• Journal of Korea Water Resources Association
• /
• v.57 no.1
• /
• pp.35-44
• /
• 2024

Research in dam inflow prediction has actively explored the utilization of data-driven machine learning and deep learning (ML&DL) tools across diverse domains. Enhancing not just the inherent model performance but also accounting for model characteristics and preprocessing data are crucial elements for precise dam inflow prediction. Particularly, existing rainfall data, derived from snowfall amounts through heating facilities, introduces distortions in the correlation between snow accumulation and rainfall, especially in dam basins influenced by snow accumulation, such as Soyang Dam. This study focuses on the preprocessing of rainfall data essential for the application of ML&DL models in predicting dam inflow in basins affected by snow accumulation. This is vital to address phenomena like reduced outflow during winter due to low snowfall and increased outflow during spring despite minimal or no rain, both of which are physical occurrences. Three machine learning models (SVM, RF, LGBM) and two deep learning models (LSTM, TCN) were built by combining rainfall and inflow series. With optimal hyperparameter tuning, the appropriate model was selected, resulting in a high level of predictive performance with NSE ranging from 0.842 to 0.894. Moreover, to generate rainfall correction data considering snow accumulation, a simulated snow accumulation algorithm was developed. Applying this correction to machine learning and deep learning models yielded NSE values ranging from 0.841 to 0.896, indicating a similarly high level of predictive performance compared to the pre-snow accumulation application. Notably, during the snow accumulation period, adjusting rainfall during the training phase was observed to lead to a more accurate simulation of observed inflow when predicted. This underscores the importance of thoughtful data preprocessing, taking into account physical factors such as snowfall and snowmelt, in constructing data models.

• Journal of Korean Society of Environmental Engineers
• /
• v.31 no.12
• /
• pp.1089-1094
• /
• 2009

The SWMM (Storm Water Management Model) was applied to Princeton University campus, USA to predict the change of the runoff characteristics. Topography and infra structure of urban area are used in detail and watershed is made as form of regular square to improve the efficiency of data. Princeton campus was divided into 131 sub-basins and model input parameters were obtained from DEM (Digital Elevation Model), land use type, and campus management map, etc.. The model was validated based on the measured meteorological data. The validated model was used to analyze the change of the runoff characteristics according to urbanization, which are two different scenarios: 50% and 100% increase of impervious area. The increase of impervious area causes the increase of runoff, especially in the first-flush.

• Journal of the Korean Society of Hazard Mitigation
• /
• v.7 no.1 s.24
• /
• pp.67-72
• /
• 2007

A model for predicting potentials of land erosion and deposition over a natural basin was developed based on the mass balance principle. The program was developed based on sediment mass balance principle for each cell in a GIS. Sediment yield from a cell was estimated with RUSLE. The outflow sediment from a cell was calculated by multiplying the sediment yield of the cell by the sediment delivery ratio (SDR) of the cell. The outflow sediment from the upstream cell becomes the incoming sediment of the downstream cell. Therefore the erosion and deposition potential of each cell could be determined from the sediment mass balance i.e., the difference between the incoming and outflow of sediments of each cell. The developed model was validated by comparing the predicted sediment yields for three basins with measured data.

• Land and Housing Review
• /
• v.8 no.4
• /
• pp.249-256
• /
• 2017

Urbanization can be remarkable affected flood, pollutant loading, ecological system, and green infrastructure by distortion of hydrologic cycle. In order to mitigate these problems in urban, Low Impact Development(LID) technique has been introduced and applied in the world. SWMM model was calibrated with sets of field monitoring data and applied for calculation of runoff and pollutant loading in Asan-tangjung LID city under 2016 rainfall. Runoff reduction of watershed and catchment basins were showed efficiency 12.2% and 62.0%, respectively. Reduction of COD and TP loading also high efficiency in catchment basins were evaluated 74.9 and 71.4%. The results of this study can be used effectively in decision making processes of urban development project by comparing watershed runoff and pollutant reduction by designs of sort of LID technique, LID volume and location.

• Journal of Korea Water Resources Association
• /
• v.55 no.12
• /
• pp.1115-1124
• /
• 2022

The impact of climate change on water resources was evaluated for Chungju Dam and Soyang-gang Dam basins by constructing an integrated modeling framework consisting of a dam inflow prediction model based on the Variable Infiltration Capacity (VIC) model, a distributed hydrologic model, and an LSTM based dam outflow prediction model. Considering the uncertainty of future climate data, four models of CMIP6 GCM were used as input data of VIC model for future period (2021-2100). As a result of applying future climate data, the average inflow for period increased as the future progressed, and the inflow in the far future (2070-2100) increased by up to 22% compared to that of the observation period (1986-2020). The minimum value of dam discharge lasting 4~50 days was significantly lower than the observed value. This indicates that droughts may occur over a longer period than observed in the past, meaning that citizens of Seoul metropolitan areas may experience severe water shortages due to future droughts. In addition, compared to the near and middle futures, the change in water storage has occurred rapidly in the far future, suggesting that the difficulties of water resource management may increase.

