• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.6
• /
• pp.665-670
• /
• 2014

A SVM is a kind of binary classifier in order to find optimal hyperplane which separates training data into two groups. Due to its remarkable performance, the SVM is applied in various fields such as inductive inference, binary classification or making predictions. Also it is a representative black box model; there are plenty of actively discussed researches about analyzing trained SVM classifier. This paper conducts a study on a method that is automatically detecting the line-shaped echoes, sun strobe echo and radial interference echo, using the SVM algorithm because the line-shaped echoes appear relatively often and disturb weather forecasting process. Using a spatial clustering method and corrected reflectivity data in the weather radar, the training data is made up with mean reflectivity, size, appearance, centroid altitude and so forth. With actual occurrence cases of the line-shaped echoes, the trained SVM classifier is verified, and analyzed its characteristics using the decision tree method.

• Journal of the Korean earth science society
• /
• v.34 no.1
• /
• pp.13-27
• /
• 2013

Quantitative precipitation estimation (QPE) is one of the most important elements in meteorological and hydrological applications. In this study, we adjusted the QPE from an S-band weather radar based on co-kriging method using the geostatistical structure function of error distribution of radar rainrate. In order to estimate the accurate quantitative precipitation, the error of radar rainrate which is a primary variable of co-kriging was determined by the difference of rain rates from rain gauge and radar. Also, the gauge rainfield, a secondary variable of co-kriging is derived from the ordinary kriging based on raingauge network. The error distribution of radar rain rate was produced by co-kriging with the derived theoretical variogram determined by experimental variogram. The error of radar rain rate was then applied to the radar estimated precipitation field. Locally heavy rainfall case during 6-7 July 2009 is chosen to verify this study. Correlation between adjusted one-hour radar rainfall accumulation and rain gauge rainfall accumulation improved from 0.55 to 0.84 when compared to prior adjustment of radar error with the adjustment of root mean square error from 7.45 to 3.93 mm.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.26 no.1
• /
• pp.1-8
• /
• 2016

The Korea Meteorological Administration has operated the Automatic Weather Stations, of the average 13 km horizontal resolution, to observe rainfall. However, an additional RADAR network also has been operated in all-weather conditions, because AWS network could not observed rainfall over the sea. In general, the rain rate is obtained by estimating the relationship between the radar reflectivity (Z) and the rainfall (R). But this empirical relationship needs to be optimized on the rainfall over the Korean peninsula. This study was carried out to optimize the Z-R relationship in the summer of 2014 using a parallel Micro Genetic Algorithm. The optimized Z-R relationship, $Z=120R^{1.56}$, using a micro genetic algorithm was different from the various Z-R relationships that have been previously used. However, the landscape of the fitness function found in this study looked like a flat plateau. So there was a limit to the fine estimation including the complex development and decay processes of precipitation between the ground and an altitude of 1.5km.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.63 no.5
• /
• pp.676-682
• /
• 2014

In this paper, precipitation echo(PRE) and non-precipitaion echo(N-PRE)(including ground echo and clear echo) through weather radar data are identified with the aid of neuro-fuzzy algorithm. The accuracy of the radar information is lowered because meteorological radar data is mixed with the PRE and N-PRE. So this problem is resolved by using RBFNN and judgement module. Structure expression of weather radar data are analyzed in order to classify PRE and N-PRE. Input variables such as Standard deviation of reflectivity(SDZ), Vertical gradient of reflectivity(VGZ), Spin change(SPN), Frequency(FR), cumulation reflectivity during 1 hour(1hDZ), and cumulation reflectivity during 2 hour(2hDZ) are made by using weather radar data and then each characteristic of input variable is analyzed. Input data is built up from the selected input variables among these input variables, which have a critical effect on the classification between PRE and N-PRE. Echo judgment module is developed to do echo classification between PRE and N-PRE by using testing dataset. Polynomial-based radial basis function neural networks(RBFNNs) are used as neuro-fuzzy algorithm, and the proposed neuro-fuzzy echo pattern classifier is designed by combining RBFNN with echo judgement module. Finally, the results of the proposed classifier are compared with both CZ and DZ, as well as QC data, and analyzed from the view point of output performance.

