JOURNAL OF HYDRAULICS AND WATER SCIENCE
Year:2024 | Volume:34 | Issue:4|Papers Count:6

Background and ObjectivesIn recent years, the use of artificial intelligence methods, such as artificial neural network models, have become increasingly prevalent in simulating complex natural phenomena, including daily streamflow. The streamflow directly correlates with flood occurrences, and mitigating financial and human losses due to floods is crucial. Accurate streamflow simulation is essential for water resource management and river management. Consequently, in hydrology, deep learning methods have emerged as novel tools to address the longstanding challenge of daily streamflow modeling and are widely used in simulations.Advancements in streamflow modeling with Artificial Intelligence (AI): In recent years, the field of hydrology has witnessed a significant shift toward leveraging AI techniques for streamflow modeling. Among these methods, artificial neural network (ANN) models have gained prominence due to their ability to capture complex relationships within hydrological systems. Streamflow, which represents the flow of water in rivers and streams, is a critical variable for understanding water availability, flood risk, and ecosystem health. By accurately simulating streamflow, researchers and water resource managers can make informed decisions regarding water allocation, flood preparedness, and environmental conservation. Hydrological processes are inherently nonlinear and influenced by various factors such as precipitation, temperature, land cover, and soil properties. Traditional hydrological models often struggle to capture these complexities. However, deep learning methods, including convolutional neural networks (CNNs) and recurrent neural networks (RNNs), offer promising solutions. These models can learn intricate patterns from historical streamflow data, adapt to changing conditions, and provide accurate predictions. As a result, they have become indispensable tools for addressing the longstanding challenge of daily streamflow modeling. Researchers continue to explore novel architectures, data augmentation techniques, and hybrid approaches to enhance the performance and robustness of AI-based streamflow simulations. In summary, the integration of deep learning methods into hydrological research has revolutionized streamflow modeling, enabling more accurate predictions and informed decision-making in water management and flood risk assessment.MethodologyIn this study, we focused on selecting an appropriate input scenario for deep learning models and simulate daily streamflow on the Kashkan River using LSTM (Long Short-Term Memory) and GRU (Gated Recurrent Unit) deep learning methods. Prior to this, deep learning modeling with the GRU approach using native streamflow measurements had not been performed for Kashkan river. The study area is a flood-prone and mountainous region, specifically the western part of Iran, where a hydrological station with a history of flood events is situated on the Kashkan River. We employ four approaches for handling outliers (Mahalanobis, critical interval removal, Z-Score, and no removal) and four different preprocessing techniques for input data to train two models: LSTM and GRU. Ultimately, eight distinct models are generated and validated against historical data. The input features include regional average precipitation, normalized vegetation cover index, surface soil moisture, groundwater flow, and the Kashkan River’s own flow at the hydrological station, with the best features selected using statistical correlation control.FindingsThe results demonstrate that among the deep learning models generated with a 10-day time step, the model with the least error and consistent low error retention in error metrics is observed. Furthermore, the best performance is achieved using different approaches, in the following order: the GRU model with Z-Score-corrected inputs, followed by the Mahalanobis removal approach with average RMSE (Root Mean Square Error) and KGE (Kling-Gupta Efficiency) values of 5.41 and 0.99, respectively, and the critical interval removal approach with RMSE of 6.23 and KGE of 0.7.The results showed that among the deep learning models produced with a time step of 10 days in the model, the lowest amount of error and the persistence of low error can be seen in the error statistics, and among the different approaches used, the best performance is the GRU model with input modified by Z-Score elimination of outlier method, Mahalanobis elimination method with average RMSE and KGE values of 5.41, 0.99, 6.23, and 0.7 in the training phase and 8.17, 0.79, 4.21, and 0.81 in the validation phase and 5.01, 0.68, and 7.21 and 0.52 are in the testing phase. The obtained results do not reject the LSTM method in simulating the river flow, but state that the listed scenarios, especially in the GRU method, have a higher power in dealing with the data and recognizing the complex pattern of daily river flow, taking into account the limitation in use They have seven years of regular daily data, and future research will show how the behavior of GRU and LSTM models will differ if data with higher convergence is used.ConclusionGRU in future studies can make difference by enhanced flood forecasting accuracy, efficient computation and real-time applications, integration with lag time preprocessing, adaptability to changing climate and urbanization. Future studies will be on data driven method in flood prone areas. There remains ample room for future research and innovation. Here are some directions for further exploration: hydrological data fusion, spatially explicit models, uncertainty quantification, climate change resilience.

