draws on Artificial Neural Networks (ANNs) as a data-driven model. The employed ANNs were trained and tested by means of a statistically-based input selection procedure to accurately predict reservoir inflow and consequently drought conditions. Thirty three years’ data of inflow volume on a monthly time resolution were selected to enable calculation of the standardized streamflow index (SSI) for the Markov chain model. Availability of hydro-climatic data from the Doroodzan reservoir in the Fars province, Iran, allowed us to develop a reservoir specific ANN model. Results demonstrated that both models accurately predicted drought conditions, by employing a randomization procedure that facilitated the selection of the required data for the ANN to forecast reservoir inflow close to the observed values over a validation period. The results confirmed that combining the two models improved short-term prediction reliability. This was in contrast to single model applications that resulted into substantial uncertainty. This research emphasized the importance of the correct selection of data or data mining, prior to entering a specific modeling routine.     

Uncertainty Analysis of Climate Change Impacts on Flood Frequency by Using Hybrid Machine Learning Methods

In the present study, for the first time, a new framework is used by combining metaheuristic algorithms, decomposition and machine learning for flood frequency analysis under climate-change conditions and application of HadCM3 (A2 and B2 scenarios), CGCM3 (A2 and A1B scenarios) and CanESM2 (RCP2.6, RCP4.5 and RCP8.5 scenarios) in global climate models (GCM). In the proposed framework, Multivariate Adaptive Regression Splines (MARS) and M5 Model tree are used for classification of precipitation (wet and dry days), whale optimization algorithm (WOA) is considered for training least square support vector machine (LSSVM), wavelet transform (WT) is used for decomposition of precipitation and temperature, LSSVM-WOA, LSSVM, K nearest neighbor (KNN) and artificial neural network (ANN) are performed for downscaling precipitation and temperature, and discharge is simulated under present period (1972–2000), near future (2020–2040) and far future (2070–2100). Log normal distribution is used for flood frequency analysis. Furthermore, analysis of variance (ANOVA) and fuzzy method are employed for uncertainty analysis. Karun3 Basin, in southwest of Iran, is considered as a case study. Results indicated that MARS performed better than M5 model tree. In downscaling, ANN and LSSVM_WOA slightly outperformed other machine learning algorithms. Results of simulating the discharge showed superiority of LSSVM_WOA_WT algorithm (Nash-Sutcliffe efficiency (NSE)?=?0.911). Results of flood frequency analysis revealed that 200-year discharge decreases for all scenarios, except CanESM2 RCP2.6 scenario, in the near future. In the near and far future periods, it is obvious from ANOVA uncertainty analysis that hydrological models are one of the most important sources of uncertainty. Based on the fuzzy uncertainty analysis, HadCM3 model has lower uncertainty in higher return periods (up to 60% lower than other models in 1000-year return period).

     

Uncertainties in the Methods of Flood Discharge Measurement

This study demonstrates an application of uncertainty analysis in evaluating methods of discharge measurement including: the velocity-area, rating curve and efficient methods based on the probabilistic velocity distribution equation. The measurement of river discharge plays a large part in the distribution of water resources. The conventional methods of discharge measurement are costly, time-consuming, and dangerous. Therefore the efficient method of discharge measurement which bases on the relationship between maximum and mean velocities being constant was employed to justify its alternative for the conventional methods: velocity-area and rating curve methods. Distribution test was applied to investigate the statistical properties of the uncertainties involved in the three methods of discharge measurement. Latin hypercube sampling (LHS) method was employed accordingly to assess the discharge features of the three methods of discharge measurement. The main purpose of this study is to quantify the uncertainty involved in several discharge measurement methods and justify the availability and reliability of using the efficient method as an alternative of the conventional methods. Results show that the correlation analysis also validates that the efficient method is a more reliable method than the rating curve method to yield accurate discharge measurements. Moreover, it also yielded comparably accurate measurements as those by the velocity-area method.     