• Journal of Korean Society of Environmental Engineers
• /
• v.27 no.12
• /
• pp.1340-1346
• /
• 2005

A water quality modeling study was performed for Chunggye stream during combined sewer overflow(CSO) period, utilizing the diagnostic system for water management in small watershed, CREEK-1(Cyber River for Environment and Economy in Korea). This system integrated geogaphic information system, data base, landscape ecological model(FRAGSTATS), watershed model(SWMM), water quality model (WASP5), and computer graphic. In this study, the watershed model and water quality model were extensively utilized so as to simulate water qualities and flow in Chunggye stream during wet periods. The Chunggye stream watershed was divided into 18 sub-basins in the watershed model and the stream reach into 11 segments in the water quality model. The watershed model was validated against field measurements of BOD, TN, TP, and flow at the downstream location, where the model results showed a reasonable agreement with the field measurements at all parameters. From this study, it was shown that the stream water quality would change along with elapsed time from rainfall start as well as rainfall intensity. The model results indicated that the water quality would significantly upgrade due to the first flush and high sewage ratio of CSO at the beginning of rainfall event, but become degraded along with the runoff increase due to dilution effect.

• Journal of Korea Water Resources Association
• /
• v.33 no.6
• /
• pp.775-782
• /
• 2000

Water quality models can be applied to manage the regional water quality problems and to estimate the target and allowable pollution load in watershed effectively. Since the models such as QUAL2E, W ASP5 and HSPF need many data and are not easily applied in real systems, the water quality model, which would be simple and easily applicable, has been required. Thus, an Analytical Model of BOD(AMB) considering travel time of pollutant was developed for the total pollution load regulation in drainage basin. It was found that the main stream length of a sub-basin in the AMB should be shorter than 7km and the length of distributed pollutant load should also be shorter than 3.5km in a sub-basin. The basin in the AMB could be divided into sub-basins with almost same hydraulic characteristics and reaction rate constant satisfying the proposed stream length. The running results of the AMB in a small stream were very close to the results of QUAL2E, which is widely used one in the world. Therefore, the AlVIB can be used to regulate the total pollution load in drainage district by local government.rnment.

• The Journal of Engineering Geology
• /
• v.28 no.3
• /
• pp.397-409
• /
• 2018

Izumi sedimentary basin (ISB), west of Shikoku, Japan, is widely distributed across the western side of the Sakuragi Bend of the Japan Median Tectonic Line (MTL). It is not obvious how the ISB formed, but this feature is similar to an asymmetric pull-apart basin. The stratigraphic succession and tuff layers show that ages tend to decrease toward the Sakuragi Bend. We investigate whether the ISB is an asymmetric pull-apart basin using analogue model experiments with running sand. A pull-apart basin of length 60 cm and width 20 cm is formed, and secondary normal faults appear on the surrounding surface. A cross-section parallel to the direction of displacement shows that the stratigraphic succession of the pull-apart basin becomes younger toward the releasing bend. A listric normal fault, which has the opposite dip to the master fault, is observed in a cross-section perpendicular to the direction of displacement. These results are consistent with the observed properties of the ISB west of Shikoku, thereby supporting the possibility that the ISB is an asymmetric pull-apart basin.

• Journal of the Korean Geographical Society
• /
• v.44 no.1
• /
• pp.31-55
• /
• 2009

This research aims 1) to characterize the spatial distribution of low relief landforms (plains) via analyses of a Digital Elevation Model (DEM), 2) to classify plains according to morphological and genetic similarity, and 3) to develop a model to explain forming processes of plains in the Korean peninsula. Plains can easily be separated from high relief mountaneous areas by analyzing the DEM. The overall morphological and locational characteristics of plains can be categorized into lava plains, fluvial-marine plains, erosional plains, intermontane basins, and higher ground plains. It is concluded that the characteristic of each plain type is decided by base-level changes caused by tectonic uplift and sea-level changes, and topological relationship of different rock types. Different plain types do not exist independently, but connected with each others along stream networks. The model developed is able to combine the morphological characteristics of plains with the channel network to conceptualize characteristics and development pathways of plains in the Korean Peninsula.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.39 no.2
• /
• pp.74-85
• /
• 1997

This study aims at the development of a mathematical approach for the optimal water allocation in the river basin where available water is not in sufficient. Its optimal allocation model is determined from the comparison and analysis of mathematical programming techniques such as transportation programming and dynamic programming models at its optimal allocation models. The water allocation system used in this study is designed to be the optimal water allocation which can satisfy the water deficit in each district through inter-basin water transfer between Kumho river basin which is a tributary catchment of Nakdong river basin, and the adjacent Hyungsan river basin, Milyang river basin and Nakdong upstream river basin. A general rule of water allocation is obtained for each district in the basins as the result of analysis of the optimal water allocation in the water allocation system. Also a comparison of the developed models proves that there is no big difference between the models Therefore transportation programming model indicates most adequate to the complex water allocation system in terms of its characteristics It can be seen, however, that dynamic programming model shows water allocation effect which produces greater net benefit more or less.