Search

The Qanat (also known as kariz) is one of the significant water resources in many arid and semiarid regions. The present research aims to use machine learning techniques for Qanat discharge (QD) prediction and find a practical model that predicts QD well. Gene expression programming (GEP), artificial neural network (ANN), group method of data handling (GMDH), least-square support vector machine (LSSVM) and adaptive neuro-fuzzy inference system (ANFIS), are employed to predict one-, two-, and five-months time-step ahead QD in an unconfined aquifer. QD for one, two, and three lag-times (QDt−1, QDt−2, QDt−3), QD for adjacent Qanat, the main meteorological components (Tt, ETt, Pt) and GWL for one, two, and three lag-times are utilized as input dataset to accomplish accurate QD prediction. The GMDH model, according to its best results, had promising accuracy in predicting multi-step ahead monthly QD, followed by the LSSVM, ANFIS, ANN and GEP, respectively. [ABSTRACT FROM AUTHOR]

Compound channels play an important role in hydraulic and hydrological engineering. Many rivers and streams have naturally compound channels, which are important for both flood control and river ecosystems. When predicting discharge, interval predictions are usually more realistic and reliable than traditional point predictions made using ML models. In this study, the authors used three well-known machine learning techniques (Adaptive Neuro Fuzzy Inference System (ANFIS), Support Vector Machine (SVM) and Artificial Neural Networks (ANN)) to predict compound channel discharge. To accurately assess uncertainty, two resampling techniques were employed: Bootstrap (B) and Jackknife after Bootstrap (JB). The proposed approach reflects the impacts of statistical uncertainties using prediction confidence intervals and the reliability of point prediction. The JB resampling technique has proven to offer superior accuracy in terms of point prediction and prediction intervals when compared to Bootstrap. Bootstrapping did not yield superior outcomes compared with the standalone ML models. The JB-ANN and JB-ANFIS models outperformed the JB-SVM and Bootstrap ML models in the testing phase, as shown by various statistical measures. This study suggests that the JB-ANN approach is a reliable and robust tool for accurately predicting compound discharges. The findings reveal that when the models have low accuracy in point prediction, the level of uncertainty in predicting the interval also increases. The innovation of this research is enhancing the examination of uncertainties in compound channel discharge by incorporating two resampling techniques to analyze three different ML models. [ABSTRACT FROM AUTHOR]

The generation of streamflow is linked with different factors such as water level, rainfall intensity, meteorological variables, and many more. In this study, we have developed a new hybrid approach (named LASSO-HF-SAA) by integrating the least absolute shrinkage and selection operator (LASSO) and Hampel filter (HF) with three data-driven models i.e. support vector machine (SVM), artificial neural network (ANN) and autoregressive distributed lag (ARDL). Firstly, LASSO selects meteorological variables important in daily streamflow prediction. Next, the HF detects and correct outliers in the variables to handle the randomness and noise of data. Thirdly, the HF-corrected data is fed to SVM, ANN, and ARDL models to obtain the predictions of the proposed LASSO-HF-SVM, LASSO-HF-ANN, and LASSO-HF-ARDL models. The performance of these models is checked using performance indices and the Diebold-Mariano (DM) test. The proposed hybrid approach is illustrated on the streamflow data of the Kabul River (Nowshera station) of Pakistan. Based on Nash-Sutcliffe efficiency (NSE), it is revealed that the prediction accuracy of the LASSO-HF-SVM hybrid model (NSE = 0.52) is better than SVM (NSE = 0.43), HF-SVM (NSE = 0.49) and LASSO-SVM (NSE = 0.47) models in testing phase. Similar findings are for the proposed LASSO-HF-ARDL and LASSO-HF-ANN hybrid models. Overall, the suggested LASSO-HF-ARDL hybrid model has shown winning performance compared to all models in the study. The root mean squared error (RMSE) and NSE of the proposed LASSO-HF-ARDL model is 443.5m3/s and 0.68 on the test data. The DM test confirms that the prediction accuracy of the proposed hybrid models is better than their respective single, HF-based, and LASSO-based models versions of SVM, ANN, and ARDL models respectively. [ABSTRACT FROM AUTHOR]

This study compares the ability of Long Short-Term Memory (LSTM) tuned with Grey Wolf Optimization (GWO) and machine learning models, artificial neural network (ANN), Adaptive-Network-based Fuzzy Inference System (ANFIS), and support vector machine (SVM) enhanced with GWO in the prediction of monthly streamflow. Precipitation, temperature, and streamflow data obtained from stations in Pakistan are used as inputs to the implemented models. The first application focused on streamflow prediction using local data and applying four different input scenarios; i) precipitation-based inputs, ii) temperature-based inputs, iii) lagged streamflow inputs, and iv) combinations of previous input scenarios. In the second application, streamflow data of one station are predicted using data from another station and similar input scenarios are considered. In the first application, the LSTM-GWO performed the best in one station, whereas the ANN-GWO produced the best predictions in another station. The mean absolute percentage error in LSTM-GWO and ANN-GWO are 17.9% and 22.63, respectively. Furthermore, it is observed that the LSTM-GWO generally offers the best accuracy in peak streamflow prediction with a maximum relative error of 113%. In the second application, the LSTM-GWO is the superior model which improves the mean absolute percentage error by 13%, 60%, and 101% compared to SVR-GWO, ANN-GWO, and ANFIS-GWO, respectively. [ABSTRACT FROM AUTHOR]

The study focuses on the critical concern of designing secure and resilient bridge piers, especially regarding scour phenomena. Traditional equations for estimating scour depth are limited, often leading to inaccuracies. To address these shortcomings, modern data-driven models (DDMs) have emerged. This research conducts a comprehensive comparison involving DDMs, including support vector machine (SVM), gene expression programming (GEP), multilayer perceptron (MLP), gradient boosting trees (GBT) and multivariate adaptive regression spline (MARS) models, against two regression equations for predicting scour depth around cylindrical bridge piers. Evaluation employs statistical indices, such as root-mean-square error (RMSE), coefficient of determination (R2), mean average error (MAE) and normalized discrepancy ratio (S(DDRmax)), to assess their predictive performance. A total of 455 datasets from previous research papers are employed for assessment. Dimensionless parameters Froude number F r = U gy , Pier Froude number F r P = U g ′ D , and the ratio of scour depth to pier diameter (y D) are carefully selected as influential model inputs through dimensional analysis and the gamma test. The results highlight the superior performance of the SVM model. In the training phase, it exhibits an RMSE of 0.1009, MAE of 0.0726, R2 of 0.9401, and SDDR of 2.9237. During testing, the SVM model shows an RMSE of 0.023, MAE of 0.017, R2 of 0.984, and SDDR of 5.301. Additionally, it has an average error of − 0.065 and a total error of − 20.642 in the training set and an average error of − 0.005 and a total error of − 0.707 in the testing set. Conversely, the M5 model exhibits the lowest accuracy. The statistical metrics unequivocally establish the SVM model as significantly outperforming the experimental models, placing it in a higher echelon of predictive accuracy. [ABSTRACT FROM AUTHOR]

