This study develops a novel explainable stacking ensemble model that combines the stacked generalization ensemble method with SHapley Additive exPlanations (SHAP) to enhance the prediction and interpretation of gully erosion susceptibility. Applied to Jefferson County, Illinois, our approach leverages Random Forest (RF), Gradient Boosting Machine (GBM), Logistic Regression (LR), and Deep Neural Networks (DNN) as both base and meta-learners in various configurations, resulting in 44 distinct stacking models. The comparative analysis demonstrated the superior predictive performance of the stacked models when evaluated at 200 randomly gully sites selected points based on LiDAR difference observations; all but three exceeded the highest area under the curve (AUC) value of 0.86 achieved by the best-performing base model (GBM). The LR stacking model, combining RF and GBM as base models with LR as the meta-learner, emerged as the most effective, achieving an AUC of 0.916. The resulting gully erosion susceptibility map by the LR stacking model classified 33 % of the agricultural land (89,208 ha) as the "very high" class, compared to 27 %, 87 %, 27 %, and 55 % predicted by individual RF, LR, GBM, and DNN models, respectively. Crucially, SHAP analysis elucidated how changes in feature values influence model behavior, considering feature interactions within both the base models and the meta-learner. The SHAP identified the annual leaf area index (LAI) as the most influential feature in both RF and GBM base models. Additionally, it highlights the significance of the GBM model in comparison to the RF base model in the final decision-making process of the stacking model. By offering a transparent mechanism to evaluate how different features and models contribute to final decisions, this approach can be extended to broader environmental management and policy-making contexts, facilitating more informed and responsible resource allocation.

The impact of the channel geometry on water quantity and quality simulation of the Soil and Water Assessment Tool (SWAT) was evaluated for the Andong Dam watershed. The new equations to determine the bankfull width of the channels and the bottom width of the floodplains were developed using aerial photographs, and its performance was compared with the current equations of SWAT. The new equations were more exact than the current equations since the current equations tended to overestimate the widths of the channel and floodplain. When compared with the observed data, the streamflow of the scenario 2 (S2, applying the new equations) showed lower deviation and higher accuracy than scenario 1 (S1, applying the current equations) because the peak flow of S2 captured the observed data better due to the impact of the change geometry. Moreover, the water quality results of S2 outperformed S1 regarding suspended solid, total nitrogen, and dissolved oxygen. This is attributed to the variables, such as flow travel time, which is directly related to the channel geometry. Additionally, SWAT was modified to consider the various channel cross-sectional shapes. The results of this study suggest that the channel geometry information for the water quantity and quality estimation should be carefully applied, which could improve the model performance regarding streamflow and water quality simulations.

The South Korean government has recently focused on environmental protection efforts to improve water quality which has been degraded by nonpoint sources of water pollution from runoff. In order to take care of environmental issues, many physically-based models have been used. However, the physically-based models take a large amount of work to carry out site simulations, and there is a need to find faster and more efficient approaches. For an alternative approach for sediment management using the physically-based models, the machine learning-based models were used for estimating sediment trapping efficiency of vegetative filter strips. The seven nonlinear regression algorithms of machine learning models (e.g., decision tree, multilayer perceptron, k-nearest neighbors, support vector machine, random forest, AdaBoost and gradient boosting) were applied to select the model which best estimates the sediment trapping efficiency of vegetative filter strips. The sediment trapping efficiencies calculated by the machine learning models showed similar results as those of vegetative filter strip modeling system (VFSMOD-W) model. As a result of the accuracy evaluation among the seven machine learning models, the multilayer perceptron model-derived the best fit with VFSMOD-W model. It is expected that the sediment trapping efficiency of the vegetative filter strips in various cases in agricultural fields in South Korea can be predicted easier, faster and accurately by the machine learning models developed in this study. Machine learning models can be used to evaluate sediment trapping efficiency without complicated physically-based model design and high computational cost. Therefore, decision makers can maximize the quality of their outputs by minimizing their efforts in the decision-making process.

Soil erosion due to rainfall is a critical environmental issue in North Korea, exacerbated by deforestation and climate change. This study aims to estimate rainfall erosivity (RE) in North Korea using automated machine learning (AutoML), with a particular focus on regional soil erosion risks. North Korean data were sourced from the European Centre for Medium-Range Weather Forecasts (ECMWF) ReAnalysis 5 dataset, while South Korean data were obtained from the Korea Meteorological Administration. Data from 50 stations in South Korea (2013–2019) and 27 stations in North Korea (1980–2020) were used. The GradientBoostingRegressor (GBR) model, optimized using the Tree-based Pipeline Optimization Tool (TPOT), was trained on South Korean data. The model’s performance was evaluated using metrics such as the root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R²), achieving high predictive accuracy across eight stations in South Korea. Using the optimized model, RE in North Korea was estimated, and the spatial distribution of RE was analyzed using the Kriging interpolation. Results reveal significant regional variability, with the southern and western areas displaying the highest erosivity. These findings provide valuable insights into soil erosion management and the development of sustainable agricultural and environmental strategies in North Korea.

As water scarcity intensifies due to climate change, improving the efficiency of agricultural water use has become increasingly critical. Current irrigation systems often experience significant water losses, especially in paddy fields in South Korea that largely rely on open-channel water supply networks. However, previous studies have simulated irrigation improvement strategies by aggregating multiple paddy fields into larger unites rather than modeling them individually, which limits the accurate representation of field conditions. To address this limitation, we applied a granular simulation approach, collecting detailed input data through field surveys. Using this granular model, the study evaluated strategies to enhance irrigation efficiency in paddy fields serviced by the Baekma Agricultural Reservoir in South Korea. We assessed four scenarios: the current open-channel system, conversion of open channels to closed conduits, installation of farm ponds, and a combination of closed conduits and farm ponds. These scenarios were simulated using the Strom Water Management Model (SWMM), meticulously configured to represent individual paddy fields and channel networks. The results show that converting open channels to closed conduits increased irrigation efficiency by 5.4% compared to the current open-channel system. Combining closed conduits with farm ponds achieved the highest efficiency, although the independent effect of farm ponds was minimal. These findings suggested that converting open-channels to closed conduits was a highly effective solution for reducing water losses, while farm ponds played a limited role. This study provides valuable insights into the development of precise irrigation strategies, offering a detailed assessment of real-world conditions in paddy fields.

The increasing frequency of extreme rainfall events poses significant challenges to sustainable water resource management, leading to severe natural disasters. To mitigate these challenges, understanding the hydrological characteristics of watersheds, especially baseflow, is critical for enhancing watershed resilience and supporting sustainable water quality and resource management. However, conventional watershed models often neglect the accurate simulation of baseflow recession. This study proposes a method for calculating and applying the alpha factor for each hydrologic response unit (HRU) in the Soil and Water Assessment Tool (SWAT), considering both temporal and spatial variability in baseflow. The study watershed has undergone significant development, increasing the need for effective water management strategies that promote long-term sustainability. The alpha factor was computed using BFlow2021, and its effectiveness was evaluated by comparing recession and baseflow estimates under different methods. The results indicate that incorporating monthly HRU-specific alpha factors significantly improves model predictions of recession characteristics, highlighting the need for a more spatially and temporally detailed approach in hydrological modeling. The proposed methodology can help clarify the connection between recession and baseflow and can be applied to ungauged stations, offering a valuable tool for sustainable watershed and water quality management.