Artificial Neural Network and Support Vector Machine Models for Inflow Prediction of Dam Reservoir (Case Study: Zayandehroud Dam Reservoir)

Inflow prediction of reservoirs is of considerable importance due to its application in water resources management related to downstream water release planning and flood protection. Therefore, in this research, different new input patterns for predicting inflow to Zayandehroud dam reservoir is proposed employing artificial neural network (ANN) and support vector machine (SVM) models. Nine different models with different patterns of input data such as inflow to the dam reservoir considering time duration lags, time index, and monthly rainfall of Ghaleh-Shahrokh station have been proposed to predict the inflow to the dam reservoir. Comparison of the results indicates that the ninth proposed model has the least error for inflow prediction in which the results of SVM model outperform those of ANN model. That is, the least error has been obtained using the ninth SVM (ANN) model with correlation coefficient (R) values of 0.8962 (0.89296), 0.9303 (0.92983) and 0.9622 (0.95333) and root mean squared error (RMSE) values of 47.9346 (48.5441), 42.69093 (43.748) and 23.56193 (28.5125) for training, validation and test data, respectively.

     

Monthly Rainfall Prediction Using Wavelet Neural Network Analysis

Rainfall is one of the most significant parameters in a hydrological model. Several models have been developed to analyze and predict the rainfall forecast. In recent years, wavelet techniques have been widely applied to various water resources research because of their time-frequency representation. In this paper an attempt has been made to find an alternative method for rainfall prediction by combining the wavelet technique with Artificial Neural Network (ANN). The wavelet and ANN models have been applied to monthly rainfall data of Darjeeling rain gauge station. The calibration and validation performance of the models is evaluated with appropriate statistical methods. The results of monthly rainfall series modeling indicate that the performances of wavelet neural network models are more effective than the ANN models.     

Daily Outflow Prediction by Multi Layer Perceptron with Logistic Sigmoid and Tangent Sigmoid Activation Functions

This paper discusses the use of artificial neural network (ANN) models for predicting daily flows from Khosrow Shirin watershed located in the northwest part of Fars province in Iran. A Multi-Layer Perceptron (MLP) neural network was developed using five input vectors leading to five ANN models: MLP1, MLP2, MLP3, MLP4, and MLP5. Two activation functions were used and they were logistic sigmoid and tangent sigmoid. The MLP_Levenberg–Marquardt (LM) algorithm was used for the training of ANN models. A 5-year data record, selected randomly, was used for ANN training and testing. The predicted outflow showed that the tangent sigmoid activation function performed better than did the logistic sigmoid activation function. The values of R ² and RMSE for MLP4 with the tangent sigmoid activation function for the validation period were equal to 0.89 and 1.7 m³/s, respectively. Appropriate input vectors for MLPs were determined by correlation analysis. It was found that antecedent precipitation and discharge with 1 day time lag as an input vector best predicted daily flows. Also, comparison of MLPs showed that an increase in input data was not always useful.     

多元线性回归与BP神经网络模型在次洪量预测中的对比研究

针对半干旱地区次洪量预测问题,选取岔巴沟流域曹坪水文站1980-2010年中15场洪水资料,根据实测次暴雨、洪量资料,考虑淤地坝控制面积、次暴雨量、暴雨中心位置、前期影响雨量等因子,利用SPSS及MATLAB软件,建立用以预测次洪量的多元线性回归模型和BP神经网络模型。模型预测结果比较表明:多元线性回归模型和BP神经网络模型都能较好地应用于次洪量的预测,进一步得出BP神经网络模型的预测效果优于多元线性回归模型。研究结果可为淤地坝的安全度汛提供决策依据。     

基于AM-MCMC算法的贝叶斯概率洪水预报模型

本文在贝叶斯预报系统的框架下,利用BP网络能描述非线性映射的特性建立了基于BP网络的先验密度和似然函数的模型,并采用基于自适应采样算法(Adaptive Metropolis algorithm,简称AM)的马尔可夫链蒙特卡罗模拟方法(Markov Chain Monte Carlo,简称MCMC)求解流量的后验密度,最后给出流量的概率预报。实例表明,基于AM-MCMC的BP贝叶斯概率水文预报的精度高,且能给出预报的方差,使得防洪决策可以考虑预报的不确定性。     

