• Proceedings of the Korean Society of Agricultural Engineers Conference
• /
• 2002.10a
• /
• pp.277-280
• /
• 2002

An artificial neural network model was developed to analyze and forecast real time river runoff from the Naju watershed, in Korea. Model forecasts are very accurate (i.e., relative error is less than 3% and $R^2$ is great than 0.99) for calibration data sets. Increasing the time horizon for validation data sets, thus making the model suitable for flood forecasting, decreases the accuracy of the model. The resulting optimal EBPN models for forecasting real time runoff consists of ten rainfall and four and ten runoff data (ANN0410 and ANN1010 models). Performances of the ANN0410 and ANN1010 models remain satisfactory up to 6 hours (i.e., $R^2$ is great than 0.92).

• Journal of Wetlands Research
• /
• v.19 no.1
• /
• pp.132-143
• /
• 2017

It is essential to obtain periodic sediment discharge data in a river in order to minimize problems that may arise from the erosion, transport, and deposition of sediment. However, it is difficult to estimate sediment discharge by the sediment discharge measurement plan in Korea at present, and empirical fomulas or numerical models are used to replace them. This paper has applied the K-DRUM model, a grid-based rainfall-runoff-sediment model, to estimate sediment discharge and ensure the continuity of the data in the watershed. Discharge and sediment load in 17 watersheds were estimated and the applicability of the model was analyzed through comparisons with measured data. For quantitative evaluation, NSE, PBIAS and RSR items were used, and discharge results reflected the tendency of rainfall and showed high statistical value. In case of sediment discharge, the soil erosion process of the watershed is physically well reflected. When the calibration was performed using the measure data, the applicability seems to be excellent in estimating the continuous sediment discharge data in the real watershed.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.20 no.3
• /
• pp.141-152
• /
• 2017

This paper presents a technical method for flood estimation based on satellite rainfall and satellite rainfall correction method for watersheds lacking measurement data. The study area was the Sebou Watershed, Morocco. The Integrated Flood Analysis System(IFAS) and Grid-based Rainfall-Runoff Model(GRM) were applied to estimate watershed runoff. Daily rainfall from ground gauges and satellite-derived hourly data were used. In the runoff simulation using satellite rainfall data, the composites of the daily gauge rainfall and the hourly satellite data were applied. The Shuttle Radar Topographic Mission Digital Elevation Model(SRTM DEM) with a 90m spatial resolution and 1km resolution data from Global map land cover and United States Food and Agriculture Organization(US FAO) Harmonized World Soil Database(HWSD) were used. Underestimated satellite rainfall data were calibrated using ground gauge data. The simulation results using the revised satellite rainfall data were $5,878{\sim}7,434m^3/s$ and $6,140{\sim}7,437m^3/s$ based on the IFAS and GRM, respectively. The peak discharge during flooding of Sebou River Watershed in 2009~2010 was estimated to range from $5,800m^3/s$ to $7,500m^3/s$. The flood estimations from the two hydrologic models using satellite-derived rainfall data were similar. Therefore, the calibration method using satellite rainfall suggested in this study can be applied to estimate the flood discharge of watersheds lacking observational data.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.55 no.6
• /
• pp.19-30
• /
• 2013

This study is to evaluate the use of dual-polarization radar data for storm runoff modeling in Namgang dam (2,293 $km^2$) watershed using KIMSTORM (Grid-based KIneMatic wave STOrm Runoff Model). The Bisl dual-polarization radar data for 3 typhoons (Khanun, Bolaven, Sanba) and 1 heavy rain event in 2012 were obtained from Han River Flood Control Office. Even the radar data were overall less than the ground data in areal average, the spatio-temporal pattern between the two data was good showing the coefficient of determination ($R^2$) and bias with 0.97 and 0.84 respectively. For the case of heavy rain, the radar data caught the rain passing through the ground stations. The KIMSTORM was set to $500{\times}500$ m resolution and a total of 21,372 cells (156 rows${\times}$137 columns) for the watershed. Using 28 ground rainfall data, the model was calibrated using discharge data at 5 stations with $R^2$, Nash and Sutcliffe Model Efficiency (ME) and Volume Conservation Index (VCI) with 0.85, 0.78 and 1.09 respectively. The calibration results by radar rainfall showed $R^2$, ME and VCI were 0.85, 0.79, and 1.04 respectively. The VCI by radar data was enhanced by 5 %.

• Journal of Korea Water Resources Association
• /
• v.37 no.2
• /
• pp.133-144
• /
• 2004

A Tank Model composed of 4 tanks with soil moisture structure was applied to Daecheong Dam and Soyanggang Dam watersheds. Calibration and verification were repeated 332 and 472 times for each watershed using SCE-UA global optimization method for different calibration periods and objective functions. Four different methods of evapotranspiration calculation were used and evaluated. They are pan evaporation, 1963 Penman, FAO-24 Penman-Monteith, and FAO-56 Penman-Monteith methods. Tank model with soil moisture structure showed better results than the standard tank model for daily rainfall-runoff simulation. Two types of objective function for model calibration were found. Proper calibration period are 3 years, in which dry year and flood year are included. If a calibrationperiod has an inadequate runoff rate, the period should be more than 8 years. The four methods of eyapotranspiraton computation showed similar results, but 1963 Penman method was slightly inferior to the other methods.