Groundwater management is key to attaining sustainable development goals, especially in arid and semi-arid countries. Hence, a precise estimate of the aquifer hydrodynamic parameters (hydraulic conductivity, transmissivity, specific yield, and storage coefficient) is required for proper groundwater resource management. The central goal of this research is to utilize machine learning models to estimate transmissivity by pumping test data in unconfined alluvial aquifers. Artificial neural network (ANN), gene expression programming (GEP), adaptive neuro-fuzzy inference system (ANFIS), fuzzy logic (FL), least square support vector machine (LSSVM), and group method of data handling (GMDH) methods were utilized to estimate transmissivity. To achieve this goal, pumping tests and hydrogeological data from 96 pumping wells located in the central plateau of Iran were collected and normalized. Using the existing normalized data, several combinations were utilized as inputs to the models, and then randomly, 70% of the data were used for the training step and 30% for the testing step. Finally, ten combinations that provided a better answer were selected from whole combinations. The mean absolute error, root means square error and correlation coefficient were used to assess the models' precision. The comparison criteria revealed that while all developed methods could provide desirable transmissivity estimations, the GMDH model was the most accurate and precise. The result of the models suggested that the best combination in transmissivity estimating was combination 3 (well discharge, thickness, depth of pumping well, and minimum and maximum time-drawdown data). Therefore, using this method and without graphic methods, it is feasible to estimate the transmissivity with acceptable accuracy in unconfined aquifers with similar hydrogeological and geological characteristics. [ABSTRACT FROM AUTHOR]

Accurate streamflow estimation is crucial for proper water management for irrigation, hydropower, drinking and industrial purposes. The main aim of this study to adopt new data preprocessing techniques (e.g., EMD, EEMD and EWT) to capture the data noise and to enhance the prediction accuracy of machine learning methods for streamflow estimation which is a challenging task in high-altitude basins due to the influence of many external climatic and geographical parameters. The prediction accuracy of support vector regression (SVR), twin support vector machine (T), extreme learning machine (ELM), asymmetric Huber loss function-based ELM (AHELM) and ε-insensitive Huber loss function-based ELM (ε-AHELM) methods are investigated in monthly streamflow prediction. Among the standalone methods, the ε-AHELM performs superior to the SVR, TSVR, ELM, and AHELM in streamflow prediction; improvements in root mean square error are 6.9%, 4.9%, 6% and 4.2%, respectively. The study outcomes reveal that the preprocessing methods considerably improve the prediction accuracy of the implemented standalone models. Among the data preprocessing techniques, it is found that the EWT outperforms the EMD and EEMD techniques by reducing the prediction errors in the best ε-AHELM, EMD-ε-AHELM and EEMD-ε-AHELM models by 68–61.3%, 64.7–63.4% and 59.4–58.6%, respectively. The overall results of the study recommend the use of EWT-ε-AHELM in streamflow estimation. [ABSTRACT FROM AUTHOR]

Side orifices are commonly installed in the side of a main channel to spill or divert some of the flow from the source channel to lateral channels. The aim of the present study is the accurate estimation of the discharge coefficient for flow through triangular (Δ-shaped) side orifices by applying three data-driven models including support vector machine (SVM), least squares support vector machine (LSSVM) and least squares support vector machine improved by gravity search algorithm (LSSVM-GSA). The discharge coefficient was estimated by utilizing five dimensionless variables resulted from experimental data (570 runs). Five different scenarios were applied based on the input variables. The models were evaluated through several statistical indices and graphical charts. The results showed that all of the models could successfully estimate the discharge coefficient of Δ-shaped side orifices with adequate accuracy. However, the LSSVM-GSA produced the best performance for the input combination of all variables with the highest coefficients of determination (R2) and Nash–Sutcliffe efficiency (NSE), equal to 0.965 and 0.993, and the least root mean square error (RMSE) and mean absolute error (MAE), equal to 0.0099 and 0.0077, respectively. The LSSVM-GSA improved the RMSE of the SVM and LSSVM by 26% and 20% in estimating the discharge coefficient. Furthermore, the ratio of orifice crest height to orifice height (W/H) was identified as having the highest influence on the discharge coefficient of triangular side orifices among the various input variables. [ABSTRACT FROM AUTHOR]

Machines learning models have recently been proposed for predicting rivers water temperature (Tw) using only air temperature (Ta). The proposed models relied on a nonlinear relationship between the Tw and Ta and they have proven to be robust modelling tools. The main motivation for this study was to evaluate how the variational mode decomposition (VMD) contributed to the improvement of machines learning performances for river Tw modelling. Measured data collected at five stations located in Poland from 1987 to 2014 were acquired and used for the analysis. Six machines learning models were used and compared namely, K-nearest neighbor's regression (KNNR), least square support vector machine (LSSVM), generalized regression neural network (GRNN), cascade correlation artificial neural networks (CCNN), relevance vector machine (RVM), and locally weighted polynomials regression (LWPR). The six models were developed according to three scenarios. First, the models were calibrated using only the Ta as input and obtained results show that the models were able to predict consistently water temperature, showing a high determination coefficient (R2) and Nash–Sutcliffe efficiency (NSE) with values near or above 0.910 and 0.915, respectively, and in overall the six models worked equally without clear superiority of one above another. Second, the air temperature was combined with the periodicity (i.e., day, month and year number) as input variable and a significant improvement was achieved. Both models show their ability to accurately predict river Tw with an overall accuracy of 0.956 for R2 and 0.955 for NSE values, but the LSSVM2 have some advantages such as a small errors metrics, and high fitting capabilities and it slightly surpasses the others models. Thirdly, air temperature was decomposed into several intrinsic mode functions (IMF) using the VMD method and the performances of the models were evaluated. The VMD parameters appeared to cause much influence on the prediction accuracy, exhibiting an improvement of about 40.50% and 39.12% in terms of RMSE and MAE between the first and the third scenarios, however, some models, i.e., GRNN and KNNR have not benefited from the VMD. This research has demonstrated the high capability of the VMD algorithm as a preprocessing approach in improving the accuracies of the machine learning models for river water temperature prediction. [ABSTRACT FROM AUTHOR]

Precise estimation of groundwater level (GWL) might be of great importance for attaining sustainable development goals and integrated water resources management. Compared with alternative numerical models, soft computing methods are promising tools for GWL prediction, which need more hydrogeological and aquifer characteristics. The central aim of this research is to explore the performance of such well-accepted data-driven models to predict monthly GWL with emphasis on major meteorological components, including; precipitation (P), temperature (T), and evapotranspiration (ET). Artificial neural network (ANN), fuzzy logic (FL), adaptive neuro-fuzzy inference system (ANFIS), group method of data handling (GMDH), and least-square support vector machine (LSSVM) are used to predict one-, two-, and three-month ahead GWL in an unconfined aquifer. The main meteorological components (Tt, ETt, Pt, Pt-1) and GWL for one, two, and three lag-time (GWLt-1, GWLt-2, GWLt-3) are used as input to attain precise prediction. The results show that all models could have the best prediction for one month ahead in scenario 5, comprising inputs of GWLt-1, GWLt-2, GWLt-3, Tt, ETt, Pt, Tt-1, ETt-1, Pt-1. Based on different evaluation criteria, all employed models could predict the GWL with a desirable accuracy, and the results of LSSVM are the superior one. [ABSTRACT FROM AUTHOR]

