This study aimed to estimate the discharge in ungauged watersheds. To this end, we herein deviated from the model development methodology of previous studies and used convolution neural network (CNN), a deep training algorithm, and hydrological images. As the CNN model was developed for solving classification issues in general, it is unsuitable for simulating the discharge, which is a continuous variable. Therefore, the fully connected layer of the CNN model was improved. Moreover, images reflecting the hydrological conditions rather than a general photograph were used as input data for the CNN model. Three study areas that have discharge gauged data were set for the model’s training and testing. The data from two of the three study areas were used for CNN model training, and the data of the other were used to evaluate model prediction performance. The results of this study demonstrate a moderate predictive success of the discharge of an ungauged watershed using the CNN model and hydrological images. Therefore, it can be suitable as a methodology for the discharge estimation of ungauged watersheds. Simultaneously, it is expected that our methodology can be applied to the field of remote sensing or to the field of real-time discharge simulation using satellite imagery on a global scale or across a wide area.

This study developed a runoff model using a convolution neural network (CNN), which had previously only been used for classification problems, to get away from artificial neural networks (ANNs) that have been extensively used for the development of runoff models, and to secure diversity and demonstrate the suitability of the model. For this model’s input data, photographs typically used in the CNN model could not be used; due to the nature of the study, hydrological images reflecting effects such as watershed conditions and rainfall were required, which posed further difficulties. To address this, the method of a generating hydrological image using the curve number (CN) published by the Soil Conservation Service (SCS) was suggested in this study, and the hydrological images using CN were found to be sufficient as input data for the CNN model. Furthermore, this study was able to present a new application for the CN, which had been used only for estimating runoff. The model was trained and generalized stably overall, and R², which indicates the relationship between the actual and predicted values, was relatively high at 0.82. The Pearson correlation coefficient, Nash–Sutcliffe efficiency (NSE), and root mean square error (RMSE), were 0.87, 0.60, and 16.20 m³/s, respectively, demonstrating a good overall model prediction performance.

The existing methods of river evaluation tend to focus exclusively on water quantity; therefore, they do not provide a suitable methodology for integrated water management. In this study, research was carried out to develop an integrated river evaluation system that can simultaneously consider water quantity and water quality to improve the existing river evaluation methods. To this end, specific indicators were established to evaluate water quantity and water quality; moreover, an integrated evaluation formula was developed to express each indicator as an index. The integrated evaluation formula used additive functions and enabled integrated and comprehensive river evaluation through the sum of each indicator’s indices. The research subjects were rivers in the Paldang watershed, which surrounds important water resources in rep. of Korea. The rivers were analyzed using the study’s integrated river evaluation formula to identify the deteriorated grade of the water quality as well as the water quantity. Finally, the results of the integrated river evaluation rating were found to be poor or very poor. Based on this, the study determined that an integrated river management policy is required to simultaneously consider water quantity and water quality to restore the integrity of the rivers in the Special Countermeasures Area. The existing evaluations of rivers, which had been conducted only with a focus on water quantity, could be judged narrow or incomplete results. Based on this finding, it was also possible to identify an urgent need for a basic river management plan that can consider both water quantity and water quality organically. Ultimately, the study demonstrated that its methodology was able to make highly intuitive judgments about rivers’ current conditions; thus, it can be utilized to generate basic data for the establishment of customized river management policies.

Various factors like climate change and population increase have limited water management evaluation. In South Korea particularly, although the management of water quality and water quantity has recently been integrated, a comprehensive policy has not yet been identified. This study, therefore, aims to propose a methodology for evaluating water social service for 18 basins near major water resources in South Korea. It aims to promote advanced water resource management, secure water equity, and improve inadequate policy implementation. In addition, it proposes a methodology for comprehensive water management evaluation linked with integrated river evaluation with respect to water quality and water quantity. Accordingly, contrary to the common assumption that the entire population has easy access to the supplied water, the status of water service was assessed objectively. The status of water management per sector was also visually represented, through which the vulnerabilities of water management could be intuitively diagnosed. Based on the possibility of utilizing the study results to determine the basic direction for water management, the methodology of this study has been proposed as a tool for establishing an efficient water management policy.

Changes in the pore water pressure of soil are essential factors that affect the movement of structures during and after construction in terms of stability and safety. Soil permeability represents the quantity of water transferred using pore water pressure. However, these changes cannot be easily identified and require considerable time and money. This study predicted and evaluated the soil permeability coefficient using a multiple regression (MR) model, adaptive network-based fuzzy inference system (ANFIS), general deep neural network (DNN) model, and DNN using the dendrite concept (DNN−T, which was proposed in this study). The void ratio, unit weight, and particle size were obtained from 164 undisturbed samples collected from the embankments of reservoirs in South Korea as input variables for the aforementioned models. The data used in this study included seven input variables, and the ratios of the training data to the validation data were randomly extracted, such as 6:4, 7:3, and 8:2, and were used. The analysis results for each model showed a median correlation of r = 0.6 or less and a low model efficiency of Nash–Sutcliffe efficiency (NSE) = 0.35 or less as a result of predicting MR and ANFIS. The DNN and DNN−T both have good performance, with a strong correlation of r = 0.75 or higher. Evidently, the DNN−T performance in terms of r, NSE, and root mean square error (RMSE) improved more than that of the DNN. However, the difference between the mean absolute percent error (MAPE) of DNN−T and the DNN was that the error of the DNN was small (11%). Regarding the ratio of the training data to the verification data, 7:3 and 8:2 showed better results compared to 6:4 for indicators, such as r, NSE, RMSE, and MAPE. We assumed that this phenomenon was caused by the DNN−T thinking layer. This study shows that DNN−T, which changes the structure of the DNN, is an alternative for estimating the soil permeability coefficient in the safety inspection of construction sites and is an excellent methodology that can save time and budget.

The analysis of the local regulation effects is required for sustainable and effective land utilization because land use/land cover (LULC) changes are not only determined by human activity but are also affected by national policy and regulation; however, previous studies for land use/land cover (LULC) have mainly been conducted on the LULC changes using past experience. This study, therefore, analyzed the effects of local regulations aimed at preserving the water quality in South Korea. To this end, changes in LULC were simulated using the CA-Markov model under conditions in which two local regulations, the special countermeasure area (SCA) and total maximum daily load (TMDL), were not applied and examined the differences between the simulated LULC and the actual LULC as of 2018. In addition, the differences in the generation of pollutant loads were driven for Biochemical Oxygen Demand (BOD), Total Nitrogen (TN), and Total Phosphorus (TP) using pollutant unit-load. As a result, without SCA, the agricultural area increased by 379.0 km², the urban area decreased by 101.8 km², and the meadow area decreased by 176.0 km². In addition, without TMDL, the urban area increased by 169.2 km² and the meadow area decreased to 158.8 km².Differences in BOD, TN, and TP pollution loads without SCA applications were shown to decrease to 22,710.5 kg·km⁻² day⁻¹, 1133.9 kg·km⁻² day⁻¹, and 429.8 kg·km⁻² day⁻¹, respectively, while BOD, TN, and TP pollution loads without TMDL applications decreased to 14,435.7 kg·km⁻² day⁻¹, 2543.6 kg·km⁻² day⁻¹, and 368.2 kg·km⁻² day⁻¹, respectively. As such, this study presents a methodology for analyzing the effects of local regulations using the CA-Markov model, which can intuitively and efficiently examine the effects of regulations by predicting LULC changes.