• Journal of the Korean Society of Hazard Mitigation
• /
• v.1 no.3 s.3
• /
• pp.73-82
• /
• 2001

This study examines the applicability of the Nash model and the Diskin model, which are linear and nonlinear runoff models, respectively, by applying optimization techniques to the parameter calibration of the two models. Nonlinear programming which is one of traditional optimization techniques and Genetic Algorithm which has been actively applied recently are used in this study. The Nash and Diskin models which use the calibrated parameter with a flood events are applied to a different flood event in Soyang Dam basin. The results obtained from the parameter calibration show slight discrepancy depending upon the flood events. It has been found in the comparion between the observed hydrograph and the hydrographs obtained from the parameter calibration that the Diskin model can better simulate the observed hydrograph than the Nash model can, especially, for the peak flow. This can be analyzed that the Diskin model which is a nonlinear runoff model is better off in simulating the nonlinear characteristic of the rainfall-runoff process.

• Journal of Korea Water Resources Association
• /
• v.38 no.10 s.159
• /
• pp.861-869
• /
• 2005

A rainfall-runoff model should be calibrated so that the model simulates the hydrological behavior of the basin as accurately as possible. In this study, to calibrate the five parameters of the SSARR model, a multi-objective function and the genetic algorithm were used. The solution of the multi-objective function will not, in general, be a single unique set of parameters but will consist of the so-called Pareto solution according to various trade-offs between the different objectives. The calibration strategy using multi-objective function could decrease calibrating time and effort. From the Pareto solution, a single solution could be selected to simulate a specific flow condition.

• Journal of Wetlands Research
• /
• v.16 no.1
• /
• pp.19-32
• /
• 2014

The purpose of this paper is to obtain reliable rainfall data for runoff simulation and other hydrological analysis by the calibration of gauge rainfall. The calibrated gauge rainfall could be close to the actual value with rainfall on the ground. In order to analyze the wind effect of ground rain gauge, we selected the rain gauge sites with and without a windshield and standard rain gauge data from Chupungryeong weather station installed by standard of WMO. Simple linear regression model and artificial neural networks were used for the calibration of rainfalls, and we verified the reliability of the calibrated rainfalls through the runoff analysis using $Vflo^{TM}$. Rainfall calibrated by linear regression is higher amount of rainfall in 5%~18% than actual rainfall, and the wind remarkably affects the rainfall amount in the range of wind speed of 1.6~3.3m/s. It is hard to apply the linear regression model over 5.5m/s wind speed, because there is an insufficient wind speed data over 5.5m/s and there are also some outliers. On the other hand, rainfall calibrated by neural networks is estimated lower rainfall amount in 10~20% than actual rainfall. The results of the statistical evaluations are that neural networks model is more suitable for relatively big standard deviation and average rainfall. However, the linear regression model shows more suitable for extreme values. For getting more reliable rainfall data, we may need to select the suitable model for rainfall calibration. We expect the reliable hydrologic analysis could be performed by applying the calibration method suggested in this research.

• Journal of Korean Society on Water Environment
• /
• v.39 no.1
• /
• pp.38-45
• /
• 2023

In this study, the removal efficiency of road sweeping and sand filter facility for removing total suspended solid (TSS) as nonpoint source pollution from expressway was evaluated for the last 10 years (2012~2021) using ROADMOD. ROADMOD is a screening level model and was calibrated for runoff rate and TSS loading both at the inlet, which is the loading from the drainage area, and the outlet, from the sand filter facility. The drainage area is 715 m² and the dimensions of sand filter facility are 1.5 m (wide) × 3.8 m (length) × 1.5 m (depth). The monitoring period for model calibration was the rainfall event during Aug. 31~Sep. 1, 2021 and the amount of rainfall was 74.5 mm. As a result of calibration, the determination coefficients (R²) of the flow rate were 0.66 and 0.86, for the inlet and outlet, respectively, and those of TSS loading were 0.50 and 0.84, for the inlet and outlet, respectively. Considering that ROADMOD is a screening level model, the calibration results were reasonable to evaluate the best management practices (BMPs) on the expressway. Using ROADMOD simulation results for 2012~2021, the average yearly runoff rate from the expressway was 82% and removal efficiency was 9% for road sweeping, 35% for sand filter facility, and 39% for both road sweeping and sand filter facility.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2006.05a
• /
• pp.770-774
• /
• 2006

The determination of feasible design flood is the most important to control flood damage in river management. Model parameters should be calibrated using observed discharge but due to deficiency of observed data the parameters have been adopted by engineer's empirical sense. Storage coefficient in the Clark unit hydrograph method mainly affects magnitude of peak flood. This study is to estimate the storage coefficients based on the observed rainfall-runoff events at the four stage stations in the Hantan river basin. Model calibration is the process of adjusting model parameter values until model results match historical data. An objective function which is the percent difference between the observed and computed peak flows is available for measuring the goodness-of-fit between computed and observed hydrographs. By sensitivity analysis for the storage coefficient, it has been shown that the storage coefficients affect the peak flows. The Clark parameters adopted in the River Rectification Basic Plan have been estimated through an iterative process designed to produce a hydrograph with the peak flow.