Ehsan Afaridegan

The online version contains supplementary material available at 10.1038/s41598-026-50458-6.

1. Scour phenomenon is a complex hydraulic engineering problem because it depends on many different factors, such as water flows, the shape and size of hydraulic structures, and the physical proper...

Accurate estimation of flow discharge, Q, through hydraulic structures such as spillways and gates is of great importance in water resources engineering. Each hydraulic structure, due to its unique characteristics, requires a specific and comprehensive study. In this regard, the present study innovatively focuses on predicting Q through Rectangular Top-Hinged Gates (RTHGs) using advanced Gradient Boosting (GB) models. The GB models evaluated in this study include Categorical Boosting (CatBoost), Histogram-based Gradient Boosting (HistGBoost), Light Gradient Boosting Machine (LightGBoost), Natural Gradient Boosting (NGBoost), and Extreme Gradient Boosting (XGBoost). One of the essential factors in developing artificial intelligence models is the accurate and proper tuning of their hyperparameters. Therefore, four powerful metaheuristic algorithms—Covariance Matrix Adaptation Evolution Strategy (CMA-ES), Sparrow Search Algorithm (SSA), Particle Swarm Optimization (PSO), and Genetic Algorithm (GA)—were evaluated and compared for hyperparameter tuning, using LightGBoost as the baseline model. An assessment of error metrics, convergence speed, stability, and computational cost revealed that SSA achieved the best performance for the hyperparameter optimization of GB models. Consequently, hybrid models combining GB algorithms with SSA were developed to predict Q through RTHGs. Random split was used to divide the dataset into two sets, with 70% for training and 30% for testing. Prediction uncertainty was quantified via Confidence Intervals (CI) and the R-Factor index. CatBoost-SSA produced the most accurate prediction performance among the models (R² = 0.999 training, 0.984 testing), and NGBoost-SSA provided the lowest uncertainty (CI = 0.616, R-Factor = 3.596). The SHapley Additive exPlanations (SHAP) method identified h/B (upstream water depth to channel width ratio) and channel slope, S, as the most influential predictors. Overall, this study confirms the effectiveness of SSA-optimized boosting models for reliable and interpretable hydraulic modeling, offering a robust tool for the design and operation of gated flow control systems.

Dust pollution is a major environmental issue that threatens public health and ecosystems. This study investigates the relationship between wetland area reduction and the intensification of dust pollution in southwestern Iran, utilizing satellite data from Landsat, Sentinel-5, and MODIS (2010–2022). High-resolution imagery from Google Earth Engine (GEE) was used to extract dust event frequency data, while Aerosol Optical Depth (AOD) and Absorbing Aerosol Index (AAI) were employed to monitor dust pollution levels. Normalized Difference Water Index (NDWI) analysis revealed that the Hoor Al-Azim and Shadegan wetlands experienced their driest conditions in 2010, 2011, 2021, and 2022. A strong inverse correlation was observed between wetland area and dust event frequency, with correlation coefficients of − 0.85 for Hoor Al-Azim and − 0.90 for Shadegan. When the area of Hoor Al-Azim dropped below 60 km², dust events increased to 20–35 occurrences. Similarly, in Shadegan, dust events exceeded 20 when the wetland area fell below 100 km². Wind rose analysis indicated that prevailing northeast and northwest winds, combined with dry conditions in northern parts of Shadegan, significantly contribute to dust formation. Furthermore, the study established a clear inverse relationship between wetland area reduction and the number of polluted days for both PM10 and PM2.5. For each 1 km² decrease in wetland area, PM10 polluted days increased by 0.03, explaining 56% of the variation, while PM2.5 polluted days increased by 0.05, accounting for 64%. Principal Component Analysis (PCA) further demonstrated that wetland area changes explain 87.5% and 90.1% of the variance in PM10 and PM2.5 levels, respectively. These findings highlight the critical role of wetland preservation in mitigating dust pollution and improving air quality.

This study presents a rigorous investigation into determining the optimal placement of a drainage well and cutoff wall to effectively mitigate the uplift force and seepage discharge in small concrete dams. A sophisticated numerical model based on the two-dimensional Laplace equation was developed for this purpose. The Laplace equation was discretized using the finite difference method with a second-order central schema, and the resulting system of equations was efficiently solved using the Gauss–Seidel method with an over-relaxation factor of 1.95. The Neumann boundary conditions were applied to the dam body and cutoff wall, while Dirichlet boundary conditions were imposed on the drainage well, as well as the upstream and downstream sections of the dam. The results exhibited an excellent agreement between the numerical simulations and the observed data, with a mean absolute percentage error of 3.54%. The findings from the numerical simulations revealed that the optimal location for the drainage well is at a distance of 0.2L from the upstream face of the dam, where L represents the dam length. This location resulted in a notable reduction of approximately 38% in the uplift force. Additionally, utilizing a cutoff wall at the upstream portion of the dam led to a reduction of about 15% in the uplift force. Remarkably, when both a cutoff wall and a drainage well were employed at their respective optimal locations, the uplift force decreased by an impressive 53%.