• Proceedings of the Korea Water Resources Association Conference
• /
• 2016.05a
• /
• pp.238-238
• /
• 2016

토양수분, 증발산량, 유출량 등의 고해상도 수문기상요소 산출을 위한 지면모델 활용 기술은 기상 및 수문분야에서 널리 활용 중에 있다. 본 연구에서는 미국 국립대기과학연구소(NCAR)에서 개발된 기상-수문 결합모델 WRF-Hydro(Weather Research and Forecasting Model Hydrological modeling extension package)을 활용하여 낙동강 유역에서 발생한 돌발홍수 사례 실험에 적용하여 강우량 및 수문기상요소 전체를 모의함으로써 기상-수문-지면 결합모델을 활용한 수문기상요소 산출하고자 하였다. 이를 기존의 기상모델로부터 입력강제자료를 제공받아 Off-line 형태로 결합된 지면모델(TOPLATS, TOPmodel-based Land Atmosphere Transfer Scheme) 결과와 비교하였고 기상-수문 결합모델의 국내 적용성을 검토하였다. 기상-수문-지면 결합모델(WRF-Hydro)의 초기장 및 경계장은 기상청 현업 모델에서 생성된 국지예보모델자료 1.5km 자료(LDAPS, Local Data Assimilation and Prediction System)를 사용하였으며, 모델의 적분기간은 돌발홍수 사례에 따라 24~36시간을 수행하였다. WRF-Hydro 모델의 물리모수화 방안은 작년까지 기상청에서 현업운영되는 KWRF의 방안들을 준용하였으며, WRF-Hydro 수행을 위해 Advanced Spaceborne Thermal Emission and Reflection Radiometer(ASTER)에서 제공되는 30 m 해상도의 수치표고자료를 GIS(Geographic Information System)를 활용하여 지표유출방향을 설정하였다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2019.05a
• /
• pp.115-115
• /
• 2019

하천수 흐름예측에 대한 연구는 대부분 WRF-Hydro와 같은 과정기반 모델링 시스템을 이용한다. 과정기반 모델링 시스템은 물리적 현상을 일반화한 수식으로 구성되어있다. 일반화된 수식은 불확실성을 내포하고 있으며 지역적 특성도 반영하지 못한다. 특히 수식에 사용되는 입력자료는 측정값으로 오차가 존재한다. 따라서 과정기반 모델링 시스템 예측결과는 계통오차와 우연오차가 존재한다. 현재 매개변수 보정을 통해 예측결과를 개선하는 방법을 사용하고 있으나 한계가 있다. 본 연구는 이러한 한계를 극복하기 위해 상호보완적인 Data-driven 모델을 구축하여 과정기반 모델링 시스템 결과를 개선하고자 하였다. Data-driven 모델 구축을 위해 머신러닝 기법인 instance-based weighting(IBW)과 support vector regression(SVR)을 사용하였다. 구축된 Data-driven 모델은 한반도 지역 주요 저수지 및 호수의 하천수 흐름예측을 통해 검증하였다. 검증을 위해 과정기반 모델링 시스템으로 WRF-Hydro를 구동하였다. 입력자료는 기상청의 국지수치예측모델자료(LDAPS), HydroSHEDS의 수치표고모델자료(DEM), 국가지리정보원의 저수지 및 호수 연속수치지형도를 사용하였다. 본 연구를 통해 구축된 Data-driven모델은 기존 과정기반 모델링 시스템의 오류수정 한계를 머신러닝을 이용하여 개선할 수 있는 가능성을 제시하였다.

• Atmosphere
• /
• v.31 no.1
• /
• pp.73-83
• /
• 2021

Numerical weather prediction (NWP) models play an essential role in predicting weather factors, but using them is challenging due to various factors. To overcome the difficulties of NWP models, deep learning models have been deployed in weather forecasting by several recent studies. This study adapts long short-term memory (LSTM), which demonstrates remarkable performance in time-series prediction. The combination of LSTM model input of meteorological features and activation functions have a significant impact on the performance therefore, the results from 5 combinations of input features and 4 activation functions are analyzed in 9 Automated Surface Observing System (ASOS) stations corresponding to cities/islands/mountains. The optimized LSTM model produces better performance within eight forecast hours than Local Data Assimilation and Prediction System (LDAPS) operated by Korean meteorological administration. Therefore, this study illustrates that this LSTM model can be usefully applied to very short-term weather forecasting, and further studies about CNN-LSTM model with 2-D spatial convolution neural network (CNN) coupled in LSTM are required for improvement.