This paper focuses on the development of a robust accurate streamflow prediction model by balancing the abilities of exploitation and exploration to find the best parameters of a machine learning model. To do so, the simulated annealing (SA) algorithm is integrated with the mayfly optimization algorithm (MOA) as SAMOA to determine the optimal hyper-parameters of support vector regression (SVR) to overcome the exploration weakness of the MOA method. The proposed method is compared with the classical SVR and hybrid SVR-MOA. To examine the accuracy of the selected methods, monthly hydroclimatic data from Jhelum River Basin is used to predict the monthly streamflow on the basis of RMSE, MAE, NSE, and R2 indices. Test results show that the SVR-SAMOA outperformed the SVR-MOA and SVR models. SVR-SAMOA reduced the prediction errors of the SVR-MOA and SVR models by decreasing the RMSE and the MSE from 21.4% to 14.7% and from 21.7% to 15.1%, respectively, in the test stage. [ABSTRACT FROM AUTHOR]

This study employs two heuristic algorithms, including the genetic algorithm (GA) and ant colony optimization for continuous domains (ACOR), for optimizing the parameters of two soft computing models, namely adaptive neuro-fuzzy inference system (ANFIS) and least-squares support vector machine (LSSVM), which were used for modeling monthly precipitation for all 12 months of the year. Data from 40 meteorological stations situated in different parts of Iran were used. The effectiveness of input data was determined by internal correlationcoefficient and nonlinear sensitivity analysis. Selected input data were further evaluated by another sensitivity analysis method, cosine amplitude (CA). Considering different evaluation months, LSSVM was more accurate and reliable than ANFIS. It was also found that both algorithms improved the performance of models for most months of the year. ACOR was better and more reliable than was GA in optimizing the models. ACOR produced the best results in autumn that led to the improvement of performance of ANFIS in terms of correlation coefficient (R) and root-mean square error (RMSE) by 35% and 0.40 mm for October; 42% and 0.99 mm for November; and 31% and 0.74 mm for December. The performance of LSSVM was enhanced by 6% and 0.28 mm for October; 22% and 0.20 mm for November; and 4% and 0.10 mm for December, respectively. For July and August, the suggested algorithms could not improve the performance of ANFIS. The algorithms did optimize LSSVM in all months, so the RMSE and mean absolute error were improved by 0.15 and 0.28 mm for July and 0.28 and 0.56 mm for August, respectively. [ABSTRACT FROM AUTHOR]

Sediment yield is important for maintaining soil health, reservoir sustainability, environmental pollution, and conservation of natural resources. The main aim of the present work is to develop four machine learning models, artificial neural networks (ANNs), radial basis function (RBF), support vector machine (SVM) and multiple model (MM)-ANNs for forecasting daily sediment yield. These models were applied to the Shakkar and Manot watersheds covering 25 years (1990–2015) and 10 years (2000–2010) of rainfall and discharge data, respectively. Results showed that the MM-ANNs model satisfactorily predicted sediment yield and outperformed the other models providing the highest correlation coefficient (0.921, 0.883) and Nash-Sutcliffe efficiency (0.744, 0.763) and the lowest relative absolute error (0.360, 0.344) and root mean square error (23,609.5, 269,671.5) for the Shakkar and Manot during the test period, respectively. Hence, the MM-ANNs model can be successfully used for sediment prediction. [ABSTRACT FROM AUTHOR]

This study investigates the ability of support vector machine (SVM), random forest (RF) and geo-statistical (GS) methods in simulating long-term monthly air temperatures. Data of 30 locations in Iran were utilized and data scanning procedure was employed in the applications. Geographical coordinates, e.g. latitude, longitude and altitude as well as the periodicity component (month of the year), were utilized as inputs to the applied models. SVM and RF models obtained using k-fold testing were also compared with artificial neural networks (NNs) and neuro fuzzy (NF) methods employed in a previous study. Comparisons demonstrated the superiority of SVM and RF models over the NN and NF models in the present study. In all stations, the RF model with average scatter index of 0.111, average mean absolute error of 1.456 and average variance account for of 0.968 provided better estimates than the SVM model with corresponding values of 0.142, 1.855 and 0.945, respectively. Nonetheless, the obtained results revealed that the geo-statistics-based kriging method is able to spatially simulate air temperature. [ABSTRACT FROM AUTHOR]

Owing to the lack of lysimetric data in many regions, the standard Penman–Monteith equation‐adopted by Food and Agriculture Organization of the United Nations (FAO56‐PM model) is usually used to calculate the reference evapotranspiration (ETo). However, as this model needs many meteorological parameters that cannot be easily obtained in many regions, other simple models along with the soft computing models are used to obtain ETo values. The paper presents a comprehensive comparison of 12 soft computing models: gene expression programming (GEP), neuro‐fuzzy with grid partitioning (NF‐GP), neuro‐fuzzy with sub‐clustering (NF‐SC), multivariate adaptive regression spline (MARS), boosted regression tree (BT), random forest (RF), model tree (MT), support vector machine (SVM), SVM‐firefly algorithm (SVM‐FA), extreme learning machine (ELM), neural network‐particle swarm optimization (NN‐PSO) and neural network‐differential evolution (NN‐DE), to estimate daily ETo values in humid regions. Therefore, daily meteorological data from two weather stations (during a 12 year period) were used to assess the models. The obtained results revealed that very good efficiency was obtained from all the applied methods. The temperature‐based SVM‐FA (root mean square error (RMSE) = 0.324 mm, Nash–Sutcliffe co‐efficient (NS) = 0.960) and NF‐GP (RMSE = 0.272 mm, NS = 0.974) models generally provided the best accuracy when estimating the ETo of Sari and Bablosar, two humid stations in northern Iran, respectively. The accuracy ranks of the other models (from best to worst) were found to be: NN‐PSO, NF‐SC, ELM, NN‐DE, MARS, GEP, RF, SVM, BT and MT. Among the radiation‐based models, the NF‐GP provided the best accuracy at estimating the ETo of both stations. The other models were ranked as: ELM, SVM‐FA, NN‐DE, NN‐PSO, MARS, RF, BT, NF‐SC, SVM and MT, respectively. [ABSTRACT FROM AUTHOR]

Modelling rainfall, an important element of the hydrological cycle, is of crucial importance in hydrology, water resources engineering, irrigation scheduling and environmental issues. In the current study, the monthly rainfall amounts were simulated through the least squares support vector machine (LSSVM), the model tree (MT) and the geostatistical kriging approaches using geographical inputs. So, the latitude, longitude, altitude and the periodicity component were introduced as model inputs for simulating monthly rainfall values. Long‐term rainfall records from 73 weather stations covering different climatic contexts of Iran were used for developing and evaluating the proposed methodology. Two modelling scenarios, namely the at station and pooled scenarios, were carried out for assessing the capability of the applied models in modelling rainfall values. First, two different heuristic methods, the LSSVM and MT methods, were compared with each other. The results obtained revealed that the LSSVM model produced more accurate results than the MT for the at station scenario. Both methods gave better results in arid/semi‐arid regions than humid stations. In the pooled scenario, the rainfall data of humid and arid training stations were pooled and used for calibration and then tested at each station. In this scenario, the LSSVM was superior to the MT in modelling long‐term monthly rainfall. The LSSVM and MT models were then compared with the geostatistical kriging method in both scenarios. It was observed that the kriging method generally performed better than the heuristic methods in spatial modelling of rainfall. [ABSTRACT FROM AUTHOR]

