• 한국방재학회:학술대회논문집
• /
• 2008.02a
• /
• pp.101-104
• /
• 2008

In this study, rainfall adjust and forecasting using artificial neural network(ANN) which includes a correlation coefficient is application in Seoul region. It analyzed one-hour rainfall data which has been reported in 25 region in seoul during from 2000 to 2006 at rainfall observatory by AWS. The ANN learning algorithm apply for input data that each region using cross-correlation will use the highest correlation coefficient region. In addition, rainfall adjust analyzed the minimum error based on correlation coefficient and determination coefficient related to the input region. ANN model used back-propagation algorithm for learning algorithm. In case of the back-propagation algorithm, many attempts and efforts are required to find the optimum neural network structure as applied model. This is calculated similar to the observed rainfall that the correlation coefficient was 0.98 in missing rainfall adjust at 10 region. As a result, ANN model has been for suitable for rainfall adjust. It is considered that the result will be more accurate when it includes climate data affecting rainfall.

• Journal of Environmental Science International
• /
• v.6 no.3
• /
• pp.205-211
• /
• 1997

For the prediction of multi-site rainfall with radar data and ground meteorological data, a rainfall prediction model was proposed, which uses the neural network theory, a kind of artifical Intelligence technique. The Input layer of the prediction model was constructed with current ground meteorological data, their variation, moving vectors of rain- fall field and digital terrain of the measuring site, and the output layer was constructed with the predicted rainfall up to 3 hours. In the application of the prediction model to the Pyungchang river basin, the learning results of neural network prediction model showed more Improved results than the parameter estimation results of an existing physically based model. And the proposed model comparisonally well predicted the time distribution of ralnfall.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2018.05a
• /
• pp.499-499
• /
• 2018

Hydro-meteorological extremes are trivial in these days. Therefore, it is important to identify extreme hydrological events in advance to mitigate the damage due to the extreme events. In this context, exploring temporal distribution of sub-daily extreme rainfall at multiple rain gauges would informative to identify different states to describe severity of the disaster. This study proposehidden Markov chain model (HMM) based rainfall analysis tool to understand the temporal sub-daily rainfall patterns over South Korea. Hourly and daily rainfall data between 1961 and 2017 for 92 stations were used for the study. HMM was applied to daily rainfall series to identify an observed hidden state associated with rainfall frequency and intensity, and further utilized the estimated hidden states to derive a temporal distribution of daily extreme rainfall. Transition between states over time was clearly identified, because HMM obviously identifies the temporal dependence in the daily rainfall states. The proposed HMM was very useful tool to derive the temporal attributes of the daily rainfall in South Korea. Further, daily rainfall series were disaggregated into sub-daily rainfall sequences based on the temporal distribution of hourly rainfall data.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.183-183
• /
• 2021

Deep learning methods and their application have become an essential part of prediction and modeling in water-related research areas, including hydrological processes, climate change, etc. It is known that application of deep learning leads to high availability of data sources in hydrology, which shows its usefulness in analysis of precipitation, runoff, groundwater level, evapotranspiration, and so on. However, there is still a limitation on microclimate analysis and prediction with deep learning methods because of deficiency of gauge-based data and shortcomings of existing technologies. In this study, a real-time rainfall prediction model was developed from a sky image data set with convolutional neural networks (CNNs). These daily image data were collected at Chung-Ang University and Korea University. For high accuracy of the proposed model, it considers data classification, image processing, ratio adjustment of no-rain data. Rainfall prediction data were compared with minutely rainfall data at rain gauge stations close to image sensors. It indicates that the proposed model could offer an interpolation of current rainfall observation system and have large potential to fill an observation gap. Information from small-scaled areas leads to advance in accurate weather forecasting and hydrological modeling at a micro scale.

• Journal of Korea Water Resources Association
• /
• v.55 no.3
• /
• pp.177-189
• /
• 2022

The purpose of this study is to apply the d4PDF (Data for Policy Decision Making for Future Change) constructed from a large-scale ensemble climate simulation to estimate the probable rainfall with low frequency and high intensity. In addition, this study analyzes the uncertainty caused by the application of the frequency analysis by comparing the probable rainfall estimated using the d4PDF with that estimated using the observed data and frequency analysis at Geunsam, Imsil, Jeonju, and Jangsu stations. The d4PDF data consists of a total of 50 ensembles, and one ensemble provides climate and weather data for 60 years such as rainfall and temperature. Thus, it was possible to collect 3,000 annual maximum daily rainfall for each station. By using these characteristics, this study does not apply the frequency analysis for estimating the probability rainfall, and we estimated the probability rainfall with a return period of 10 to 1000 years by distributing 3,000 rainfall by the magnitude based on a non-parametric approach. Then, the estimated probability rainfall using d4PDF was compared with those estimated using the Gumbel or GEV distribution and the observed rainfall, and the deviation between two probability rainfall was estimated. As a result, this deviation increased as the difference between the return period and the observation period increased. Meanwhile, the d4PDF reasonably suggested the probability rainfall with a low frequency and high intensity by minimizing the uncertainty occurred by applying the frequency analysis and the observed data with the short data period.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2004.05b
• /
• pp.256-261
• /
• 2004

