• Magazine of the Korean Society of Agricultural Engineers
• /
• v.45 no.2
• /
• pp.45-57
• /
• 2003

An artificial neural network model was developed to analyze and forecast Short-term river runoff from the Naju watershed, in Korea. Error back propagation neural networks (EBPN) of hourly rainfall and runoff data were found to have a high performance In forecasting runoff. The number of hidden nodes were optimized using total error and Bayesian information criterion. Model forecasts are very accurate (i.e., relative error is less than 3% and $R^2$is greater than 0.99) for calibration and verification data sets. Increasing the time horizon for application data sets, thus mating the model suitable for flood forecasting. decreases the accuracy of the model. The resulting optimal EBPN models for forecasting hourly runoff consists of ten rainfall and four runoff data(ANN0410 model) and ten rainfall and ten runoff data(ANN1010 model). Performances of the ANN0410 and ANN1010 models remain satisfactory up to 6 hours (i.e., $R^2$is greater than 0.92).

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.31-31
• /
• 2023

During December 2022, the northeast monsoon, which dominates the south and the Gulf of Thailand, had significant rainfall that impacted the lower southern region, causing flash floods, landslides, blustery winds, and the river exceeding its bank. The Golok River, located in Narathiwat, divides the border between Thailand and Malaysia was also affected by rainfall. In flood management, instruments for measuring precipitation and water level have become important for assessing and forecasting the trend of situations and areas of risk. However, such regions are international borders, so the installed measuring telemetry system cannot measure the rainfall and water level of the entire area. This study aims to predict 72 hours of water level and evaluate the situation as information to support the government in making water management decisions, publicizing them to relevant agencies, and warning citizens during crisis events. This research is applied to machine learning (ML) for water level prediction of the Golok River, Lan Tu Bridge area, Sungai Golok Subdistrict, Su-ngai Golok District, Narathiwat Province, which is one of the major monitored rivers. The eXtreme Gradient Boosting (XGBoost) algorithm, a tree-based ensemble machine learning algorithm, was exploited to predict hourly water levels through the R programming language. Model training and testing were carried out utilizing observed hourly rainfall from the STH010 station and hourly water level data from the X.119A station between 2020 and 2022 as main prediction inputs. Furthermore, this model applies hourly spatial rainfall forecasting data from Weather Research and Forecasting and Regional Ocean Model System models (WRF-ROMs) provided by Hydro-Informatics Institute (HII) as input, allowing the model to predict the hourly water level in the Golok River. The evaluation of the predicted performances using the statistical performance metrics, delivering an R-square of 0.96 can validate the results as robust forecasting outcomes. The result shows that the predicted water level at the X.119A telemetry station (Golok River) is in a steady decline, which relates to the input data of predicted 72-hour rainfall from WRF-ROMs having decreased. In short, the relationship between input and result can be used to evaluate flood situations. Here, the data is contributed to the Operational support to the Special Water Resources Management Operation Center in Southern Thailand for flood preparedness and response to make intelligent decisions on water management during crisis occurrences, as well as to be prepared and prevent loss and harm to citizens.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.51 no.6
• /
• pp.11-16
• /
• 2009

This study is to suggest the SWAT model as inputs for the estimation of CN (Curve number) if we do not have hourly rainfall and runoff data in the ungaged watershed. The daily CNs were estimated by using SWAT model for Chungju dam watershed and the CNs by hourly rainfall and runoff data in the same period with daily CN estimation were also estimated. Then the daily and hourly CNs were compared each other. The CNs by SWAT model were larger than the actual CNs. 7.4% larger in AMC-I, 1.2% in AMC-II, and 6.3% in AMC-III respectively. If we consider various uncertainties in the estimation of CN, the error of 6.8% could be acceptable for the application in the field.

• Journal of Korea Water Resources Association
• /
• v.43 no.6
• /
• pp.533-541
• /
• 2010

It is basic for a flood prediction to calculate direct runoff from rainfall in a basin by the rainfall-runoff model. The direct runoff is calculated from rainfall excess or effective rainfall based on a rainfall-runoff model. The total rainfall minus rainfall loss equals rainfall excess with time. This loss can be treated equal to an infiltration loss under the assumption that the infiltration is a major one among the losses in the rainfall-runoff model. Practically obtaining the infiltration loss $\Phi$ index method, W index method or modified ones of these have been used. In this study it is assumed the loss of rainfall in a basin be a well-known Horton infiltration mechanism. And in case that the parameter set is given in the Horton infiltration model a procedure and assumption for calculating hourly infiltration loss and rainfall excess are offered and the results of its application are compared with those of $\Phi$ index method. By this study it is well shown the value of Horton infiltration function is exponentially decay with time as the Horton infiltration mechanism.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.66 no.4
• /
• pp.51-57
• /
• 2024

