• Journal of Korea Water Resources Association
• /
• v.54 no.5
• /
• pp.301-309
• /
• 2021

In this study, we tried to improve the performance of the existing U-net-based deep learning rainfall prediction model, which can weaken the meaning of time series order. For this, ConvLSTM2D U-Net structure model considering temporal consistency of data was applied, and we evaluated accuracy of the ConvLSTM2D U-Net model using a RainNet model and an extrapolation-based advection model. In addition, we tried to improve the uncertainty in the model training process by performing learning not only with a single model but also with 10 ensemble models. The trained neural network rainfall prediction model was optimized to generate 10-minute advance prediction data using four consecutive data of the past 30 minutes from the present. The results of deep learning rainfall prediction models are difficult to identify schematically distinct differences, but with ConvLSTM2D U-Net, the magnitude of the prediction error is the smallest and the location of rainfall is relatively accurate. In particular, the ensemble ConvLSTM2D U-Net showed high CSI, low MAE, and a narrow error range, and predicted rainfall more accurately and stable prediction performance than other models. However, the prediction performance for a specific point was very low compared to the prediction performance for the entire area, and the deep learning rainfall prediction model also had limitations. Through this study, it was confirmed that the ConvLSTM2D U-Net neural network structure to account for the change of time could increase the prediction accuracy, but there is still a limitation of the convolution deep neural network model due to spatial smoothing in the strong rainfall region or detailed rainfall prediction.

• Journal of Korea Water Resources Association
• /
• v.53 no.12
• /
• pp.1211-1220
• /
• 2020

This research detects the features of the bright band (BB) through analysis of the vertical profile of range height indicator (RHI) and the slant range beam profile of plane position indicator (PPI) of the polarimetric radar measurements-horizontal reflectivity (ZH), differential reflectivity (ZDR), and cross-correlation coefficient (ρHV). As a result of the analysis, it is possible to clearly detect the bright band using the polarimetric radar measurements, and it is confirmed that the result is consistent by double searching for the BB using the RHI and PPI scan data at the same time. Based on these results, the accuracy of QPE (quantification of precipitation estimation) can be improved by applying the BB search method by the PPI slant range in this research to large rainfall radars that only scan PPI volumes in the field without RHI observations.

• 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.

• Journal of the Korean Society of Hazard Mitigation
• /
• v.11 no.3
• /
• pp.17-28
• /
• 2011

This study applied the Kalman Filter for real-time forecasting the G/R (ground rain gauge rainfall/radar rainfall) ratio to correct the mean field bias of the very-short-range-forecast (VSRF) rainfall. The MAPLE-forecasted rainfall was used as the VSRF rainfall, also the methodology for deciding the G/R ratio was improved by evaluating the change of G/R ratio characteristics depending on the threshold and accumulation time. This analysis was done for the inland, mountain, and coastal regions, separately, for their comparison. As the results, more stable G/R ratio could be estimated by applying the threshold and accumulation time, whose forecasting accuracy could also be secured. The accuracy of the corrected rainfall forecasting by the forecasted G/R ratio was the best in the inland region but the worst in the coastal region.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.5
• /
• pp.819-826
• /
• 2020

The UHF band wind profiler radars of the Korea Meteorological Administration (KMA), which produces the vertical profile of the wind, need to be calibrated for better performance. The capabilities of the radar in detecting even light precipitation were used for the calibration of which reference takes the hourly series of ground rainfall rate measured by a rain gauge at the radar site. This calibration must be renewed regularly according to the methodology implemented in this work since errors occur on the wind vectors in the clear sky without reflectivity calibration. Comparing the wind by wind profiler with that by radiosonde, the optimal radar constant contributed to the improvement of wind accuracy.

• Journal of Korea Water Resources Association
• /
• v.48 no.5
• /
• pp.409-423
• /
• 2015

In this study the very short-term rainfall forecasting and storm water forecasting using the weather radar data were implemented in an urban stream basin. As forecasting time increasing, the very short-term rainfall forecasting results show that the correlation coefficient was decreased and the root mean square error was increased and then the forecasting model accuracy was decreased. However, as a result of the correlation coefficient up to 60-minute forecasting time is maintained 0.5 or higher was obtained. As a result of storm water forecasting in an urban area, the reduction in peak flow and outflow volume with increasing forecasting time occurs, the peak time was analyzed that relatively matched. In the application of storm water forecasting by radar rainfall forecast, the errors has occurred that we determined some of the external factors. In the future, we believed to be necessary to perform that the continuous algorithm improvement such as simulation of rapid generation and disappearance phenomenon by precipitation echo, the improvement of extreme rainfall forecasting in urban areas, and the rainfall-runoff model parameter optimizations. The results of this study, not only urban stream basin, but also we obtained the observed data, and expand the real-time flood alarm system over the ungaged basins. In addition, it is possible to take advantage of development of as multi-sensor based very short-term rainfall forecasting technology.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.21 no.4
• /
• pp.191-201
• /
• 2018