Background and ObjectivesRegarding water supply, water resource management is very important. Rainfall and floods are among the climate variables that play an important role in water management and agriculture. The inconsistency between the occurrence of the maximum rainfall and the flooding is one of the noteworthy manifestations of changes in land use. Knowing the reasons for the inconsistency between the time interval of rains and extreme floods can potentially be used to advance agricultural programs, water resources management, flood prevention, groundwater feeding, natural resources management, land use, industry and national economy should be important. Any extreme rainfall causes a flood at any subsequent time. In the case of annual analysis of the occurrence of extreme rainfall and floods, valuable information can be obtained from the state of land use or water resources.MethodologyToday, various statistical methods along with efficient software are used to check rainfall and flood data. In the meantime, we can mention the method of directional statistics in MATLAB environment with coding, which has attracted the attention of many experts and researchers. In fact, circular statistics is a branch of statistics that is dedicated to the development of statistics and supports special data such as directional data. Therefore, for the statistical analysis of extreme data or any data that has a time frequency, the use of directional statistics, which is also called circular statistics, is applicable. In this study, Circular statistics have been used to investigate the range of time of occurrence and distribution of threshold data regarding rainfall and runoff from 21 rain gauge stations and 17 hydrometric stations in the Gavkhoni wetland. FindingsThe average index of seasonality (dispersion) of time occurrence in most stations is above 0.6 or (60%) and their variance is less than 0.4. Due to the fact that the rainfall in the region is recorded in the stations, the occurrence values of the maximum rainfall during different years have been almost without manipulation or any human factors (influenced by human factors). Therefore, the time of occurrence of maximum rainfall was mostly in the area of 0 to -π/2, and the highest value of was from -81.80⁰ to -19.65⁰ for Lange and Isfahan stations, respectively. The average seasonality index (dispersion) of time occurrence was calculated from 0.27 at Diziche station to 0.87 at Ghale Shahrokh station. The input in some hydrometric stations has undergone changes in different years, or it may have been blocked or deviated before entering some stations. So the time of maximum runoff in those stations will not coincide with the time of rainfall. Therefore, the occurrence time of extreme floods is mostly scattered, and the highest value of is calculated from 161.25⁰ to -43.63⁰ for Diziche and Heydari stations, respectively. In Gavkhoni wetland basin, 9 out of 17 stations are not affected by the dam and 8 other stations are affected by the dam (hydrometric stations immediately after the reservoir or diversion dam).The Rayleigh test rejects the null hypothesis (non-uniformity of flood occurrence in the perimeter of the circle) in all the stations, except for Diziche station. For this reason, it is not possible to calculate the upper and lower confidence band at Diziche station.ConclusionThe last hydrometric station for draining the runoff to the Gavkhoni lagoon is the Varzane station, which has an average seasonality index and the time of the maximum flood output on the 2nd day of the new year. In general, it is affected by the changes of land use in the catchment area. In almost three seasons, spring, autumn and winter, it has output and runoff extreme data, so it shows that the water resources stored upstream of Varzaneh station are consumed and the remaining sewage is discharged from the station with a very low flow rat and it is transferred to Gavakhuni wetland. Therefore, the reason for the dryness of the wetland is being deprived of the natural source of Zayandeh River and it can be concluded that with directional statistics and histogram curves of rainfall and runoff extreme data, land use changes can be detected compared to rainfall and runoff and from the average seasonality index. And another use of indicates that the water intake valves for agricultural and industrial uses should be closed so that the runoff is transferred to the wetland.