Understanding streamflow patterns by incorporating climate signal information can contribute remarkably to the knowledge of future local environmental flows. Three machine learning models, the multivariate adaptive regression splines (MARS), the M5 Model Tree and the least squares support vector machine (LSSVM) are established to predict the streamflow pattern over the Mediterranean region of Turkey (Besiri and Baykan stations). The structure of the predictive models is built using synoptic-scale climate signal information and river flow data from antecedent records. The predictive models are evaluated and assessed using quantitative and graphical statistics. The correlation analysis demonstrates that the North Pacific (NP) and the East Central Tropical Pacific Sea Surface Temperature (Niño3.4) indices have a substantial influence on the streamflow patterns, in addition to the historical information obtained from the river flow data. The model results reveal the utility of the LSSVM model over the other models through incorporating climate signal information for modelling streamflow. [ABSTRACT FROM AUTHOR]

This study compares single and hybrid soft computing models for estimating daily solar radiation flux for two scenarios. Scenario I developed single soft computing models, including multilayer perceptron (MLP), support vector machines (SVM), adaptive neuro-fuzzy inference system (ANFIS), and multivariate adaptive regression splines (MARS), for estimating daily solar radiation flux at two stations from the USA and South Korea. The MLP model was used to evaluate the effect of factors controlling daily solar radiation flux. Using different combinations of controlling factors as input, the MLP and SVM models, based on evaluation measures, were found to be superior to the ANFIS and MARS models at Big Bend station, USA. In addition, the MLP, SVM, and MARS models performed better than did the ANFIS model at Incheon station, South Korea. Scenario II combined the discrete wavelet transform (DWT) and single soft computing models (e.g., MLP and SVM) for improved performance using 4-input combination. The wavelet-based MLP (WMLP) and SVM (WSVM) models were superior to other single soft computing models (MLP, SVM, ANFIS, and MARS) at two stations. Taylor diagrams, violin plots and point density plots were also utilized to examine the similarity between the observed and estimated solar radiation flux values. Results showed that scenarios I and II can be alternatives for estimating daily solar radiation flux based on different meteorological information, such as peninsular and continental conditions. [ABSTRACT FROM AUTHOR]

Reference evapotranspiration (ET0) plays a fundamental role in irrigated agriculture. The objective of this study is to simulate monthly ET0 at a meteorological station in India using a new method, an improved support vector machine (SVM) based on the cuckoo algorithm (CA), which is known as SVM-CA. Maximum temperature, minimum temperature, relative humidity, wind speed and sunshine hours were selected as inputs for the models used in the simulation. The results of the simulation using SVM-CA were compared with those from experimental models, genetic programming (GP), model tree (M5T) and the adaptive neuro-fuzzy inference system (ANFIS). The achieved results demonstrate that the proposed SVM-CA model is able to simulate ET0 more accurately than the GP, M5T and ANFIS models. Two major indicators, namely, root mean square error (RMSE) and mean absolute error (MAE), indicated that the SVM-CA outperformed the other methods with respective reductions of 5–15% and 5–17% compared with the GP model, 12–21% and 10–22% compared with the M5T model, and 7–15% and 5–18% compared with the ANFIS model, respectively. Therefore, the proposed SVM-CA model has high potential for accurate simulation of monthly ET0 values compared with the other models. [ABSTRACT FROM AUTHOR]

Accurate prediction of pan evaporation (PE) is one of the crucial factors in water resources management and planning in agriculture. In this research, two hybrid models, self-adaptive timefrequency methodology, ensemble empirical mode decomposition (EEMD) coupled with support vector machine (EEMD-SVM) and EEMD model tree (EEMD-MT), were employed to forecast monthly PE. The EEMD-SVM and EEMD-MT were compared with single SVM and MT models in forecasting monthly PE, measured between 1975 and 2008, at Siirt and Diyarbakir stations in Turkey. The results were evaluated using four assessment criteria, Nash-Sutcliffe Efficiency (NSE), root mean square error (RMSE), performance index (PI), Willmott's index (WI), and Legates-McCabe's index (LMI). The EEMD-MT model respectively improved the accuracy of MT by 36 and 44.7% with respect to NSE and WI in the testing stage for the Siirt station. For the Diyarbakir station, the improvements in results were less spectacular, with improvements in NSE (1.7%) and WI (2.2%), respectively, in the testing stage. The overall results indicate that the proposed pre-processing technique is very promising for complex time series forecasting and further studies incorporating this technique are recommended. [ABSTRACT FROM AUTHOR]

Solar radiation, moisture and temperature are the most vital meteorological variables which affect plant growth. Due to the fact that the global solar radiation (GSR) is scarcely gauged at meteorological stations in developing countries, it is commonly estimated by data-driven techniques or by empirical equations. In this study, support vector regression (SVR), model trees (MT), gene expression programming (GEP) and adaptive neuro-fuzzy inference system (ANFIS) and several empirical equations were applied to assess the relations between GSR and several meteorological variables including minimum temperature (Tmin), maximum temperature (Tmax), relative humidity (RH), sunshine hours (n), maximum sunshine hours (N), corrected clear-sky solar irradiation (ICSKY), day of year (DOY) and extra-terrestrial radiation (Ra). For this purpose, the daily GSR measured from the beginning of 2011 to the end of 2013 at Tabriz synoptic station, which is located in semi-arid regions of Iran, were used. A direct strong relationship was observed to exist between the GSR and n. For evaluating the performances of studied techniques, three different statistical indicators were used namely root mean square error (RMSE), mean absolute error (MAE), and correlation coefficient (CC). Additionally, a Taylor diagram was utilized to test the similarity between the observed and predicted GSR values. Results indicated that the SVR-6 with input parameters of Ra, RH, Tmin, Tmax, n/N had better accuracy in predicting GSR with RMSE of 1.656, MAE of 0.990, CC of 0.980 and WI of 0.990 than the other models. Moreover, MT-6 ranked as the second best model in the prediction of GSR values. As an interesting point, studied empirical equations had lower accuracies comparing with the SVR, GEP, MT and ANFIS methods. For instance, GSR values were computed by Angstrom and Prescott equation, as the best empirical equation, with RMSE of 1.786, MAE of 1.156, CC of 0.977 and WI of 0.988. Conclusively, results from the current study proved that the SVR provided reasonable trends for GSR modeling at Tabriz synoptic station. Furthermore, MT models with linear equations can be implemented with a high degree of simplicity and acceptable precision in GSR estimation. [ABSTRACT FROM AUTHOR]

Forecasting stream flow is a very importance issue in water resources planning and management. The ability of three soft computing methods, least square support vector machine (LSSVM), fuzzy genetic algorithm (FGA) and M5 model tree (M5T), in forecasting daily and monthly stream flows of poorly gauged mountainous watershed using nearby hydro-meteorological data is investigated in the current study. In the first application, monthly stream flows of Hunza river are forecasted using local stream flow data of Hunza and precipitation and temperature data of nearby station. LSSVM provides slightly better forecasts than the FGA and M5T models. Stream flow and temperature inputs generally give better forecasts compared to other inputs. In the second application, daily stream flows of Hunza river are forecasted using local stream flow data of Hunza and precipitation and temperature data of nearby station. Better results are obtained from the models comprising only stream flow inputs. In general, a better accuracy is obtained from LSSVM models in relative to the FGA and M5T. The results indicate that the monthly and daily stream flows of Hunza can be accurately forecasted by using only nearby climatic data. In the third application, daily stream flows of Hunza river are forecasted using local stream flow and climatic data and the models’ accuracy is slightly increased in relative to the previous applications. LSSVM generally performs superior to the FGA and M5T in forecasting daily stream flow of Hunza river using local stream flow and climatic inputs. [ABSTRACT FROM AUTHOR]