This study was conducted on how to simulate runoff, which was done using existing physical models, using an LSTM (Long Short-Term Memory) model based on deep learning. Tancheon, the first tributary of the Han River, was selected as the target area for the model application. To apply the model, one water level observatory and four rainfall observatories were selected, and hourly data from 2020 to 2023 were collected to apply the model. River water level of the outlet of the Tancheon basin was simulated by inputting precipitation data from four rainfall observation stations in the basin and average preceding 72-hour precipitation data for each hour. As a result of water level simulation using 2021 to 2023 data for learning and testing with 2020 data, it was confirmed that reliable simulation results were produced through appropriate learning steps, reaching a certain mean absolute error in a short period time. Despite the short data period, it was found that the mean absolute percentage error was 0.5544~0.6226%, showing an accuracy of over 99.4%. As a result of comparing the simulated and observed values of the rapidly changing river water level during a specific heavy rain period, the coefficient of determination was found to be 0.9754 and 0.9884. It was determined that the performance of LSTM, which aims to simulate river water levels, could be improved by including preceding precipitation in the input data and using precipitation data from various rainfall observation stations within the basin.

• Proceedings of the KSR Conference
• /
• 2010.06a
• /
• pp.2014-2020
• /
• 2010

This paper aims to find out characteristics of railroad weather-related accidents and incidents and to proposes the plan of railroad disaster prevention according to the precipitation. For this, we make the database about the railroad weather-related accidents and incidents and analysis the relationship between the hourly and cumulative precipitation and railroad accidents. The results are as follows; The weather events that have the most occurrence frequency of railroad weather-related accidents and incidents is a rainfall of the precipitation and then the cause of that was the falling rocks and the collapsed roadbed. The rainfall patterns of collapsed roadbed were classified into 4 groups. When the variation of hourly rainfall is 10/15 mm/hr over, we need to consider the caution/stop of train operation and a speed limit, respectively.

• Journal of Korea Water Resources Association
• /
• v.50 no.11
• /
• pp.735-743
• /
• 2017

With growing concerns about ever-increasing anthropogenic greenhouse gas emissions, it is crucial to enhance preparedness for unprecedented extreme weathers that can bring catastrophic consequences. In this study, we proposed a stochastic framework that considers uncertainty in weather forecasts for flood analyses. First, we calibrated a simple rainfall-runoff model against observed hourly hydrographs. Then, using probability density functions of rainfall depths conditioned by 6-hourly weather forecasts, we generated many stochastic rainfall depths for upcoming 48 hours. We disaggregated the stochastic 6-hour rainfalls into an hourly scale, and input them into the runoff model to quantify a probabilistic range of runoff during upcoming 48 hours. Under this framework, we assessed two rainfall events occurred in Bocheong River Basin, South Korea in 2017. It is indicated actual flood events could be greater than expectations from weather forecasts in some cases; however, the probabilistic runoff range could be intuitive information for managing flood risks before events. This study suggests combining deterministic and stochastic methods for forecast-based flood analyses to consider uncertainty in weather forecasts.

• Journal of Korean Society on Water Environment
• /
• v.33 no.2
• /
• pp.160-169
• /
• 2017

The objective of this study are to analyze changes in future rainfall patterns in the Soyang-dam watershed according to the RCP 4.5 scenario of climate change. Second objective is to project peak flow and hourly sediment simulated for the future extreme rainfall events using the SWAT model. For these, accuracy of SWAT hourly simulation for the large scale watershed was evaluated in advance. The results of model calibration showed that simulated peak flow matched observation well with acceptable average relative error. The results of future rainfall pattern changes analysis indicated that extreme storm events will become more severe and frequent as climate change progresses. Especially, possibility of occurrence of large scale extreme storm events will be greater on the periods of 2030-2040 and 2050-2060. In addition, as shown in the SWAT hourly simulation for the future extreme storm events, more severe flood and turbid water can happen in the future compared with the most devastating storm event which occurred by the typhoon Ewiniar in 2006 year. Thus, countermeasures against future extreme storm event and turbid water are needed to cope with climate change.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2005.09a
• /
• pp.817-817
• /
• 2005

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.34 no.3
• /
• pp.821-831
• /
• 2014

Stochastic rainfall generators or stochastic simulation have been widely employed to generate synthetic rainfall sequences which can be used in hydrologic models as inputs. The calibration of Poisson cluster stochastic rainfall generator (e.g. Modified Bartlett-Lewis Rectangular Pulse, MBLRP) is seriously affected by local minima that is usually estimated from the local optimization algorithm. In this regard, global optimization techniques such as particle swarm optimization and shuffled complex evolution algorithm have been proposed to better estimate the parameters. Although the global search algorithm is designed to avoid the local minima, reliable parameter estimation of MBLRP model is not always feasible especially in a limited parameter space. In addition, uncertainty associated with parameters in the MBLRP rainfall generator has not been properly addressed yet. In this sense, this study aims to develop and test a Bayesian model based parameter estimation method for the MBLRP rainfall generator that allow us to derive the posterior distribution of the model parameters. It was found that the HBM based MBLRP model showed better performance in terms of reproducing rainfall statistic and underlying distribution of hourly rainfall series.