• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.2B
• /
• pp.225-236
• /
• 2008

The techniques to calculate the probability precipitation for the design of hydrological projects can be determined by the point frequency analysis and the regional frequency analysis. Probability precipitation usually calculated by point frequency analysis using rainfall data that is observed in rainfall observatory which is situated in the basin. Therefore, Probability precipitation through point frequency analysis need observed rainfall data for enough periods. But, lacking precipitation data can be calculated to wrong parameters. Consequently, the regional frequency analysis can supplement the lacking precipitation data. Therefore, the regional frequency analysis has weaknesses compared to point frequency analysis because of suppositions about probability distributions. In this paper, rainfall observatory in Korea did grouping by cluster analysis using position of timely precipitation observatory and characteristic time rainfall. Discordancy and heterogeneity measures verified the grouping precipitation observatory by the cluster analysis. So, there divided rainfall observatory in Korea to 6 areas, and the regional frequency analysis applies index-flood techniques and L-moment techniques. Also, the probability precipitation was calculated by the regional frequency analysis using variable kernel density function. At the results, the regional frequency analysis of the variable kernel function can utilize for decision difficulty of suitable probability distribution in other methods.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2005.05b
• /
• pp.593-598
• /
• 2005

At-site analysis is not appropriate if the record length is shorter than target return period T. If the record length is longer than 27 years, then at-site analysis may be sufficient(Institute of Hydrology, 1999). However, in such a case, regional frequency analysis is recommended for purpose of comparison. Record lengths of annual maximum rainfall data in Korea are usually shorter than 50 years. It is therefore essential to apply regional frequency analysis for estimating rainfall quantiles of more than 100 years return period. In this research, regional rainfall frequency analysis is performed for hourly rainfall data of South Korea. Homogeneous regions are idntified by clusgter analysis which is a standard method of statistical multivariate analysis for dividing a data set into groups. An appropriate distribution is chosen by goodness-of-fit test. GLO is found to be an appropriate distribution as a result of goodness-of-fit measure (Hosking & Wallis, 1997). Simulation experiments are performed to check the performance of frequency analysis techniques. The effects of discordant sites on quantiles are considered.

• Journal of Korea Water Resources Association
• /
• v.52 no.12
• /
• pp.985-999
• /
• 2019

Over the last decade, droughts have become more severe and frequent in many regions, and several studies have been conducted to explore the recent drought. Copula-based bivariate drought frequency analysis has been widely used to evaluate drought risk in the context of point frequency analysis. However, the relatively significant uncertainties in the parameters are problematic when available data are limited. For this reason, the primary purpose of this study is to develop a regional drought frequency model based on the Copula function. All parameters, including marginal and copula functions in the regional frequency model, were estimated simultaneously. Here, we present a case study of recent drought 2013-2015 over the Han-River watershed where severe drought risk is consistently found to increase. The proposed model provided a reliable way to significantly reduce the uncertainty of parameters with a Bayesian modeling framework. The uncertainty of the joint return period in the regional frequency analysis is nearly three times lower than that of the point frequency analysis. Accordingly, DIC values in the regional frequency analysis model are significantly decreased by 15. The results confirm that the proposed model is not only reliably representing characteristics of historical droughts and dependencies between drought variables, but also providing the efficacy of understanding regional drought characteristics.

• Proceedings of the Korean Society of Agricultural Engineers Conference
• /
• 2002.10a
• /
• pp.237-240
• /
• 2002

Design rainfall using LH-moments following the consecutive duration were derived by the regional and at-site analysis using the observed and simulated data resulted from Monte Carlo techniques. RRMSE, RBIAS and RR in RRMSE for the design rainfall were computed and compared in the regional and at-site frequency analysis. Consequently, it was shown that the regional analysis can substantially more reduce the RRMSE, RBIAS and RR in RRMSE than at-site analysis in the prediction of design rainfall. RE for an optimal order of L-moments was also computed by the methods of L, L1, L2, L3 and L4-moments for GEV distribution. It was found that the method of L-moments is more effective than the others for getting optimal design rainfall according to the regions and consecutive durations in the regional frequency analysis. Diagrams for the design rainfall derived by the regional frequency analysis using L-moments were drawn according to the regions and consecutive durations by GIS techniques.