Predicting streamflow values accurately is vitally important for hydrology studies. Two heuristic models, namely, gene expression programming (GEP) and support vector machine (SVM) are used and assessed utilizing data from four stations in China. The k-fold testing for local and external data management scenarios are tested extensively. Results indicate that models with inputs of current and one previous day's streamflow records provided the best accuracy. Both the GEP and SVM models can predict accurate streamflow values with respect to the observed records. GEP performed better than the SVM in all k-fold testing stages with lower skewness and standard deviation values for streamflow records. The test accuracy demonstrated high variations for the local and external k-fold case which proved the necessity of k-fold testing or data scanning procedure in daily streamflow prediction. Daily streamflow of downstream stations was also estimated using the data of upstream stations (external k-fold). The best results were obtained by the models trained using the data from the nearest upstream station. In some cases, the accuracy of the external models was found to be comparable to local models. This suggested the use of external models in streamflow prediction in the case of data scarcity. [ABSTRACT FROM AUTHOR]

Accurate estimation of evapotranspiration (ET) values is of crucial importance in hydrology, agriculture and agro-meteorology issues. The objective of this research was to evaluate the use of evolutionary support vector machine (ESVM) to model daily ET using limited climatic data. For this aim, the most common evolutionary method, genetic algorithm (GA), was used for optimization of SVM variables. For the ESVM, four input combinations of maximum air temperature (Tmax), minimum air temperature (Tmin), wind speed (U2), daily solar radiation (Rs), relative humidity (Rhmean) and mean temperature (Tmean) were tried. Climatic data covering 3-year period of October 2004-October 2007 were obtained from the extremely arid and hot region of Haji Abad located in the northern region of Hormozgan province, Iran. Artificial Neural Network (ANN) as a base model was also applied for evaluating modeling accuracy of the ESVM in estimating ET. The results of the ESVM and ANN models were evaluated by comparing their estimates with the measured lysimetric data. The root mean square error (RMSE), coefficient of efficiency (CE) and the coefficient of determination (R²) were used as comparison criteria. According to the results obtained, the ESVM2 whose input variables are Tmean and Rhmean was selected as the best model in estimating ET. [ABSTRACT FROM AUTHOR]

In the present study, three types of artificial intelligence techniques, least square support vector machine (LSSVM), multivariate adaptive regression splines (MARS) and M5 model tree (M5T) are applied for modeling daily dissolved oxygen (DO) concentration using several water quality variables as inputs. The DO concentration and water quality variables data from three stations operated by the United States Geological Survey (USGS) were used for developing the three models. The water quality data selected consisted of daily measured of water temperature (TE, °C), pH (std. unit), specific conductance (SC, μS/cm) and discharge (DI cfs), are used as inputs to the LSSVM, MARS and M5T models. The three models were applied for each station separately and compared to each other. According to the results obtained, it was found that: ( i ) the DO concentration could be successfully estimated using the three models and ( ii ) the best model among all others differs from one station to another. [ABSTRACT FROM AUTHOR]

This study investigates the ability of wavelet-artificial neural networks (WANN) for the prediction of short-term daily river flow. The WANN model is improved by conjunction of two methods, discrete wavelet transform and artificial neural networks (ANN) based on regression analyses, respectively. The proposed WANN models are applied to the daily flow data of Vanyar station, on the Ajichai River in the northwest region of Iran, and compared with the ANN and support vector machine (SVM) techniques. Mean square error (MSE), mean absolute error (MAE) and correlation coefficient ( R) statistics are used for evaluating precision of the WANN, ANN and SVM models. Comparison results demonstrate that the WANN model performs better than the ANN and SVM models in short-term (1-, 2- and 3-day ahead) daily river flow prediction. [ABSTRACT FROM AUTHOR]

Drought forecasting using standardized metrics of rainfall is a core task in hydrology and water resources management. Standardized Precipitation Index (SPI) is a rainfall-based metric that caters for different time-scales at which the drought occurs, and due to its standardization, is well-suited for forecasting drought at different periods in climatically diverse regions. This study advances drought modelling using multivariate adaptive regression splines (MARS), least square support vector machine (LSSVM), and M5Tree models by forecasting SPI in eastern Australia. MARS model incorporated rainfall as mandatory predictor with month (periodicity), Southern Oscillation Index, Pacific Decadal Oscillation Index and Indian Ocean Dipole, ENSO Modoki and Nino 3.0, 3.4 and 4.0 data added gradually. The performance was evaluated with root mean square error ( RMSE ), mean absolute error ( MAE ), and coefficient of determination ( r 2 ). Best MARS model required different input combinations, where rainfall, sea surface temperature and periodicity were used for all stations, but ENSO Modoki and Pacific Decadal Oscillation indices were not required for Bathurst, Collarenebri and Yamba, and the Southern Oscillation Index was not required for Collarenebri. Inclusion of periodicity increased the r 2 value by 0.5–8.1% and reduced RMSE by 3.0–178.5%. Comparisons showed that MARS superseded the performance of the other counterparts for three out of five stations with lower MAE by 15.0–73.9% and 7.3–42.2%, respectively. For the other stations, M5Tree was better than MARS/LSSVM with lower MAE by 13.8–13.4% and 25.7–52.2%, respectively, and for Bathurst, LSSVM yielded more accurate result. For droughts identified by SPI ≤ − 0.5, accurate forecasts were attained by MARS/M5Tree for Bathurst, Yamba and Peak Hill, whereas for Collarenebri and Barraba, M5Tree was better than LSSVM/MARS. Seasonal analysis revealed disparate results where MARS/M5Tree was better than LSSVM. The results highlight the importance of periodicity in drought forecasting and also ascertains that model accuracy scales with geographic/seasonal factors due to complexity of drought and its relationship with inputs and data attributes that can affect the evolution of drought events. [ABSTRACT FROM AUTHOR]

Pan evaporation ( Ep ) plays important roles in agricultural water resources management. One of the basic challenges is modeling Ep using limited climatic parameters because there are a number of factors affecting the evaporation rate. This study investigated the abilities of six different soft computing methods, multi-layer perceptron (MLP), generalized regression neural network (GRNN), fuzzy genetic (FG), least square support vector machine (LSSVM), multivariate adaptive regression spline (MARS), adaptive neuro-fuzzy inference systems with grid partition (ANFIS-GP), and two regression methods, multiple linear regression (MLR) and Stephens and Stewart model (SS) in predicting monthly Ep . Long-term climatic data at various sites crossing a wide range of climates during 1961–2000 are used for model development and validation. The results showed that the models have different accuracies in different climates and the MLP model performed superior to the other models in predicting monthly Ep at most stations using local input combinations (for example, the MAE (mean absolute errors), RMSE (root mean square errors), and determination coefficient (R 2 ) are 0.314 mm/day, 0.405 mm/day and 0.988, respectively for HEB station), while GRNN model performed better in Tibetan Plateau (MAE, RMSE and R 2 are 0.459 mm/day, 0.592 mm/day and 0.932, respectively). The accuracies of above models ranked as: MLP, GRNN, LSSVM, FG, ANFIS-GP, MARS and MLR. The overall results indicated that the soft computing techniques generally performed better than the regression methods, but MLR and SS models can be more preferred at some climatic zones instead of complex nonlinear models, for example, the BJ (Beijing), CQ (Chongqing) and HK (Haikou) stations. Therefore, it can be concluded that Ep could be successfully predicted using above models in hydrological modeling studies. [ABSTRACT FROM AUTHOR]