• Atmosphere
• /
• v.29 no.4
• /
• pp.355-365
• /
• 2019

Based on in-situ monitoring data produced by National Institute of Meteorological Sciences, we evaluated the performance of Joint UK Land Environment Simulator (JULES) on the surface energy balance for rice-paddy and cropland in Korea with the operational ancillary data used for Unified Model (UM) Local Data Assimilation and Prediction System (LDAPS) (CTL) and the high-resolution ancillary data from external sources (EXP). For these experiments, we employed the one-year (March 2015~February 2016) observations of eddy-covariance fluxes and soil moisture contents from a double-cropping rice-paddy in BoSeong and a cropland in AnDong. On the rice-paddy site the model performed better in the CTL experiment except for the sensible heat flux, and the latent heat flux was underestimated in both of experiments which can be inferred that the model represents flood-irrigated surface poorly. On the cropland site the model performance of the EXP experiment was worse than that of CTL experiment related to unrealistic surface type fractions. The pattern of the modeled soil moisture was similar to the observation but more variable in time. Our results shed a light on that 1) the improvement of land scheme for the flood-irrigated rice-paddy and 2) the construction of appropriate high-resolution ancillary data should be considered in the future research.

• Atmosphere
• /
• v.30 no.1
• /
• pp.17-30
• /
• 2020

KMA performed the special observation program to provide information about severe weather and to monitor typhoon PRAPIROON using the ship which called the Gisang 1 from 29 June 2018 to 4 July 2018 (UTC). For this period, upper-air was observed 21 times with 6 hour intervals using rawinsonde in the Gisang 1. We investigated the impact of upper-air observation data from the Gisang 1 on the performance of the operational convective scale model (we called LDAPS). We conducted two experiments that used all observation data including upper-air observation data from the Gisang 1 (OPER) and without it (EXPR). For a typhoon PRAPIROON case, track forecast error of OPER was lower than EXPR until forecast 24 hours. The intensity forecast error of OPER for minimum sea level pressure was lower than EXPR until forecast 12 hours. The intensity forecast error of OPER for maximum wind speed was mostly lower than EXPR until forecast 30 hours. OPER showed good performance for typhoon forecast compared with EXPR at the early lead time. Two precipitation cases occurred in the south of the Korean peninsula due to the impact of Changma on 1 July and typhoon on 3 July. The location of main precipitation band predicted from OPER was closer to observations. As assimilating upper-air data observed in the Gisang 1 to model, it showed positive results in typhoon and precipitation cases.

• Korean Journal of Remote Sensing
• /
• v.36 no.4
• /
• pp.573-586
• /
• 2020

PM (particulate matter) is of interest to everyone because it can have adverse effects on human health by the infiltration from respiratory to internal organs. To date, many studies have made efforts for the prediction of PM10 and PM2.5 concentrations. Unlike previous studies, we conducted the prediction of tomorrow's PM10 concentration for the Air Korea stations using Chinese PM10 data in addition to the satellite AOD and weather variables. We constructed 230,639 matchups from the raw data over 3 million and built an RF (random forest) model from the matchups to cope with the complexity and nonlinearity. The validation statistics from the blind test showed excellent accuracy with the RMSE (root mean square error) of 9.905 ㎍/㎥ and the CC (correlation coefficient) of 0.918. Moreover, our prediction model showed a stable performance without the dependency on seasons or the degree of PM10 concentration. However, part of coastal areas had a relatively low accuracy, which implies that a dedicated model for coastal areas will be necessary. Additional input variables such as wind direction, precipitation, and air stability should also be incorporated into the prediction model as future work.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2019.05a
• /
• pp.305-305
• /
• 2019