Optimal Selection of Predictor Variables in Statistical Downscaling Models of Precipitation

Two screening methods aimed at selection of predictor variables for use in a statistical downscaling (SD) model developed for precipitation are proposed and evaluated in this study. The SD model developed in this study relies heavily on appropriate predictors chosen and accurate relationships between site-specific predictand (i.e. precipitation) and general circulation model (GCM)-scale predictors for providing future projections at different spatial and temporal scales. Methods to characterize these relationships via rigid and flexible functional forms of relationships using mixed integer nonlinear programming (MINLP) formulation with binary variables, and artificial neural network (ANN) methods respectively are developed and evaluated in this study. The proposed methods and three additional methods based on the correlations between predictors and predictand, stepwise regression (SWR) and principal component analysis (PCA) are evaluated in this study. The screening methods are evaluated by employing them in conjunction with an SD model at 22 rain gauge locations in south Florida, USA. The predictor variables that are selected by different predictor selection methods are used in a statistical downscaling model developed in this study to downscale precipitation at a monthly temporal scale. Results suggest that optimal selection of variables using MINLP and ANN provided improved performance and error measures compared to two other models that did not use these methods for screening the variables. Results from application and evaluations of screening methods indicate improved downscaling of precipitation is possible by SD models when an optimal set of predictors are used and the selection of the predictors is site-specific.     

大尺度分布式水文模型VIC在嘉陵江流域径流模拟中的应用

大尺度分布式流域水文模型是目前评价流域环境变化的重要工具,以嘉陵江流域为研究对象,构建了嘉陵江流域大尺度分布式VIC模型(Variable Infiltration Capacity,VIC model),利用Maryland大学的全球1 km×1 km土地覆盖数据,同时参考LDAS(Land Data Assimilation System)成果,建立了嘉陵江流域VIC模型的参数库,通过4个水文站以上流域的水文模拟试验,结果表明建立的VIC模型能有效地模拟嘉陵江流域各典型站的日、月径流过程,模拟的水量平衡误差在5%以内,日径流过程模拟的确定性系数均在70%以上,月径流过程模拟的确定性系数超过90%。该模型可以用来分析环境变化对嘉陵江流域水资源及洪水过程的影响。     

Bayesian Rating Curve Inference as a Streamflow Data Quality Assessment Tool

A streamflow time-series is normally obtained by transforming a time-series of recorded stage to discharge using an estimated rating curve. The accuracy of this streamflow time-series depends on the characteristics of the available stage-discharge measurements used to fit the rating curve. The Norwegian Water Resources and Energy Directorate (NVE) has developed a method based on rating curve uncertainty for performing objective quality assessment of streamflow time-series. The method, which is based on a Bayesian statistical framework, uses the available stage-discharge measurements and the corresponding stage time-series to derive statistics utilised for a quality assurance of the streamflow time-series. Nearly one thousand streamflow time-series periods have been classified using the method. This paper presents the results.     

Influence of Time Discretization and Input Parameter on the ANN Based Synthetic Streamflow Generation

The capability of ANN to generate synthetic series of river discharge averaged over different time steps with limited data has been investigated in the present study. While an ANN model with certain input parameters can generate a monthly averaged streamflow series efficiently; it fails to generate a series of smaller time steps with the same accuracy. The scope of improving efficiency of ANN in generating synthetic streamflow by using different combinations of input data has been analyzed. The developed models have been assessed through their application in the river Subansiri in India. Efficiency of the ANN models has been evaluated by comparing ANN generated series with the historical series and the series generated by Thomas-Fiering model on the basis of three statistical parameters-periodical mean, periodical standard deviation and skewness of the series. The results reveal that the periodical mean of the series generated by both Thomas–Fiering and ANN models are in good agreement with that of the historical series. However, periodical standard deviation and skewness coefficient of the series generated by Thomas–Fiering model is inferior to that of the series generated by ANN.