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.

Rainfall erosivity factor (R-factor) is one of the Universal Soil Loss Equation (USLE) input parameters that account for impacts of rainfall intensity in estimating soil loss. Although many studies have calculated the R-factor using various empirical methods or the USLE method, these methods are time-consuming and require specialized knowledge for the user. The purpose of this study is to develop machine learning models to predict the R-factor faster and more accurately than the previous methods. For this, this study calculated R-factor using 1-min interval rainfall data for improved accuracy of the target value. First, the monthly R-factors were calculated using the USLE calculation method to identify the characteristics of monthly rainfall-runoff induced erosion. In turn, machine learning models were developed to predict the R-factor using the monthly R-factors calculated at 50 sites in Korea as target values. The machine learning algorithms used for this study were Decision Tree, K-Nearest Neighbors, Multilayer Perceptron, Random Forest, Gradient Boosting, eXtreme Gradient Boost, and Deep Neural Network. As a result of the validation with 20% randomly selected data, the Deep Neural Network (DNN), among seven models, showed the greatest prediction accuracy results. The DNN developed in this study was tested for six sites in Korea to demonstrate trained model performance with Nash–Sutcliffe Efficiency (NSE) and the coefficient of determination (R²) of 0.87. This means that our findings show that DNN can be efficiently used to estimate monthly R-factor at the desired site with much less effort and time with total monthly precipitation, maximum daily precipitation, and maximum hourly precipitation data. It will be used not only to calculate soil erosion risk but also to establish soil conservation plans and identify areas at risk of soil disasters by calculating rainfall erosivity factors.

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.

A big data system plays a significant role in various fields. This technology has also been applied to environment fields because it can discover hidden patterns between environmental factors. As the massive data set was constructed for several decades, big data analysis has widely been using for extracting useful information by analyzing different types of big data sets. In this study, we developed a big data system frame to assess the ecosystem service provided from surface soil. Among big data platforms, we used the Amazon Web Service (AWS) due to their cost-efficiency and hardware flexibility. There are five stages of the big data system (i.e. data acquisition– data storage – data processing – data analysis – visualization). In the data acquisition step, the soil sensor and Internet of Things (IoT) system were used, and we collected existing soil properties data provided by national institutes such as Rural Development Administration (RDA), Ministry of Environment (MOE), and Ministry of Land, Infrastructure, and Transport (MOLIT). AWS S3 platform, which is an object storage service and provides easy-to-use management features to users, was accepted as the data storage platform of the big data system. Amazon EMR, Amazon SageMaker, and Amazon QuickSight were used for the step of data processing, data analysis, and visualization of the big data system respectively. We tested that the developed system could predict soil bulk density and able to replace a typical environmental model by using models based on machine learning and deep learning. The results of the two tests showed positive results that the developed models could predict soil properties and simulate natural phenomena as much as the typical environmental model could.  However, since the system is at an early development stage, it needs repetitive tests in the future considering various soil properties. If this system becomes fully functional, the system will be helpful to improve soil environments.

This study is to develop the Web GIS-based surface soil erosion prediction system that informs soil information such as daily potential soil erosion, soil quality, and best management practices (BMPs). The system involves three functions that are: 1) to predict daily potential soil erosion in the study areas (e.g., Jaun-Cheon, Bukhan-Gang, Namhan-Gang, and Gyoungan-Cheon); 2) to provide the current levels of soil qualities at field scale; 3) to recommend BMPs which can improve soil qualities. This study developed a module based on MUSLE and assessed the availability of the module comparing with the measured data at sample fields (3%, 9% slope). After verification of the module, the Web GIS-based system was developed using a user-friendly interface. The users can obtain the visualized soil erosion information through the interface and compare the amount of soil erosion using the single field or multi-fields analysis tool developed in this study. Moreover, the users can find the current level of soil qualities at fields they selected and gain various applicable BMPs information. The system enables to inform non-experts to soil information without using a complex model and equation. Therefore, the system can play a significant role in recognizing the importance of soil resources and enacting laws relative to soil conservation.

Pollutants from the agricultural field are the main cause of water pollution in reach. Especially, in heavy rainfall seasons, non-point source pollutants from agricultural land contaminate stream with nitrogen, phosphorus, and other pollutants. To reduce the components of the contaminations that outflow from the agricultural areas, Best Management Practices (BMPs) have been installed. Integrated factors including weather, geographic characteristics and kind of crops should be considered for choosing proper BMPs in each field. In the fields which have long slope-length, the vegetated ridge is one of the best methods and wildly used method to reduce soil loss. In this study, the Soil & Water Assessment Tool (SWAT) was used to assess the effects of the vegetated ridge on streamflow and sediment within non-point source pollutant management areas. The LS factor in the modified Universal Soil Loss Equation (MUSLE) in SWAT was modified in order to simulate sediment loss reduction by the vegetated ridge in the target fields. This study aims to assess sediment loss reduction by implication of the vegetated ridge using SWAT and to propose the importance of vegetated ridge for reducing non-point source pollutants in agricultural fields. For further research, the development of a vegitated ridge application tool for SWAT will be conducted.