Background and ObjectivesWater shortage is the main challenge for agriculture in arid and semi-arid regions. Therefore, improving water productivity using different methods such as implementing pressurized irrigation systems is necessary. In these systems the both efficiency and distribution coefficients are vital. The purpose of this study is to investigate the performance of flow control valves in regulating the volume of water delivered in the field and saving water consumption. A flow control valve is recommanded to deliver an almost constant flow for different pressure ranges.MethodologyIn this study, the performance of automatic flow control valves to adjust the water delivered in the field was investigated. For this purpose, two valves with flow rates of 5 and 10 Ls-1 were installed on a farm equipped with a drip irrigation system to evaluate the effect of the valves in a field condition. The farm is located in Mahdiabad of Takestan. The area of this field is about 12 ha. Also, EPANET software is used to model the irrigation system in different scenarios. This software can simulate the behavior of water flows in pressurized networks. To model the irrigation system, the specifications of the reservoir, water transmission lines, manifold pipes, and laterals including the length, diameter of the pipes, and elevation, were applied as the input characteristics of the program. Numerical models need to be calibrated to check the correspondence between the measured and simulated parameters. To compare the values measured and simulated by the EPANET model, statistical indices of root mean square error (RMSE), mean absolute error (MBE), and error percentage (NRMSE) were used.FindingsFirstly, the current condition of the farm was evaluated. The results showed the water discharges were 6.4 and 12.35 L s-1 for plots B and F on-farm, respectively. These discharges were 28% and 25.3% higher than the designed discharges for parts B and F, respectively. As a result, water consumption was increased and its surplus was wasted as deep percolation or surface runoff. Then the flow control valves were installed in the suitable places. After installing the flow control valves, the water delivery condition was re-evaluated. Discharges of B and F plots were 4.86 and 9.96 L s-1, respectively. EPANET software was applied to simulate the flow in the irrigation system. The results showed that the flow control valves could be used successfully to deliver an accurate volume of water to the plots and they could compensate the effect of changes in pool water height or ruptures of irrigation tapes. By investigating the effect of changing the height of the pool water, it was found that increasing the height of the water in the pool would result in increasing the irrigation system discharge. However, by installing the flow control valve, water delivery remained almost constant by changing the water height in the pool. Additionally, by examining the performance of the valve during rupture or dislocation of the irrigation tapes, the flow through the tested irrigation fields increased. However, with the use of flow control valves, due to the structural mechanism of them, the flow rate did not exceed from the designed values. Also, the results showed that a dislocation or leakage in the irrigation tapes reduced the pressure and increased the amount of flow consumed by the irrigation system. However, the numerical modeling results showed that with the installation of the automatic control valve, the flow rate of the irrigation tapes remained in a suitable range.Conclusion                                                               Flow control valves regulate discharge irrespective of the pressure fluctuations. Experiments were performed to identify the ability of a discharge control valve to improve water distribution uniformity in an actual field condition. Field measurements revealed the successful application of the control valves on a farm scale. Also, numerical results indicated that the flow control valves could be used effectively to increase the flow uniformity inthe cases of the water height fluctuations in the reservoir pools or  ruptures of the irrigation tapes. In general, it was found that the flow control valve was a good choice to increase the water efficiency and uniformity in the field conditions. Numerical simulations could determine the suitable locations of the valves and their discharge characteristics.

Iran’s agricultural sector faces unique obstacles due to the diversity of its climatic conditions, underscoring the crucial importance of safeguarding crops against the impacts of climate change. One effective strategy to mitigate damage to the agricultural industry is the ability to accurately predict dew point temperature. The dew point refers to the temperature at which water vapor in the air condenses into liquid water or dew, given a constant air pressure. Notably, an excessively high dew point can adversely affect the performance of air conditioning systems and reduce the efficiency of coolant-based ventilation mechanisms. The formation of dew in ecosystems is influenced by a triad of key factors: radiative exchange between the Earth’s surface and the atmosphere, turbulent heat transfer, and vapor pressure. While the dew point is typically measured using a moisture meter, there also exist empirical equations that relate air temperature and humidity. However, reliable dew point forecasting often requires common meteorological parameters, such as relative humidity and precipitation, which are not consistently measured at many weather stations or may be subject to significant error. As such, regression-based estimation methods are frequently employed. Recognizing the importance of data-driven approaches in dew point estimation, this study explores the use of several predictive models, including support vector regression (SVR), the M5P tree model, the M5Rules rule-generation algorithm, Gaussian process regression (GPR), linear regression (LR), random forest (RF), and random tree (RT) models. These models