• Atmosphere
• /
• v.15 no.3
• /
• pp.155-165
• /
• 2005

The characteristics of six precipitation systems occurred around Cheongju in 2002 are analyzed after the convective/stratiform radar echo classification using radar reflectivity from the Meteorological Research Institute"s X-band Doppler weather radar. The Biggerstaff and Listemaa (2000) algorithm is applied for the classification and reveals a physical characteristics of the convective and stratiform rain diagnosed from the three-dimensional structure of the radar reflectivity. The area satisfying the vertical profile of radar reflectivity is well classified, while the area near the radar site and the topography-shielded area show a mis-classification. The seasonal characteristics of the precipitation system are also analyzed using the contoured frequency by altitude diagrams (CFADs). The heights of maximum reflectivity are 4 km and 5.5 km in spring and summer, respectively, and the vertical gradient of radar reflectivity from 1.5 km to the melting layer in spring is larger than in summer.

• Korean Journal of Remote Sensing
• /
• v.26 no.4
• /
• pp.373-385
• /
• 2010

A new method to mosaic reflectivity over the overlapped coverage of two radars was developed. The method mosaics the radar reflectivity with weights after adjustment of reflectivity differences on overlapped coverage of neighboring two radars. Their weights are inverse proportion to the difference between the height of an interpolated reflectivity and the level of CAPPI (Constant Altitude PPI). The performance of this method was compared to different mosaic methods (Mosaics by maximum value, averaged value, nearest value and distance weighted value) using the reflectivity fields of a typhoon event observed by two radar. New method was better than any other methods either as a continuity and as a bias analysis of reflectivity at the boundaries in overlapped coverage by two radars.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2008.05a
• /
• pp.997-1000
• /
• 2008

기상 레이더와 지상강우계를 이용한 실시간 강우산정기법은 전형적인 Marshall-Palmer(M-P) 방법, geostatistic 접근법을 이용한 방법, 회귀분석에 의한 방법, Kalman filter를 이용한 방법 및 실시간 weight mask를 이용한 보정 등 여러 형태가 존재한다. 본 연구에서는 실시간 강우산정을 위한 각 방법의 장단점 및 적용성을 분석하였다. 전형적인 M-P 방법은 잘 알려진 바와 같이 호우사상을 과소 추정하는 단점을 가졌으며 기존 연구자들이 제시한 바와 같이 층운형, 대류형과 같은 강우형태에 따라 다른 Z-R관계식을 가지므로 단일 Z-R관계식으로 강수를 산정함에 있어 한계를 가진다. Geostatistic 기법을 이용한 실시간 강수 산정의 경우, 지상 강우계 정보를 활용하여 강우공간분포를 개선하는 여러 기법 즉 cokriging, external drift 기법 등이 존재함에도 불구하고 과다한 계산시간, 실시간 variogram 산정과 적용상의 문제 등을 내포하고 있다. 실시간 회귀분석을 이용한 강우산정은 실제 적용에 있어 지상 강우계와 레이더 반사도사이의 선형 상관관계에 대한 결정계수가 매우 낮아 기법 적용이 간단한 장점에도 불구하고 적용에 한계를 가진다. Kalman filter기법을 이용한 실시간 레이더 강수산정은 계산시간이 여타 기법보다 많이 소요되어 실시간성을 유지하는데 한계를 가진다. 실시간 weight mask를 이용한 보정기법은 지상강우계 강우강도와 기상레이더 강우강도가 선형상관관계를 가진다는 가정이 대상지역 전체에 균일하게 적용될 수 없음에도 불구하고 기법의 적용이 간편하며 실시간 강우 공간분포를 실제 강우 관측인 지상 강우계 공간 분포 특성을 간접 강우 관측인 기상 레이더 반사도 분포와 결합하여 공간 변화 특성을 잘 나타낸다는 장점을 가지므로 실용적 적용에 있어 장점을 가진다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2016.05a
• /
• pp.226-226
• /
• 2016