Accurate estimation of suspended sediment is important for water resources projects. The accuracy of local weighted linear regression (LWLR) technique is investigated in this study for modeling streamflow-suspended sediment relationship. Daily data from two stations on the Eel River in California were used in the applications. In the first part of the study, the LWLR results were compared with those of the least square support vector machine (LSSVM), artificial neural networks (ANNs) and sediment rating curve (SRC) for modeling sediment data of upstream and downstream stations, separately. Root mean square errors (RMSE), mean absolute errors (MAE) and determination coefficient (R) statistics were used for comparison of the applied models Comparison results indicated that the LWLR model performed better than the LSSVM, ANN and SRC models. Accuracies of the sediment modeling increased by the LWLR model compared with the LSSVM model: 14 % (60 %) and 33 % (42 %) decrease in the RMSE (MAE) values for the upstream and downstream stations, respectively. The second part of the study focused on the comparison of the models in estimating downstream suspended sediment data by using data from both stations. LWLR was found to be better than the LSSVM, ANN and SRC models. The RMSE accuracy of the LSSVM model was increased by 39 % using the LWLR model. [ABSTRACT FROM AUTHOR]

They compared their hybrid model (LSSVM-GSA-EEMD) with other alternative models including artificial neural networks (ANNs), support vector machine (SVM), and LSSVM, and their versions combined with empirical mode decomposition (EMD). This paper presents a discussion of "Improving Prediction Accuracy of Hydrologic Time Series by Least-Squares Support Vector Machine Using Decomposition Reconstruction and Swarm Intelligence" by Wen-jing Niu, Zhong-kai Feng, Yin-shan Xu, Bao-fei Feng, and Yao-wu Min. https://doi.org/10.1061/(ASCE)HE.1943-5584.0002116. [Extracted from the article]

Summary This study investigates the accuracy of least square support vector machine (LSSVM), multivariate adaptive regression splines (MARS) and M5 model tree (M5Tree) in modeling river water pollution. Various combinations of water quality parameters, Free Ammonia (AMM), Total Kjeldahl Nitrogen (TKN), Water Temperature (WT), Total Coliform (TC), Fecal Coliform (FC) and Potential of Hydrogen (pH) monitored at Nizamuddin, Delhi Yamuna River in India were used as inputs to the applied models. Results indicated that the LSSVM and MARS models had almost same accuracy and they performed better than the M5Tree model in modeling monthly chemical oxygen demand (COD). The average root mean square error (RMSE) of the LSSVM and M5Tree models was decreased by 1.47% and 19.1% using MARS model, respectively. Adding TC input to the models did not increase their accuracy in modeling COD while adding FC and pH inputs to the models generally decreased the accuracy. The overall results indicated that the MARS and LSSVM models could be successfully used in estimating monthly river water pollution level by using AMM, TKN and WT parameters as inputs. [ABSTRACT FROM AUTHOR]

Accurate predicting of lake level fluctuations is essential and basic in water resources management for water supply purposes. The predicting of lake level is complicated because of it is affected by nonlinear hydrological processes. This paper applies integrated wavelet and auto regressive moving average (ARMA), adaptive neuro fuzzy inference system (ANFIS) and support vector regression (SVR) models for forecasting monthly lake level fluctuations. First, lake level time series is decomposed into low and high frequency components by using discrete wavelet transform. Then, each component is separately predicted by using ARMA, ANFIS and SVR models. Finally, the predicted components are summed to obtain estimated original lake level time series. The performance of the proposed WSVR (Wavelet-SVR), WANFIS (Wavelet-ANFIS) and WARMA (Wavelet-ARMA) models is compared with single ARMA, SVR and ANFIS models. Results show that the integrated models give better precision in forecasting lake levels in the study region when compared to single models. WSVR model is found to be slightly better than the other integrated models. [ABSTRACT FROM AUTHOR]

Accurate estimation of precipitation is an important issue in water resources engineering, management and planning. The accuracy of four different soft computing methods, adaptive neuro-fuzzy inference system (ANFIS) with grid partition (GP), ANFIS with subtractive clustering (SC), artificial neural networks (ANN) and support vector regression (SVR), is investigated in predicting long-term monthly precipitation without climatic data. The periodicity component, longitude, latitude and altitude data from 50 stations in Iran are used as inputs to the applied models. The ANFIS-GP model is found to perform generally better than the other models in predicting long-term monthly precipitation. The SVR model provides the worst estimates. The maximum correlations are found to be 0.935 and 0.944 for the ANFIS-SC and SVR models in Fasa station, respectively. The highest correlations of the ANFIS-GP and ANN models are found to be 0.964 and 0.977 for the Bam and Tabas (Zabol) stations. The minimum correlations are 0.683 and 0.661 for the ANFIS-GP and SVR models in Urmia station while the ANFIS-SC and ANN models provide the minimum correlations of 0.696 and 0.785 in the Sari and Bandar Lengeh stations, respectively. The comparison results show that the long-term monthly precipitations of any site can be successfully predicted by ANFIS-GP model without any weather data. The monthly and annual precipitations are also mapped and evaluated by using the optimal ANFIS-GP model in the study. The precipitation maps revealed that the highest amounts of precipitation occur in the north, southwestern and west regions, while the lowest values are seen in the east and southeastern parts of the Iran. [ABSTRACT FROM AUTHOR]

Forecasting lake level at various horizons is a critical issue in navigation, water resource planning and catchment management. In this article, multistep ahead predictive models of predicting daily lake levels for three prediction horizons were created. The models were developed using a novel method based on support vector machine (SVM) coupled with firefly algorithm (FA). The FA was applied to estimate the optimal SVM parameters. Daily water-level data from Urmia Lake in northwestern Iran were used to train, test and validate the used technique. The prediction results of the SVM–FA models were compared to the genetic programming (GP) and artificial neural networks (ANNs) models. The experimental results showed that an improvement in the predictive accuracy and capability of generalization can be achieved by the SVM–FA approach in comparison to the GP and ANN in 1 day ahead lake level forecast. Moreover, the findings indicated that the developed SVM–FA models can be used with confidence for further work on formulating a novel model of predictive strategy for lake level prediction. [ABSTRACT FROM AUTHOR]

