• Proceedings of the Korea Water Resources Association Conference
• /
• 1994.02a
• /
• pp.279-284
• /
• 1994

• Journal of Korea Water Resources Association
• /
• v.52 no.12
• /
• pp.963-973
• /
• 2019

This study reviews a urban flooding risk criteria estimation model to predict risk criteria in areas where flood risk criteria are not precalculated by using watershed characteristic data and limit rainfall based on damage history. The risk criteria estimation model was designed using Support Vector Machine, one of the machine learning algorithms. The learning data consisted of regional limit rainfall and watershed characteristic. The learning data were applied to the SVM algorithm after normalization. We calculated the mean absolute error and standard deviation using Leave-One-Out and K-fold cross-validation algorithms and evaluated the performance of the model. In Leave-One-Out, models with small standard deviation were selected as the optimal model, and models with less folds were selected in the K-fold. The average accuracy of the selected models by rainfall duration is over 80%, suggesting that SVM can be used to estimate flooding risk criteria.

• Journal of Korea Water Resources Association
• /
• v.42 no.11
• /
• pp.897-909
• /
• 2009

The relative contribution of between hillslope-flow and stream-flow by heterogeneity of drainage path are quantitatively assessed in the present study with GIUH model based on grid of GIS. Application watersheds are selected Pyeongchang, Bocheong and Wi river basin of IHP in Korea. The mean and variance of hillslope and stream length are estimated and analyzed in each watershed. And coupling with observation storm events, estimate hillslope and stream characteristic velocity which dynamic parameters of GIUH model. The mean and variance of distribution of travel time (i.e. IUH) calculate using estimated pass lengths and characteristic velocities. And the relative contributions are assessed by heterogeneity of drainage path. As a result, the effect of the variance that determine shape of IUH dominate with hillslope's effect in the small watershed area (within 500 $km^2$). Thus, GIUH in the small watershed area must consider hillslope-flow.

• Journal of Korean Society on Water Environment
• /
• v.27 no.2
• /
• pp.228-234
• /
• 2011

The goal of water quality management on stream and watershed is to focus not on discharged loads management but on a water quality management. Discharged loads management is not goal of water quality management but way for perform with total maximum daily loads management. It is necessary to estimate the relation between non-point source with stromwater runoff (NPSSR) and water quality to select a watershed where it is required to manage NPSSR for water quality improvement. To evaluate the effects of NPSSR on stream's water quality, we compare the aspects of water quality in dry and wet seasons using flow duration curve analysis based on flow rate variation data by actual surveying. In this study we attempt to quantify the variation characteristic of water quality and estimate the Inflow characteristic of pollution source with water quality and flow rate monitoring on 10 watersheds. We try to estimate water quality and flow rate by regression analysis and try again regression analysis with each high and low water quality data more than estimations. An analysis of relation between water quality and flow rate of 10 watersheds shows that the water quality of the Nonsan and the Ganggyeong streams had been polluted by NPSSR pollutants. Other eight streams were important point source more than NPSSR. It is wide variation range of $BOD_5$ also high average concentration of $BOD_5$. We have to quantify water quality variation by cv1 in wet season and cv365 in dry season with comparing the estimate of high water quality and low water quality. This method can be used to indicator for water quality variation according to flow rate.

• Journal of Environmental Impact Assessment
• /
• v.22 no.6
• /
• pp.695-703
• /
• 2013

Soil erosion and sediment delivery ratio(SDR) were estimated by using HSPF model in 3 tributaries of upper stream of Geum river-basin. Meteorological data and other input data were constructed from 2006 to 2011 year by the HSPF model. Flow and suspended solid results were relatively matched with the measurement data through the calibration and validation of the model. Soil erosion was proportional to the amount of rainfall and the area of watershed based on the results of model calibration and validation. SDR in Moojunamdea stream was the highest and one in Cho stream was the lowest. This was effected by the geographical characteristic. SDR was 17.6% Moojunamdea stream, 9.1% Cho stream and 13.2 % Bocheong stream. As the SDR was effected by watershed area and shape factor in this study area.

• Journal of Korean Society on Water Environment
• /
• v.23 no.5
• /
• pp.650-658
• /
• 2007