최근 이상기상현상과 기후변화로 인하여 국지적인 집중호우의 빈도 및 규모가 증가하고 있으며, 이로 인한 돌발 홍수피해가 증가하고 있다. 이러한 홍수 피해를 줄이기 위해서는 정확도가 우수한 초단시간(1~2시간 이내) 예측 강우량 정보가 필요하다. 본 연구에서는 집중호우에 대한 초단시간예보 및 실황 예측을 위해 시공간적으로 고해상도 자료를 제공할 수 있는 기상레이더 강우자료와 위성영상 자료를 결합하여 초단기 강수 예측기법 개발 연구를 수행하였다. 또한 기상레이더 강우량은 지상강우관측에 비해 정확성이 낮고, 많은 불확실성을 포함하고 있으므로, 위성영상에서 산출되는 강우자료와 결합하여 강우추정의 정확도를 개선하고자 하였다. 레이더 볼륨자료에서 반사도 자료를 추출하여, 1.5km CAPPI(Constant Altitude Plan Position Indicator) 자료를 생성하고, 반사도 CAPPI 자료의 패턴 상관분석을 통하여 강우시스템의 최적 이동벡터를 산출하였다. 또한 이동벡터를 고려하여 시공간적으로 외삽하여 강우이동 예측 모델을 개발하고, 초기자료로 레이더와 천리안 위성(Communication, Ocean and Meteorological Satellite, COMS) 영상자료에서 생성되는 강우자료를 결합한 강수장 자료를 이용하여 강수 예측장을 생성하였다. 레이더-위성 결합 초단기 강우예측 모델의 정확성 검증을 위하여 2014년 8월 25일 부산 및 영남 지역에 발생한 집중호우 사례에 대하여 지상기상자동관측시스템(Automatic Weather System, AWS) 강우 측정 결과를 비교 분석 하였으며, 그 적용 가능성을 검증하였다. 초단기 강우예측 분석 결과 지상강우자료와의 오차가 발생하나, 추후 여러 통계적 후처리 과정을 통하여 그 성능이 개선될 것으로 보이며, 보다 정확한 강우량 예측을 위해서는 지속적인 알고리즘 개선 및 모형의 검 보정이 필요할 것으로 사료된다.

• Korean Journal of Remote Sensing
• /
• v.37 no.5_3
• /
• pp.1405-1423
• /
• 2021

Precipitation is one of the main factors that affect water and energy cycles, and its estimation plays a very important role in securing water resources and timely responding to water disasters. Satellite-based quantitative precipitation estimation (QPE) has the advantage of covering large areas at high spatiotemporal resolution. In this study, machine learning-based rainfall intensity models were developed using Himawari-8 Advanced Himawari Imager (AHI) water vapor channel (6.7 ㎛), infrared channel (10.8 ㎛), and weather radar Column Max (CMAX) composite data based on random forest (RF). The target variables were weather radar reflectivity (dBZ) and rainfall intensity (mm/hr) converted by the Z-R relationship. The results showed that the model which learned CMAX reflectivity produced the Critical Success Index (CSI) of 0.34 and the Mean-Absolute-Error (MAE) of 4.82 mm/hr. When compared to the GeoKompsat-2 and Precipitation Estimation from Remotely Sensed Information Using Artificial Neural Networks (PERSIANN)-Cloud Classification System (CCS) rainfall intensity products, the accuracies improved by 21.73% and 10.81% for CSI, and 31.33% and 23.49% for MAE, respectively. The spatial distribution of the estimated rainfall intensity was much more similar to the radar data than the existing products.

• Journal of Korea Water Resources Association
• /
• v.53 no.12
• /
• pp.1159-1172
• /
• 2020

This study applied deep convolution neural network based on U-Net and SegNet using long period weather radar data to very short-term rainfall prediction. And the results were compared and evaluated with the translation model. For training and validation of deep neural network, Mt. Gwanak and Mt. Gwangdeoksan radar data were collected from 2010 to 2016 and converted to a gray-scale image file in an HDF5 format with a 1km spatial resolution. The deep neural network model was trained to predict precipitation after 10 minutes by using the four consecutive radar image data, and the recursive method of repeating forecasts was applied to carry out lead time 60 minutes with the pretrained deep neural network model. To evaluate the performance of deep neural network prediction model, 24 rain cases in 2017 were forecast for rainfall up to 60 minutes in advance. As a result of evaluating the predicted performance by calculating the mean absolute error (MAE) and critical success index (CSI) at the threshold of 0.1, 1, and 5 mm/hr, the deep neural network model showed better performance in the case of rainfall threshold of 0.1, 1 mm/hr in terms of MAE, and showed better performance than the translation model for lead time 50 minutes in terms of CSI. In particular, although the deep neural network prediction model performed generally better than the translation model for weak rainfall of 5 mm/hr or less, the deep neural network prediction model had limitations in predicting distinct precipitation characteristics of high intensity as a result of the evaluation of threshold of 5 mm/hr. The longer lead time, the spatial smoothness increase with lead time thereby reducing the accuracy of rainfall prediction The translation model turned out to be superior in predicting the exceedance of higher intensity thresholds (> 5 mm/hr) because it preserves distinct precipitation characteristics, but the rainfall position tends to shift incorrectly. This study are expected to be helpful for the improvement of radar rainfall prediction model using deep neural networks in the future. In addition, the massive weather radar data established in this study will be provided through open repositories for future use in subsequent studies.