Summary Pan evaporation (Ep) modeling is an important issue in reservoir management, regional water resources planning and evaluation of drinking-water supplies. The main purpose of this study is to investigate the accuracy of least square support vector machine (LSSVM), multivariate adaptive regression splines (MARS) and M5 Model Tree (M5Tree) in modeling Ep. The first part of the study focused on testing the ability of the LSSVM, MARS and M5Tree models in estimating the Ep data of Mersin and Antalya stations located in Mediterranean Region of Turkey by using cross-validation method. The LSSVM models outperformed the MARS and M5Tree models in estimating Ep of Mersin and Antalya stations with local input and output data. The average root mean square error (RMSE) of the M5Tree and MARS models was decreased by 24–32.1% and 10.8–18.9% using LSSVM models for the Mersin and Antalya stations, respectively. The ability of three different methods was examined in estimation of Ep using input air temperature, solar radiation, relative humidity and wind speed data from nearby station in the second part of the study (cross-station application without local input data). The results showed that the MARS models provided better accuracy than the LSSVM and M5Tree models with respect to RMSE, mean absolute error (MAE) and determination coefficient ( R 2 ) criteria. The average RMSE accuracy of the LSSVM and M5Tree was increased by 3.7% and 16.5% using MARS. In the case of without local input data, the average RMSE accuracy of the LSSVM and M5Tree was respectively increased by 11.4% and 18.4% using MARS. In the third part of the study, the ability of the applied models was examined in Ep estimation using input and output data of nearby station. The results reported that the MARS models performed better than the other models with respect to RMSE, MAE and R 2 criteria. The average RMSE of the LSSVM and M5Tree was respectively decreased by 54% and 3.4% using MARS. The overall results indicated that the LSSVM could be successfully used in estimating Ep by using local input and output data while the MARS model performed better than the LSSVM in the case of without local input and outputs. [ABSTRACT FROM AUTHOR]

The accuracy of a least square support vector machine (LSSVM) in modeling of reference evapotranspiration (ET) was examined in this study. The daily weather data, solar radiation, air temperature, relative humidity and wind speed of two stations, Glendale and Oxnard, in southern district of California, were used as inputs to the LSSVM models to estimate ET obtained using the FAO-56 Penman-Monteith equation. In the first part of the study, LSSVM estimates were compared with those of the following empirical models: Priestley-Taylor, Hargreaves and Ritchie methods. The comparison results indicated that the LSSVM performed better than the empirical models. In the second part of the study, the LSSVM results were compared with those of the conventional feed-forward artificial neural networks (ANN). It was found that the LSSVM models were superior to the ANN in modeling ET process. [ABSTRACT FROM AUTHOR]

Summary: The ability of least square support vector machine (LSSVM) is investigated in this paper for modeling discharge-suspended sediment relationship. The daily stream flow and suspended sediment concentration data from two stations on the Eel River in California were used as case studies. In the first part of the study, the LSSVM was compared with those of the artificial neural networks (ANNs) and sediment rating curve (SRC) in the prediction of upstream and downstream station sediment data, separately. Two different algorithms, Levenberg–Marquardt and Conjugate Gradient, were employed for the ANN applications. For evaluating the ability of the models, root mean square errors, mean absolute errors and determination coefficient statistics were used. Comparison results showed that the LSSVM model was able to produce better results than the ANN models. LSSVM and ANN models were found to be better than the SRC model for the upstream station. For the downstream station, however, SRC model outperformed the LSSVM and ANN models. In the second part of the study, the models were compared to each other in estimation of downstream suspended sediment data by using data from both stations. It was found that the LSSVM model performs slightly better than the ANN models and both models performed much better than the SRC model. [ABSTRACT FROM AUTHOR]

Summary: The accurate estimation of reference evapotranspiration (ETo ) becomes imperative in the planning and management of irrigation practices. The Penman–Monteith FAO 56 (PMF-56) model which incorporates thermodynamic and aerodynamic aspects is recommended for estimating ETo across the world. However, the use of the PMF-56 model is restricted by the unavailability of input climatic variables in many locations and the option is to use simple approaches with limited data requirements. In the current study, the potential of support vector machines (SVM), adaptive neuro-fuzzy inference system (ANFIS), multiple linear regression (MLR) and multiple non-linear regression (MNLR) for estimating ETo were investigated using six input vectors of climatic data in a semi-arid highland environment in Iran. In addition, four temperature-based and eight radiation-based ETo equations were tested against the PMF-56 model. The accuracies of the models were evaluated by using three commonly used criteria: root mean square error (RMSE), mean absolute error (MAE) and correlation coefficient (r). The results obtained with the SVM and ANFIS models for ETo estimation were better than those achieved using the regression and climate based models and confirmed the ability of these techniques to provide useful tools in ETo modeling in semi-arid environments. Based on the comparison of the overall performances, it was found that the SVM6 and ANFIS6 models which require mean air temperature, relative humidity, wind speed and solar radiation input variables had the best accuracy. [Copyright &y& Elsevier]

Abstract: A new wavelet-support vector machine conjunction model for daily precipitation forecast is proposed in this study. The conjunction method combining two methods, discrete wavelet transform and support vector machine, is compared with the single support vector machine for one-day-ahead precipitation forecasting. Daily precipitation data from Izmir and Afyon stations in Turkey are used in the study. The root mean square errors (RMSE), mean absolute errors (MAE), and correlation coefficient (R) statistics are used for the comparing criteria. The comparison results indicate that the conjunction method could increase the forecast accuracy and perform better than the single support vector machine. For the Izmir and Afyon stations, it is found that the conjunction models with RMSE=46.5mm, MAE=13.6mm, R=0.782 and RMSE=21.4mm, MAE=9.0mm, R=0.815 in test period is superior in forecasting daily precipitations than the best accurate support vector regression models with RMSE=71.6mm, MAE=19.6mm, R=0.276 and RMSE=38.7mm, MAE=14.2mm, R=0.103, respectively. The ANN method was also employed for the same data set and found that there is a slight difference between ANN and SVR methods. [Copyright &y& Elsevier]

Forecasting intermittent streamflows is an important issue for water quality management, water supplies, hydropower and irrigation systems. This paper compares the accuracy of several data driven techniques, that is, adaptive neuro fuzzy inference system (ANFIS), artificial neural networks (ANNs) and support vector machine (SVM) for forecasting daily intermittent streamflows. The results are also compared with those of the local linear regression (LLR) and the dynamic local linear regression (DLLR). Intermittent streamflow data from two stations, Uzunkopru and Babaeski, in Thrace region located in north-western Turkey are used in the study. The root mean square error and correlation coefficient were used as comparison criteria. The comparison results indicated that the ANFIS, ANN and SVM models performed better than the LLR and DLLR models in forecasting daily intermittent streamflows. The ANN and ANFIS gave the best forecasts for the Uzunkopru and Babaeski stations, respectively. [ABSTRACT FROM AUTHOR]

Summary: The study investigates the accuracy of wavelet and support vector machine conjunction model in monthly streamflow forecasting. The conjunction method is obtained by combining two methods, discrete wavelet transform and support vector machine, and compared with the single support vector machine. Monthly flow data from two stations, Gerdelli Station on Canakdere River and Isakoy Station on Goksudere River, in Eastern Black Sea region of Turkey are used in the study. The root mean square error (RMSE), mean absolute error (MAE) and correlation coefficient (R) statistics are used for the comparing criteria. The comparison of results reveals that the conjunction model could increase the forecast accuracy of the support vector machine model in monthly streamflow forecasting. For the Gerdelli and Isakoy stations, it is found that the conjunction models with RMSE=13.9m3/s, MAE=8.14m3/s, R =0.700 and RMSE=8.43m3/s, MAE=5.62m3/s, R =0.768 in test period is superior in forecasting monthly streamflows than the most accurate support vector regression models with RMSE=15.7m3/s, MAE=10m3/s, R =0.590 and RMSE=11.6m3/s, MAE=7.74m3/s, R =0.525, respectively. [Copyright &y& Elsevier]