Recently, the frequency of heavy rainfall is increasing due to the effects of climate change, and heavy rainfall in urban areas has an unexpected and local characteristic. Floods caused by localized heavy rains in urban areas occur rapidly and frequently, so that life and property damage is also increasing. It is crucial how fast and precise observations can be made on successful flood management in urban areas. Local heavy rainfall is predominant in low-level storms, and the present large-scale radars are vulnerable to low-level rainfall detection and observations. Therefore, it is necessary to introduce a new urban flood forecasting system to minimize urban flood damage by upgrading the urban flood response system and improving observation and forecasting accuracy by quickly observing and predicting the local storm in urban areas. Currently, the WHAP (Water Hazard Information Platform) Project is promoting the goal of securing new concept water disaster response technology by linking high resolution hydrological information with rainfall prediction and urban flood model. In the WHAP Project, local rainfall detection and prediction, urban flood prediction and operation technology are being developed based on high-resolution small radar for observing the local rainfall. This study is expected to provide more accurate and detailed urban flood warning system by enabling high-resolution observation of urban areas.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.14 no.1
• /
• pp.61-68
• /
• 2019

In addition to non-meteorological echoes, meteorological echoes with large scattering effects, such as precipitation, cause errors in wind data measured by wind profiler. In the rainfall situation, the Doppler spectrum of wind profiler shows both the rainfall signal and the atmospheric signal as two peaks. The vertical radial velocity is very large due to the falling rain drop. The radial velocity contaminated by rainfall decreases the accuracy of the horizontal wind vector and leads to inaccurate weather analysis. In this study, we developed an algorithm to process raw data of wind profiler and distinguished rainfall signal and wind signal by partitioning bimodal peak for Doppler spectrum in rainfall environment.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.267-267
• /
• 2023

최근 이상기상현상과 기후변화로 인하여 국지적인 집중호우의 빈도 및 규모가 증가하고 있으며, 이로 인한 돌발 홍수 피해가 증가하고 있다. 레이더는 넓은 영역에 대해 고해상도의 강우 정보를 제공할 수 있으므로 위험기상 감시 및 실황 예측 모형의 입력자료로써 활용도가 높다. 레이더 강우량은 대기 중 강수입자에 대한 레이더 반사도와 강우강도의 Z-R 관계식으로 추정되기 때문에 강우 추정 과정에 불확실성을 내포하고 있다. 특히, 우리나라의 여름철 한반도의 집중호우는 층운형 강우와 함께 대류형 강우가 동반되는 복합적인 강우시스템에서 자주 발생하지만, 레이더 강우는 일반적으로 단일 강우시스템에 대한 고정된 Z-R 관계식으로 추정하므로, 이러한 현상에 대해 과대 추정 혹은 과소 추정이 발생한다. 본 연구에서는 집중호우에 적합한 강우를 추정하기 위해 2021년 8월 21일부터 8월 25일까지 경남 호우사례를 대상으로 층운형, 대류형, 열대형의 Z-R관계식과 반사도 조건에 따라 층운형과 적운형을 구분하여 Z-R 관계식을 적용하여 레이더 강우량 자료를 산출하였으며, 지상강우자료를 이용하여 정확도를 평가하였다. 레이더 자료 처리를 위해 Radar Software Library (RSL)를 이용하여 수평으로 1km 해상도의 1.5km CAPPI (Constant Altitude Plan Position Indicator) 자료로 변환하였다. 레이더 강우 추정의 정확도를 평가하기 위해 레이더 지점으로부터 100 km 이내에 위치하고 있는 기상관서와 자동기상관측소의 강우관측 결과와 비교·분석하였다.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.33 no.3
• /
• pp.101-109
• /
• 2021

Wave observations using a marine X-band radar are conducted by analyzing the backscattered radar signal from sea surfaces. Wave parameters are extracted using Modulation Transfer Function obtained from 3D wave number and frequency spectra which are calculated by 3D FFT of time series of sea surface images (42 images per minute). The accuracy of estimation of the significant wave height is, therefore, critically dependent on the quality of radar images. Wave observations during Typhoon Maysak and Haishen in the summer of 2020 show large errors in the estimation of the significant wave heights. It is because of the deteriorated radar images due to raindrops falling on the sea surface. This paper presents the algorithm developed to increase the accuracy of wave heights estimation from radar images by adopting convolution neural network(CNN) which automatically classify radar images into rain and non-rain cases. Then, an algorithm for deriving the Hs is proposed by creating different ANN models and selectively applying them according to the rain or non-rain cases. The developed algorithm applied to heavy rain cases during typhoons and showed critically improved results.