The Universal Soil Loss Equation (USLE) has been used in over 100 countries to estimate potential long-term soil erosion from the field. However, the USLE estimated soil erosion cannot be used to estimate the sediment delivered to the stream networks. For an effective erosion control, it is necessary to compute sediment delivery ratio (SDR) for watershed and sediment yield at watershed outlet. Thus, the Sediment Assessment Tool for Effective Erosion Control (SATEEC) was developed to compute the sediment yield at any point in watershed. In this study, the SATEEC was applied to the Sudong watershed, Chuncheon Gangwon to compare the sediment yield using area-based sediment delivery ratio (SDRA) and slope-based sediment delivery ratio (SDRS) at watershed outlet. The sediment yield using the SDRA by Vanoni, SYA and the sediment yield using the SDRS by Willams and Berndt, SYS were compared for the same sized watersheds. The 19 subwatersheds was 2.19 ha in size, the soil loss and sediment yield were estimated for each subwatershed. Average slope of main stream was about 0.86~3.17%. Soil loss and sediment yield using SDRA and SDRS were distinguished depending on topography, especially in steep and flat areas. The SDRA for all subwatersheds was 0.762, however the SDRS were estimated in the range of 0.553~0.999. The difference between SYA and SYS was -79.74~27.45%. Thus site specific slope-based SDR is more effective in sediment yield estimation than area-based SDR. However it is recommended that watershed characteristic need to be considered in estimating yield behaviors.

• Journal of Korea Water Resources Association
• /
• v.39 no.12 s.173
• /
• pp.1031-1042
• /
• 2006

In recent years, a growing concern exists in watershed and stream improvement projects. Under these circumstances, this paper estimates monetary value of the attributes of alternatives for rehabilitation of hydrologic cycle using choice experiments. Choice experiments shows vivid image and estimates a willingness to pay based on their preference for environmental goods. A preliminary survey shows that the attributes of the Anyangcheon watershed are flood-damage possibilities, Instreamflow, water quality, river characteristic and estimates the tax for the Anyangcheon watershed improvements. We surveyed 200 citizens were selected as samples of watershed beneficing in Seoul and Gyeonggi Province and used conditional logit model to analyze the implicit values of the attributive per household. The benefit of the attributes by province based on the implicit price obtained from estimated parameters were calculated. This study is expected to contribute to the decision-making process for policy-makers by providing useful methodological framework and quantitative information related to watershed improvement projects.

• Journal of Environmental Science International
• /
• v.20 no.10
• /
• pp.1347-1355
• /
• 2011

In Jeju island, runoff has frequently happened when the rainfall depth is over a threshold value. To simulated this characteristic rainfall-runoff model structure has to be modified. In this study, the TRSM (Threshold Runoff Simulation Method) was developed to overcome the limitations of SWAT in applying to the hydrologic characteristics of Jeju island. When the precipitation and soil water are less than threshold value, we revised the SWAT routine not to make surface/lateral or groundwater discharge. For Hancheon watershed, the threshold value was set as 80% of soil water through the analysis of rainfall-runoff relationship. Through the simulation of test watershed, it was proven that TRSM performed much better in simulating pulse type stream flow for the Hancheon watershed.

• Journal of Korea Water Resources Association
• /
• v.34 no.6
• /
• pp.627-640
• /
• 2001

Generally, the synthetic unit hydrograph method is presented to estimate the design flood in the ungaged watershed. However, due to the lack of rainfall-runoff data, the models developed in other countries such as U.S.A. and Japan have been widely used in Korea. Therefore, it may be essential to develope the rainfall-runoff model suitable for the hydrological char-acteristics in Korea. In this study, the representative unit hydrographs are derived from rainfall-runoff data at 19 basins in Selma-Cheon and 3-IHP experimental watersheds using ridge-regression method and Nash model. And a new synthetic unit hydrograph for Korea is suggested by integrating the described results and previous studies on unit hydrograph. The newly developed method is represented as two regression forms with three independent variables of watershed area, channel length, and channel slope by multiple regression analysis is carried out for each watershed, the coefficients of determination are not improved in all cases compared out for each watershed, the coefficients of determination are not improved n all cased the synthetic unit hydrograph for each watershed. Therefore, when the new method is applied to some watersheds, the result analyzed for all data has to be used.

• Journal of Korean Society on Water Environment
• /
• v.24 no.6
• /
• pp.750-757
• /
• 2008

This study is conducted to characterize and evaluate delivered pollutant loads of point and nonpoint source on the upper watershed of lake paldang. The study area consists of 12 watersheds in Namhan-river and Kyungahcheon, which are approximately 80% of total area of Namhan-river and Kyungahcheon. Based on daily delivered loads from watersheds, 61% of $BOD_5$, 81% of T-N and 70% of T-P were from nonpoint sources, suggesting that delivered loads of nonpoint pollutants be crucial to water quality. On the other hand, 78% of $BOD_5$, 92% of T-N and 87% of T-P as delivered load were from nonpoint sources in an upper watershed of Namhan-river, while 48% of $BOD_5$, 70% of T-N and 57% of T-P as delivered load were from nonpoint sources in a lower watershed of Namhan-river, suggesting higher dependency of point sources than upper watershed of Namhan-river. In the characteristic of delivered loading pollutants from point and nonpoint pollution sources, delivered load of nonpoint pollutants differed significantly by seasonal flow, and as though discharged load of point pollutants were yearly uniform, delivered load of point pollutants was found to be flow-dependent because its delivery ratio was changed.