Regional Data Assimilation And Prediction System Research Articles

Optimization of SWAN Wave Model to Improve the Accuracy of Winter Storm Wave Prediction in the East Sea

In recent years, as human casualties and property damage caused by hazardous waves have increased in the East Sea, precise wave prediction skills have become necessary. In this study, the Simulating WAves Nearshore (SWAN) third-generation numerical wave model was calibrated and optimized to enhance the accuracy of winter storm wave prediction in the East Sea. We used Source Term 6 (ST6) and physical observations from a large-scale experiment conducted in Australia and compared its results to Komen’s formula, a default in SWAN. As input wind data, we used Korean Meteorological Agency's (KMA’s) operational meteorological model called Regional Data Assimilation and Prediction System (RDAPS), the European Centre for Medium Range Weather Forecasts’ newest 5th generation re-analysis data (ERA5), and Japanese Meteorological Agency's (JMA’s) meso-scale forecasting data. We analyzed the accuracy of each model’s results by comparing them to observation data. For quantitative analysis and assessment, the observed wave data for 6 locations from KMA and Korea Hydrographic and Oceanographic Agency (KHOA) were used, and statistical analysis was conducted to assess model accuracy. As a result, ST6 models had a smaller root mean square error and higher correlation coefficient than the default model in significant wave height prediction. However, for peak wave period simulation, the results were incoherent among each model and location. In simulations with different wind data, the simulation using ERA5 for input wind datashowed the most accurate results overall but underestimated the wave height in predicting high wave events compared to the simulation using RDAPS and JMA meso-scale model. In addition, it showed that the spatial resolution of wind plays a more significant role in predicting high wave events. Nevertheless, the numerical model optimized in this study highlighted some limitations in predicting high waves that rise rapidly in time caused by meteorological events. This suggests that further research is necessary to enhance the accuracy of wave prediction in various climate conditions, such as extreme weather.

The Spatial Maps of Paddy Rice Yield over Northeast Asia Using COMS Geostationary Satellite and Reanalysis Meteorological Data

This study estimated rice yield maps for Northeast Asia by using the Communication, Ocean and Meteorological satellite (COMS), Terra satellite, and Regional Data Assimilation and Prediction System (RDAPS) of the numerical model. The rice yield is highly useful in the study for crop information monitoring according to climate change as well as agriculture information, industry, and economy. This study produced rice yield maps for Northeast Asia including Korea, North Korea, Japan, and three northeastern provinces of China (Heilongjiang, Jilin, and Liaoning) from 2011 to 2017. The estimated spatial resolution of the rice yield maps in Northeast Asia is 500 m. The spatial observation range is 25 ° N ~ 47 ° N and 115 ° E ~ 145 ° E. In order to estimate rice yield, Remote Sensing-integrated Crop Model was employed in this study. The inputs of the RSCM are vegetation indices from Geostationary Ocean Color Imager (GOCI) of the COMS, solar radiation from Meteorological Imager of the COMS, Land Surface Water Index from the MODerate Resolution Imaging Spectroradiometer, and the temperature from the RDAPS were considered as input data. In particular, this study applied the Bidirectional Reflectance Distribution Function to the GOCI time-series images to calculate more improved vegetation indices by minimizing the directional error generated in the satellite observation location. These indices were very effective in the simulation of the rice yield.

Temperature Prediction Based on Bidirectional Long Short-Term Memory and Convolutional Neural Network Combining Observed and Numerical Forecast Data.

Weather is affected by a complex interplay of factors, including topography, location, and time. For the prediction of temperature in Korea, it is necessary to use data from multiple regions. To this end, we investigate the use of deep neural-network-based temperature prediction model time-series weather data obtained from an automatic weather station and image data from a regional data assimilation and prediction system (RDAPS). To accommodate such different types of data into a single model, a bidirectional long short-term memory (BLSTM) model and a convolutional neural network (CNN) model are chosen to represent the features from the time-series observed data and the RDAPS image data. The two types of features are combined to produce temperature predictions for up to 14 days in the future. The performance of the proposed temperature prediction model is evaluated by objective measures, including the root mean squared error and mean bias error. The experiments demonstrated that the proposed model combining both the observed and RDAPS image data is better in all performance measures for all prediction periods compared with the BLSTM-based model using observed data and the CNN-BLSTM-based model using RDAPS image data alone.