태풍에 의한 재해는 우리나라에서 발생하는 자연재해 중 발생빈도가 가장 높은 것으로 나타나며, 최근 들어 태풍 및 집중호우로 인한 홍수가 급증하고 있는 실정이다. 최근에는 치수증대사업으로 하천 범람의 재해가 감소하는 추세이지만, 도시지역의 경우 도시개발에 따른 내수 범람 피해가 증가하고 있고, 산지에서는 토석류 등의 토사 재해가 증가하고 있다. 이러한 홍수피해를 경감하기 위해서는 치수사업 등과 같은 구조적인 대책도 필요하지만, 정확한 홍수 예 경보를 통한 대비시간의 확보 등과 같은 비구조적인 대책도 중요하며, 홍수 예 경보를 통한 선행시간(Lead time)확보를 위해 강우 및 홍수예측 시스템 구축이 하나의 대안으로 대두되고 있다. 강우예측 기법으로는 레이더(Radar)를 통해 관측된 자료를 외삽하는 초단기 강우예측기법이 최근까지 많이 수행되어 왔다. 하지만 컴퓨터 계산 능력이 향상되면서 수치예보(Numerical Weather Prediction; NWP) 모델을 이용한 강우예측 및 수문학적 적용에 관한 연구들이 대두되고 있다. 본 연구에서는 수치예보모델을 이용하여 기상 및 수자원 간의 연계를 통한 강우 및 홍수 예측에 활용방안을 검토하기 위해 한국 기상청에서 제공하는 국지예보모델(LDAPS)과 예측 도메인에 한국을 포함하는 일본 기상청의 중규모 모델(MSM)을 이용하여 남강댐 유역 내 산청 유역에 대해 강우 및 홍수 예측 정확도를 평가하고 비교 검토하였다. 본 연구에서 적용한 LDAPS와 MSM은 사용하는 수치모델, 물리과정 매개변수, 자료동화 기법 및 지배 방정식 등이 다르기 때문에 직접적인 비교를 하는데 무리가 있지만 국내의 강우 및 홍수 예측 분야에서의 각 수치예보모델의 활용성을 검토하고자 한다.

• Atmosphere
• /
• v.31 no.4
• /
• pp.421-431
• /
• 2021

The Yeongdong region is frequently vulnerable to heavy snowfall in winter in terms of societal and economical damages. By virtue of a lot of previous efforts, snowfall forecast has been significantly improved, but the performance of light snowfall forecast is still poor since it is very conducive to synoptic and mesoscale interactions, largely attributable to Taeback mountains and East Sea effects. An intensive observation has been made in cooperation with Gangwon Regional Meteorological Office and National Institute of Meteorological Studies in winter seasons since 2019. Two distinctive Cold Air Damming (CAD) events (14 February 2019 and 6 February 2020) were observed for two years when the snowfall forecast was wrong specifically in its location and timing. For two CAD events, lower-level temperature below 2 km ranged to lowest limit in comparisons to those of the previous 6-years (2014~2019) rawinsonde soundings, along with the stronger inversion strength (> 2.0℃) and thicker inversion depth (> 700 m). Further, the northwesterly was predominant within the CAD layer, whereas the weak easterly wind was exhibited above the CAD layer. For the CAD events, strong cold air accumulation along the east side of Taeback Mountains appeared to prevent snow cloud and convergence zone from penetrating into the Yeongdong region. We need to investigate the influence of CAD on snowfall in the Yeongdong region using continuous intensive observation and modeling studies altogether. In addition, the effect of synoptic and mesoscale interactions on snowfall, such as nighttime drainage wind and land breeze, should be also examined.

• Atmosphere
• /
• v.31 no.1
• /
• pp.17-28
• /
• 2021

We investigated the impact of domain size on the simulated summer precipitation over the Korean Peninsula using the Weather Research and Forecasting (WRF) model. Two different domains are integrated up to 72-hours from 29 June 2017 to 28 July 2017 when the Changma front is active. The domain sizes are adopted from previous RDAPS (Regional Data Assimilation and Prediction System) and current LDAPS (Local Data Assimilation and Prediction System) operated by the Korea Meteorological Administration, while other model configurations are fixed identically. We found that the larger domain size showed better prediction skills, especially in precipitation forecast performance. This performance improvement is particularly noticeable over the central region of the Korean Peninsula. Comparisons of physical aspects of each variable revealed that the inflow of moisture flux from the East China Sea was well reproduced in the experiment with a large model domain due to a more realistic North Pacific high compared to the small domain experiment. These results suggest that the North Pacific anticyclone could be an important factor for the precipitation forecast during the summer-time over the Korean Peninsula.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.50 no.1
• /
• pp.67-74
• /
• 2022

To analyze the motion of aircraft through computing the dynamics equations, true airspeed is essential for obtaining aerodynamic loads. Although the airspeed is measured by on-board instruments such as pitot tubes, measurement data are difficult to obtain for commercial flights because they include sensitive data about the airline operations. One of the commonly available trajectory data, Automatic Dependent Surveillance-Broadcast data, provide aircraft's speed in the form of ground speed. The ground speed is a vector sum of the local wind velocity and the true airspeed. This paper present a method to estimate true airspeed by combining the trajectory, meteorological, and topology data available to the public. To integrate each data, we first matched the coordinate system and then unified the altitude reference to the mean sea level. We calculated the wind vector for all trajectory points by interpolating from the lower resolution grid of the meteorological data. Finally, we calculate the true airspeed from the ground speed and the wind vector. These processes were applied to several sample trajectories with corresponding meteorological data and the topology data, and the estimated true airspeeds are presented.