Predicting dam inflow is necessary for effective water management. This study created machine learning algorithms to predict the amount of inflow into the Soyang River Dam in South Korea, using weather and dam inflow data for 40 years. A total of six algorithms were used, as follows: decision tree (DT), multilayer perceptron (MLP), random forest (RF), gradient boosting (GB), recurrent neural network–long short-term memory (RNN–LSTM), and convolutional neural network–LSTM (CNN–LSTM). Among these models, the multilayer perceptron model showed the best results in predicting dam inflow, with the Nash–Sutcliffe efficiency (NSE) value of 0.812, root mean squared errors (RMSE) of 77.218 m³/s, mean absolute error (MAE) of 29.034 m³/s, correlation coefficient (R) of 0.924, and determination coefficient (R²) of 0.817. However, when the amount of dam inflow is below 100 m³/s, the ensemble models (random forest and gradient boosting models) performed better than MLP for the prediction of dam inflow. Therefore, two combined machine learning (CombML) models (RF_MLP and GB_MLP) were developed for the prediction of the dam inflow using the ensemble methods (RF and GB) at precipitation below 16 mm, and the MLP at precipitation above 16 mm. The precipitation of 16 mm is the average daily precipitation at the inflow of 100 m³/s or more. The results show the accuracy verification results of NSE 0.857, RMSE 68.417 m³/s, MAE 18.063 m³/s, R 0.927, and R² 0.859 in RF_MLP, and NSE 0.829, RMSE 73.918 m³/s, MAE 18.093 m³/s, R 0.912, and R² 0.831 in GB_MLP, which infers that the combination of the models predicts the dam inflow the most accurately. CombML algorithms showed that it is possible to predict inflow through inflow learning, considering flow characteristics such as flow regimes, by combining several machine learning algorithms.

Recent climate change has intensified extreme rainfall events, exacerbating soil erosion and agricultural nonpoint source pollution in South Korea’s steeply sloped farmlands. This study assessed soil erosion reduction measures by applying individual Best Management Practices (BMPs) in cropland and expanding upon existing management efforts through the implementation of additional BMPs aimed at further reducing soil erosion. Furthermore, priority management areas were identified based on soil erosion reduction efficiency within subbasins. For this evaluation, the Soil and Water Assessment Tool (SWAT) was employed, with a spatially distributed Hydrological Response Unit (SD-HRU) module and calibrated Modified Universal Soil Loss Equation (MUSLE) parameters tailored to Korean watershed conditions. Scenarios 1 and 2 were implemented in the study area to evaluate BMP effectiveness in controlling soil erosion and suspended sediment (SS) loads. Scenario 1 applied a set of BMPs already in place, while Scenario 2 involved the addition of supplementary BMPs to enhance soil erosion control. Scenario 1 resulted in a 34.6% reduction in annual soil erosion and a 35.0% decrease in SS concentration, whereas Scenario 2 achieved a 59.3% reduction in soil erosion and a 57.3% decrease in SS concentration. Subbasin-scale evaluations revealed considerable spatial variability in erosion control efficiency, ranging from 1.3% to 70.5%, highlighting the necessity for spatially targeted management strategies. These results underscore the importance of employing spatially adaptive BMP approaches and offer practical guidance for enhancing watershed sustainability, particularly in regions vulnerable to extreme hydrometeorological events.