• Proceedings of the Korean Society of Agricultural Engineers Conference
• /
• 2001.10a
• /
• pp.319-323
• /
• 2001

This study was conducted to derive the regional design rainfall by the regional frequency analysis based on the regionalization of the precipitation. Using the L-moment ratios and Kolmogorov-Smirnov test, the underlying regional probability distribution was identified to be the Generalized extreme value distribution among apt]lied distributions. regional and at-site parameters of the Generalized extreme value distribution were estimated by the method of L-moment. The regional and at-site analysis for the design rainfall were tested by Monte Carlo simulation. Relative root-mean-square error(RRMSE), relative bias(RBIAS) and relative reduction(RR) in RRMSE were computed and compared with those resulting from at-site Monte Carlo simulation. All show that the regional analysis procedure can substantially reduce the RRMSE, RBIAS and RR in RRMSE in the prediction of design rainfall. Consequently, optimal design rainfalls following the regions and consecutive durations were derived by the regional frequency analysis.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.43 no.5
• /
• pp.70-82
• /
• 2001

This study was conducted to derive the regional design rainfall by the regional frequency analysis based on the regionalization of the precipitation suggested by the first report of this project. Using the L-moment ratios and Kolmogorov-Smirnov test, the underlying regional probability distribution was identified to be the Generalized extreme value distribution among applied distributions. Regional and at-site parameters of the generalized extreme value distribution were estimated by the linear combination of the probability weighted moments, L-moment. The regional and at-site analysis for the design rainfall were tested by Monte Carlo simulation. Relative root-mean-square error(RRMSE), relative bias(RBIAS) and relative reduction(RR) in RRMSE were computed and compared with those resulting from at-site Monte Carlo simulation. All show that the regional analysis procedure can substantially reduce the RRMSE, RBIAS and RR in RRMSE in the prediction of design rainfall. Consequently, optimal design rainfalls following the legions and consecutive durations were derived by the regional frequency analysis.

• Journal of Environmental Science International
• /
• v.8 no.4
• /
• pp.431-441
• /
• 1999

In this study, the regional frequency analysis is used to determine each subbasin drought frequency with reliable monthly precipitation and the L-Moments method which is almost unbiased and has very nearly a normal distribution is used for the parameter estimation of monthly precipitation time series in Nakdong river basin. As the result of this study, the duration of '93-'94 is most severe drought year than any other water year and the drought frequency is established as compared the regional frequency analysis result of cumulative precipitation of 12th duration months in each subbasin with that of 12th duration months in the major drought duration. The Linear regression equation is induced according to linear regression analysis of drought frequency between Nakdong total basin and each subbasin of the same drought duration. Therefore, as the foundation of this study, it can be applied proposed method and procedure of this study to the water budget analysis considering safety standards for the design of impounding facilities large-scale river basin and for this purpose, above all, it is considered that expansion of reliable preciptation data is needed in watershed rainfall station.

• Journal of Korea Water Resources Association
• /
• v.47 no.5
• /
• pp.469-482
• /
• 2014

This study developed a Bayesian spatial regional frequency analysis, which aimed to analyze spatial patterns of design rainfall by incorporating geographical information (e.g. latitude, longitude and altitude) and climate characteristics (e.g. annual maximum series) within a Bayesian framework. There are disadvantages to considering geographical characteristics and to increasing uncertainties associated with areal rainfall estimation on the existing regional frequency analysis. In this sense, this study estimated the parameters of Gumbel distribution which is a function of geographical and climate characteristics, and the estimated parameters were spatially interpolated to derive design rainfall over the entire Han-river watershed. The proposed Bayesian spatial regional frequency analysis model showed similar results compared to L-moment based regional frequency analysis, and even better performance in terms of quantifying uncertainty of design rainfall and considering geographical information as a predictor.

• Journal of Korea Water Resources Association
• /
• v.51 no.5
• /
• pp.383-393
• /
• 2018

The regional frequency analysis is the method which uses not only sample of target station but also sample of neighborhood stations in which are classified as hydrological homogeneous regions. Consequently, identification of homogeneous regions is a very important process in regional frequency analysis. In this study, homogeneous regions for regional frequency analysis of precipitation were identified by the self-organizing map (SOM) which is one of the artificial neural network. Geographical information and hourly rainfall data set were used in order to perform the SOM. Quantization error and topographic error were computed for identifying the optimal SOM map. As a result, the SOM model organized by $7{\times}6$ array with 42 nodes was selected and the selected stations were classified into 6 clusters for rainfall regional frequency analysis. According to results of the heterogeneity measure, all 6 clusters were identified as homogeneous regions and showed more homogeneous regions compared with the result of previous study.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.6B
• /
• pp.723-729
• /
• 2008

Confidence intervals of growth curves are calculated to assess the uncertainty of index flood method as a regional frequency analysis. The asymptotic variance of quantile estimator for the generalized logistic distribution is introduced to evaluate confidence intervals. In addition, the variances of at-site frequency estimator and regional frequency estimator are used to evaluate an efficiency index. The efficiency indexes for 14 homogeneous regions based on 378 stations show that index flood method estimators are more efficient than at-site frequency estimators. It is shown that the number of sites in a region needs to be limited for regional gain.