Razi University

   Abstract: Objective: The main objective of this study is the field investigation and numerical simulation of the hydraulic behavior of the main canal equipped with gates, lateral canals, and a pumping station in the Dasht-e Abbas irrigation network. This aims to evaluate system performance, analyze the factors affecting the inflow to the network and operational conditions, and derive a suitable stage-discharge relationship considering the specific flow conditions in the main canal. Methodology: To acquire the required data for the numerical model, field surveys were conducted within the study area. In this phase, channel geometric data, including cross-sections, specifications, vegetation conditions, and the locations of hydraulic structures, were recorded. Flow simulation in the channel was performed using HEC-RAS software. The calibration process was carried out by comparing simulated water surface elevations with measured ones. To achieve this, Manning’s roughness coefficient and radial gate discharge coefficients were adjusted iteratively and systematically to obtain the best agreement between the model results and field data. Additionally, a genetic programming model was utilized for modeling. To evaluate the performance of the models, statistical indices including the coefficient of determination (R2R^2R2), Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), Mean Absolute Error (MAE), Root Mean Square Relative Error (RMSRE), and Nash-Sutcliffe Efficiency (NSE) were employed. Finally, to examine the conformity of the actual flow pattern with the simulated model, dye tracer injection was used at sensitive and flow-complex points, such as the end of the canal and the locations of radial and Neerpick gates. Findings: Measured data collected over a period of two years across 34 cases (under various discharges and operational conditions) were used for the calibration and validation of the numerical model. Furthermore, flow simulation was conducted using HEC-RAS in 1D for at least 216 cases covering all probable scenarios, and in 2D for 34 cases, the results of which were subsequently used for genetic programming modeling. The flow behavior under different hydraulic conditions in both observational and simulated states was then compared. Conclusion: The results indicated that hydraulic behavior is not solely a function of discharge; rather, the interaction of discharge, channel geometry, roughness, boundary conditions, gate operation, lateral canal abstractions, and pumping station performance play significant roles in determining flow conditions. Therefore, the simultaneous use of field measurements, numerical modeling, and data-driven methods can provide a more precise understanding of such a system’s performance. A comparison between 1D and 2D model performances showed that both models are capable of adequately simulating flow depth; however, in regions experiencing severe lateral flow variations, flow separation, eddies, and recirculation, the 2D model exhibits a superior capability in representing actual flow behavior. Keywords: Genetic Programming, Dasht-e Abbas Irrigation Network, Numerical Simulation, Stage-Discharge Relationship, HEC-RAS.