Due to the problems of global warming, the frequency of meteorological extremes such as droughts, floods and the annual rainfall amount are suddenly increasing. Even though the increase of greenhouse gases, for example, is thought to be the main factor for global warming, its impact on global climate has not yet been revealed clearly in rather quantitative manners. Therefore, tile objective of this study is to inquire the change of precipitation condition due to climate change by global warming. In brief, this study want to see its assumption if rainfall quantile estimates are really changing. In order to analyze the temporal change, the rainfall quantile estimates at the Seoul rain gauge stations are estimated for the 21-year data period being moved from 1908 to 2002 with 1-year lag. The main objective of this study is to analyze the variability of rainfall quantile estimates using four methods. Next, The changes in confidence interval of rainfall quantile are evaluated by increasing the data period. It has been found that confidence interval of rainfall quantile estimates is reduced as the data period increases. When the hydraulic structures are to be designed, it is important to select the data size and to re-estimate the flood prevention capacity in existing river systems.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2020.06a
• /
• pp.96-96
• /
• 2020

In order to establish the risk criteria of inundation due to typhoons or heavy rainfall, research is underway to predict the limit rainfall using basin characteristics, limit rainfall and artificial intelligence algorithms. In order to improve the model performance in estimating the limit rainfall, the learning data are used after the pre-processing. When 50.0% of the entire data was removed as an outlier in the pre-processing process, it was confirmed that the accuracy is over 90%. However, the use rate of learning data is very low, so there is a limitation that various characteristics cannot be considered. Accordingly, in order to predict the limit rainfall reflecting various watershed characteristics by increasing the use rate of learning data, the watersheds with similar characteristics were clustered. The algorithms used for clustering are K-Means, Agglomerative, DBSCAN and Spectral Clustering. The k-Means, DBSCAN and Agglomerative clustering algorithms are clustered at the impervious area ratio, and the Spectral clustering algorithm is clustered in various forms depending on the parameters. If the results of the clustering algorithm are applied to the limit rainfall prediction algorithm, various watershed characteristics will be considered, and at the same time, the performance of predicting the limit rainfall will be improved.

• The Journal of Engineering Geology
• /
• v.34 no.2
• /
• pp.173-191
• /
• 2024

Landslides pose significant threats to many countries globally, yet the development and implementation of effective landslide early warning systems (LEWS) remain challenging due to multifaceted complexities spanning scientific, technological, and political domains. Addressing these challenges demands a holistic approach. Technologically, integrating thresholds, such as rainfall thresholds, with real-time data within accessible, open-source software stands as a promising solution for LEWS. This article introduces LandScient_EWS, a PHP-based program tailored to address this need. The software facilitates the comparison of real-time measured data, such as rainfall, with predefined landslide thresholds, enabling precise calculations and graphical representation of real-time landslide advisory levels across diverse spatial scales, including regional, basin, and hillslope levels. To illustrate its efficacy, the program was applied to a case study in Medellin, Colombia, where a rainfall event on August 26, 2008, triggered a shallow landslide. Through pre-defined rainfall intensity and duration thresholds, the software simulated advisory levels during the recorded rainfall event, utilizing data from a rain gauge positioned within a small watershed and a single grid cell (representing a hillslope) within that watershed. By identifying critical conditions that may lead to landslides in real-time scenarios, LandScient_EWS offers a new paradigm for assessing and responding to landslide hazards, thereby improving the efficiency and effectiveness of LEWS. The findings underscore the software's potential to streamline the integration of rainfall thresholds into both existing and future landslide early warning systems.

• Journal of Korea Water Resources Association
• /
• v.51 no.8
• /
• pp.739-745
• /
• 2018

Recently, the incidence of heavy rainfall is increasing. Therefore, a rainfall analysis should be performed considering increasing frequency. The current rainfall analysis for hydrologic design use the hourly rainfall data of ASOS with a density of 36 km on the Korean Peninsula. Therefore, medium and small scale watershed included Thiessen network at the same rainfall point are analyzed with the same design rainfall and time distribution. This causes problem that the watershed characteristics can not be considered. In addition, there is a problem that the temporal-spatial change of the heavy rainfall occurring in the range of 10~20 km can not be considered. In this study, Author estimated design rainfall considering heavy rainfall using minutely rainfall data of AWS, which are relatively dense than ASOS. Also, author analyzed the time distribution and runoff of each case to estimate the huff's method suitable for the watershed. The research result will contribute to the estimation of the design hydrologic data considering the heavy rainfall and watershed characteristics.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2020.06a
• /
• pp.99-99
• /
• 2020

The frequency and scale of domestic flood damage continues to increase, but the criteria for responding to flood damage have not been established. To this end, research is underway to estimate the amount of rainfall in each region so that it can be used to respond to flood damage. The limit rainfall is defined as the cumulative maximum rainfall for each duration that causes flooding, and this research purpose to improve the threshold rainfall by estimating the damage based on the damage history in units of 5 years and analyzing changes over time. The limit rainfall based on the damage history was estimated by using the NDMS past damage history of the Ministry of the Interior and Safety and the rainfall minutes data of AWS and ASOS. The period for estimating the limit rainfall is 2013 ~ 2017, 2015 ~ 2019, and the limit rainfall is estimated by analyzing the relationship between the flood damage history and the rainfall event in each period. Considering changes in watershed characteristics and disaster prevention performance, the data were compared using 5-year data. As a result of the analysis, the limit rainfall based on the damage history could be estimated for less than about 10.0% of the administrative dongs nationwide. As a result of comparing the limit rainfall by period, it was confirmed that the area where the limit rainfall has increased or decreased This was analyzed as a change due to rainfall events or urbanization, and it is judged that it will be possible to improve the risk criteria of flooding.