were applied to data collected from two stations in Gorgan and Shahrekord, Iran, to estimate dew point temperature.MethodologyThe Gorgan basin is geographically situated between 25°54’ east longitude and 36°50’ north latitude, while the Shahrekord basin is located between 50°49’ east longitude and 32°20’ north latitude. Topographically, Shahrekord lies in the eastern segment of the Zagros Mountain range, along the Zagros fault margin. The input parameters for this study were obtained from the Iran Meteorological Organization, covering the period from 1990 to 2021. These parameters include daily maximum temperature (Tmax), daily minimum temperature (Tmin), daily average temperature (Tm), sunshine hours (sshn), average wind speed (ffm), average relative humidity (RHm), maximum relative humidity (RHmax), and minimum relative humidity (RHmin). To assess the accuracy of the input parameters and models, the dew point (DP) was extracted from the testing data and evaluated using regression and tree-based models. Additionally, eight potential scenarios were defined to estimate the daily DP.FindingsThe study findings revealed that scenarios 1, 2, and 3 exhibited the highest correlation with dew point temperature, while scenarios 4, 5, 6, 7, and 8 displayed the lowest correlation. However, upon evaluating the scenarios based on the established criteria, it was determined that scenarios 1, 2, and 3 performed less effectively than the other scenarios. Consequently, the placement of the majority of parameters (scenarios 6, 7, and 8) led to a decrease in the models’ errors. The results obtained for the Gorgan models showed that the R-value ranged from 1 to 0.952. In the Gorgan station, the highest RMSE was observed for RT-3 at 2.0307, and the lowest RMSE was for SVR-8 at 0.222. Furthermore, the best fit for SVR-8 was characterized by RMSE: 0.222, NSE: 0.999, MBE: 0.092, MAE: 0.147, WI: 1, and SI: 0.017, while the worst fit was for RT-3 with RMSE: 2.307, NSE: 0.882, MBE: 0.875, MAE: 1.745, WI: 0.971, and SI: 0.179. The results for the Shahrekord models indicated that the R-value ranged between 0.996 and 0.615. In the Shahrekord station, the highest RMSE was observed for RT-2 at 4.952, and the lowest RMSE was for SVR-7 at 0.550. Additionally, the best fit for SVR-7 was characterized by RMSE: 0.550, NSE: 0.989, MBE: 0.374, MAE: 0.346, WI: 0.997, and SI: -0.15, while the worst fit was for RT-2 with RMSE: 4.957, NSE: 0.131, MBE: 1.43, MAE: 3.914, WI: 0.754, and SI: -1.352.ConclusionIn this study, meteorological data were fitted using various modeling techniques, including Gaussian Process Regression (GPR), Linear Regression (LR), M5P, M5Rules, Random Forest (RF), Regression Tree (RT), and Support Vector Regression (SVR), to obtain the dew point temperature at the Gorgan and Shahrekord weather stations. The performance of these models was then evaluated under different scenarios, and the best-performing models were selected for estimating the dew point temperature. The results indicate that the estimation accuracy of the models using a single input parameter, such as minimum temperature (Tmin), was lower compared to the other models in both the Shahrekord and Gorgan stations. The SVR model with scenario 8 and the M5P model with scenario 7 demonstrated the best performance in estimating the dew point temperature for the Gorgan and Shahrekord stations, respectively. Furthermore, the comparison of the selected models revealed that the SVR model had the highest accuracy among the models evaluated. For the Gorgan station, the models were ranked from high to low accuracy as follows: SVR, M5P, M5Rules, GPR, RF, and LR. For the Shahrekord station, the ranking was M5P, M5Rules, GPR, RF, RT, and LR. Additionally, the comparison of the tree-based and regression-based models showed that the regression models, such as SVR and LR, had higher accuracy in estimating the dew point temperature compared to the tree-based models, such as M5P, M5Rules, and RT.

Background and ObjectivesWeirs are the most important structures for measuring and regulating flow rate. They are also simple hydraulic structures used to control the water level and measure the flow rate in canals. Lateral weirs are sometimes known as side weirs. Side weirs are usually made in various geometric shapes such as rectangular, arched, trapezoidal and triangular. A complete analytical solution of the equations governing the side weir discharge is not possible as there are many parameters influencing the flow phenomenon. An accurate computation of lateral discharge mainly depends on the proper estimation of its discharge coefficient. Investigation of the discharge coefficient has been the main focus by many researchers. In trapezoidal side weirs, weir height, hydraulic head, and weir wall angle affect the discharge coefficient. Although many researchers have done studies on the theoretical and practical applications of simple side weirs, there are only limited investigations on compound sharp-crested side weirs. Recently, some experimental works have been done in order to understand flow hydraulics of compound sharp- and broad- crested normal weirs, which are built across the channel. In this study, the effect of the mentioned parameters on the discharge coefficient, hydraulic characteristics of the flow including flow rate, the water surface profile and energy variations were investigated numerically.MethodologyNumerical simulation geometry which is composed of a rectangular channel and installed side weir, created and meshed using GAMBIT (ANSYS) . The experimental domain length is 2 m in order to avoid large mesh numbers. A quad-map mesh was generated for all models. In the present study, 3-dimensional numerical simulation of  flow over  trapezoidal sharp-crested side weir was evaluated using three turbulence models of  standard k-ε, RNG k-ε and  Realizable k-ε. The free surface was determined using the VOF method. The results showed that the RNG k-ε turbulence model and VOF method are suitable for predicting the discharge coefficient in trapezoidal plan side