The study investigates the accuracy of linear genetic programming (LGP), which is an extension to genetic programming (GP) technique, in daily reference evapotranspiration (ET0) modeling. The daily climatic data, solar radiation, air temperature, relative humidity, and wind speed from three stations, Windsor, Oakville, and Santa Rosa, in central California, are used as inputs to the LGP to estimate ET0 obtained using the FAO-56 Penman-Monteith equation. The accuracy of the LGP is compared with those of the support vector regression (SVR), artificial neural network (ANN), and those of the following empirical models: the California irrigation management system Penman, Hargreaves, Ritchie, and Turc methods. The root-mean-square errors, mean-absolute errors, and determination coefficient (R2) statistics are used for evaluating the accuracy of the models. Based on the comparison results, the LGP is found to be superior alternative to the SVR and ANN techniques. [ABSTRACT FROM AUTHOR]

Summary: This study compares the potential of two different data-driven techniques, support vector machines (SVM) and artificial neural network (ANN) in modeling lake level fluctuations. The SVM method which is a new regression procedure in water resources is applied to the monthly level data of Lake Van which is the biggest lake in Turkey and Lake Egirdir. The estimated lake levels are found to be in good agreement with the corresponding observed values. The results of the SVM based models are compared with those of the ANN. The mean square errors, mean absolute relative errors and determination coefficient statistics are used as comparison criteria. Based on the comparison, it is found that the SVM based model performs better than the ANN. [Copyright &y& Elsevier]

Snow is one of the essential factors in hydrology, freshwater resources, irrigation, travel, pastimes, floods, avalanches, and vegetation. In this study, the snow cover of the northern and southern slopes of Alborz Mountains in Iran was investigated by considering two issues: (1) Estimating the snow cover area and the (2) effects of droughts on snow cover. The snow cover data were monitored by images obtained from the Moderate Resolution Imaging Spectroradiometer (MODIS) sensor. The meteorological data (including the precipitation, minimum and maximum temperature, global solar radiation, relative humidity, and wind velocity) were prepared by a combination of National Centers for Environmental Prediction-Climate Forecast System Reanalysis (NCEP-CFSR) points and meteorological stations. The data scale was monthly and belonged to the 2000–2014 period. In the first part of the study, snow cover estimation was conducted by Multiple Linear Regression (MLR), Least Square Support Vector Machine (LSSVM), Group Method of Data Handling (GMDH), Multilayer Perceptron (MLP), and MLP with Grey Wolf Optimization (MLP-GWO) models. The most accurate estimations were produced by the MLP-GWO and GMDH models. The models produced better snow cover estimations for the northern slope compared to the southern slope. The GWO improved the MLP's accuracy by 10.7%. In the second part, seven drought indices, including the Palmer Drought Severity Index (PDSI), Bahlme–Mooley Drought Index (BMDI), Standardized Precipitation Index (SPI), Multivariate Standardized Precipitation Index (MSPI), Modified Standardized Precipitation Index (SPImod), Joint Deficit Index (JDI), and Standardized Precipitation-Evapotranspiration Index (SPEI) were calculated for both slopes. The results showed that the effects of a drought event on the snow cover area would remain up to 5 (or 6) months in the region. The highest impact of drought appears after two months in the snow cover area, and the drought index most related to snow cover variations is the 2–month time window of SPI (SPI2). The results of both subjects were promising and the methods can be examined in other snowy areas of the world. [ABSTRACT FROM AUTHOR]

Establishing robust models for predicting precipitation processes can yield a significant aspect for many applications in water resource engineering and environmental prospective. In particular, understanding precipitation phenomena is crucial for managing the effects of flooding in watersheds. In this research, a regional precipitation pattern modeling was undertaken using three intelligent predictive models incorporating artificial neural network (ANN), support vector machine (SVM) and random forest (RF) methods. The modeling was carried out using monthly time scale precipitation information in a semi-arid environment located in Iraq. Twenty weather stations covering the entire region were used to construct the predictive models. At the initial stage, the region was divided into three climatic districts based on documented research. Initially, modeling was carried out for each district using historical information from regionally distributed meteorological stations for calibration. Subsequently, cross-station modeling was undertaken for each district using precipitation data from other districts. The study demonstrated that cross-station modeling was an effective means of predicting the spatial distribution of precipitation in watersheds with limited meteorological data. [ABSTRACT FROM AUTHOR]

Chlorination, the basic treatment utilized for drinking water sources, is widely used for water disinfection and pathogen elimination in water distribution networks. Thereafter, the proper prediction of chlorine consumption is of great importance in water distribution network performance. In this respect, data mining techniques—which have the ability to discover the relationship between dependent variable(s) and independent variables—can be considered as alternative approaches in comparison to conventional methods (e.g., numerical methods). This study examines the applicability of three key methods, based on the data mining approach, for predicting chlorine levels in four water distribution networks. ANNs (artificial neural networks, including the multi-layer perceptron neural network, MLPNN, and radial basis function neural network, RBFNN), SVM (support vector machine), and CART (classification and regression tree) methods were used to estimate the concentration of residual chlorine in distribution networks for three villages in Kerman Province, Iran. Produced water (flow), chlorine consumption, and residual chlorine were collected daily for 3 years. An assessment of the studied models using several statistical criteria (NSC, RMSE, R2, and SEP) indicated that, in general, MLPNN has the greatest capability for predicting chlorine levels followed by CART, SVM, and RBF-ANN. Weaker performance of the data-driven methods in the water distribution networks, in some cases, could be attributed to improper chlorination management rather than the methods’ capability. [ABSTRACT FROM AUTHOR]

The article offers the author's insights on the study "Monthly Mean Streamflow Prediction Based on Bat Algorithm-Support Vector Machine," by Bing Xing and colleagues which discusses the use of bat algorithm-based support vector machine (BA-SVM) for predicting streamflows of the Yangtze River in China. Topics discussed include the comparison of results from BA-SVM with artificial neural networks (ANNs), the use of precipitation value as input, and the weighted mean absolute percentage error.

Soil bulk density (BD) is very important factor in land drainage and reclamation, irrigation scheduling (for estimating the soil volumetric water content), and assessing soil carbon and nutrient stock as well as determining the pollutant mass balance in soils. Numerous pedotransfer functions have been suggested so far to relate the soil BD values to soil parameters (e.g. soil separates, carbon content, etc). The present paper aims at simulating soil BD using easily measured soil variables through heuristic gene expression programming (GEP), neural networks (NN), random forest (RF), support vector machine (SVM), and boosted regression trees (BT) techniques. The statistical Gamma test was utilized to identify the most influential soil parameters on BD. The applied models were assessed through k-fold testing where all the available data patterns were involved in the both training and testing stages, which provide an accurate assessment of the models accuracy. Some existing pedotransfer functions were also applied and compared with the heuristic models. The obtained results revealed that the heuristic GEP model outperformed the other applied models globally and per test stage. Nevertheless, the performance accuracy of the applied heuristic models was much better than those of the applied pedotransfer functions. Using k-fold testing provides a more-in-detail judgment of the models. [ABSTRACT FROM AUTHOR]