weirs. The studied models for trapezoidal side weirs were meshed using different node values to determine the optimum number of nodes to generate mesh and to perform a mesh independence test , a negligible difference was observed by increasing the number of nodes in simulated and measured the discharge coefficient of flow over trapezoidal side weir. Therefore, a mesh composed of approximately 25000 elements was considered as an optimum mesh for all created models to resolve flow characteristics. The boundary conditions were defined for all models. At the channel inlet and outlet, pressure inlet and outlet boundary conditions were used. For free surface, pressure inlet boundary condition was definedand wall boundary condition was assigned at the channel bed, side walls and the structure of weir. The comparison of the discharge coefficient of flow that data obtained from numerical simulation agreed well with the experimental data. Also results showed that VOF method can simulate free surface variations accurately enough given that the average relative error values of measured and simulated the discharge coefficient were 2 - 6% for all considered turbulence models. For sharp-crested side weirs in subcritical flow conditions, the equation  of  De Marchi (1934)was used to compute  the flow discharge coefficient of the side weirs.FindingIn this study the water surface and flow patterns were analyzed using contour plots at different horizontal and vertical planes.  Using a non-linear regression model, they proposed a dimensionless relationship for prediction of the discharge coefficient in trapezoidal side weirs in subcritical flow conditions. In this study, the effect of weir wall angle, weir height and the hydraulic head on the discharge coefficient  of trapezoidal side weirs was investigated numerically using FLUENT software and also the numerical results were compared with the experimental data. The results of the simulation were in a good agreement with the experimental data. The discharge coefficient (Cm) is not dependent on any single hydraulic or geometric parameter, but several parameters affect it. The results also showed that the side weir with a wall slope of  z = 1 has better performance compared to the other two angles. Because it has the highest amount of the discharge coefficient among different weirs.ConclusionAs a result of the flow passing through the side weir, the main channel's flow rate and longitudinal velocity decrease. It can be concluded that the velocity values near the side weir decrease with a greater slope, and the velocity variations decrease downstream in the main channel.

Abstract Background and ObjectivesOptimum operation of dam reservoirs is one of the most significant management factors in developing the annual resource and consumption plan of dam reservoirs during operation. The decisions regarding amount of water release are made by having the volume of  the reservoir, amount of demand, and the prediction of reservoir inflow in the actual operation of dam reservoirs. Since the volume of release is related to the storage volume of the reservoirs of the dams and should be optimized simultaneously, after introducing the genetic algorithm and the particle swarm algorithm, the performance of these algorithms alone and in combination with each other in the optimal operation of the Alavian dam reservoir are compared with the modeling results in the nonlinear programming and the rule curves of the operation are developed in this study. The performance indicators of the reservoir were been used including reliability, vulnerability and stability  to evaluate the performance of the examined algorithms in the optimal operation of the reservoir. MethodologyIn this study, after introducing the genetic algorithm and the particle swarm algorithm, innovatively examines the accuracy and effectiveness of modeling by comparing the performance of these algorithms both individually and in combination. This comparison focuses on optimizing the operation of the Alavian dam reservoir over multi-step ahead, using modeling results from the software Lingo. To enhance decision-making for improved management of the Alavian dam reservoir, operation rule curves have been developed. The model utilizes a series of 25 years of data from the Alavian dam, which includes the volume of inflow, the volume of release from the reservoir, storage volume, and usage data encompassing drinking, agriculture, industry, and environmental needs. Additionally, information such as the volume of overflow from the dam reservoir and the volume of evaporation from the surface of the Alaviyan Dam reservoir has been collected on a monthly basis. FindingsThe results from these optimal solutions indicate that the combined algorithm outperforms other methods, demonstrating a better correlation with the reservoir management policy. Over the last 25 years, the combined algorithm met 85% of the water requirements for agriculture downstream of Alavian dam, compared to 82% for the Particle swarm optimization(PSO)  algorithm and 78% for the genetic algorithm (GA). In contrast, thenonlinear programming (NLP) method met 80%. The total shortages over the entire 25-year operational period for the GA, PSO, GA-PSO, and NLP algorithms were 38, 33.7, 27.1, and 35.2 million cubic meters, respectively. The GA-PSO algorithm has successfully addressed 10.87 million cubic meters more than the GA algorithm and 6.57 million cubic meters more than the PSO algorithm. ConclusionInvestigating the results obtained from the optimal solutions revealed that the hybrid algorithm model provides a more favorable result and shows a better correlation regarding the reservoir operation policy. The results indicate the high performance of the hybrid algorithm compared to other studied methods in the optimal operation of the single reservoir system of Alavian dam. Accordingly, the optimal parameters of the Alavian dam reservoir were obtained using a hybrid algorithm. It was proposed to  release volume rule curves and reservoir volume for the  multi-step ahead.