Characteristics of Meteorological Variables in the Leeward Side associated with the Downslope Windstorm over the Yeongdong Region

We investigated the characteristics of meteorological conditions related to the strong downslope wind over the leeward side of the Taebaek Mountains during the period 2005~2010. The days showing the strong wind exceeding in Gangwon province were selected as study cases. A total of 15 days of strong wind were observed at Sokcho, Gangneung, Donghae, and Taebaek located over the Yeongdong region. Seven cases related to tropical cyclone (3 cases) and heavy snowfall (2 cases) and heavy rainfall (2 cases) over the Yeongdong region were excluded. To investigate the characteristics of the remaining 8 cases, we used synoptic weather chart, Sokcho radiosonde, Gangneung wind profiler and numerical model. The cases showed no precipitation (or ). From the surface and upper level weather chart, we found the pressure distribution of southern high and northern low pattern over the Korean peninsula and warm ridge over the Yeongdong region. Inversion layer (or stable layer) and warm ridge with strong wind were located in about 1~3 km (925~700 hPa) over mountains. The Regional Data Assimilation and Prediction System (RDAPS) indicated that warm core and temperature ridge with horizontal temperature gradient were which were located on 850 hPa pressure level above mountaintop. These results were summarized as a forecasting guidance of downslope windstorm in the Yeongdong region.

Characteristics of easterly-induced snowfall in Yeongdong and its relationship to air-sea temperature difference

Characteristics of snowfall episodes have been investigated for the past ten years in order to study its association with lowlevel stability and air-sea temperature difference over the East Sea. In general, the selected snowfall episodes have similar synoptic setting such as the Siberian High extended to northern Japan along with the Low passing by the southern Korean Peninsula, eventually resulting in the easterly flow in the Yeongdong region. Especially in the heavy snowfall episodes, convective unstable layers have been identified over the East sea due to relatively warm sea surface temperature (SST) about 8∼10°C and specifically cold pool around 1∼2 km above the surface level (ASL), which can be derived from Regional Data Assimilation and Prediction System (RDAPS), but that have not been clearly exhibited in the weak snowfall episodes. The basic mechanism to initiate snowfall around Yeongdong seems to be similar to that of lake-effect snowstorms around Great Lakes in the United States (Kristovich et al., 2003). Difference of equivalent potential temperature (θ e ) between 850 hPa and surface as well as difference between air and sea temperatures altogether gradually began to increase in the pre-snowfall period and reached their maximum values in the course of the period, whose air (850 hPa) — sea temperature difference and snowfall intensity in case of the heavy snowfall episodes are almost larger than 20°C and 6 tims greater than the weak snowfall episodes, respectively. Interestingly, snowfall appeared to begin in case of an air-sea temperature difference exceeding over 15°C. The current analysis is overall consistent with the previous finding (Lee et al., 2012) that an instabilityinduced moisture supply to the lower atmosphere from the East sea, being cooled and saturated in the lower layer, so to speak, East Sea-Effect Snowfall (SES), would make a low-level ice cloud which eventually moves inland by the easterly flow. In addition, a longlasting synoptic characteristics and convergence-induced invigoration also appear to play the important roles in the severe snowstorms. Improvements in our understanding of mesoscale sea-effect snowstorms require detailed in-situ and remote sensing observations over and around East Sea since observations of the concurrent thermodynamic and microphysical characteristics have not been available there and this study emphasizes the importance of low level stability as quantitative estimation of moist static energy generation over the East Sea.

광역 위성 영상과 수치예보자료를 이용한 여름철 강수량 예측

본 연구에서는 지상의 관측 자료와 광역의 정보를 제공하는 수치 예보 모형 자료 및 인공위성 자료를 이용하고 자료와 강수예측치의 물리적 상관 특성을 나타내기 위하여 자료 사이의 비선형 거동을 잘 나타내는 신경망 모형에 적용시켜 단시간 강수 예측을 수행하였다. 이를 위하여 서울지점에 대하여 현재로부터 3시간, 6시간, 9시간, 12시간의 선행시간을 가지는 인공위성자료(MTSAT-1R) 및 수치 예보 모형 자료(RDAPS, Regional Data Assimilation and Prediction System)와 실시간 전송되는 자동 기상 관측 시스템(AWS, Automatic Weather System)의 관측치를 신경망 모형의 입력 자료로 하여 3시간, 6시간, 9시간, 12시간의 선행시간을 가지는 자료로 강수를 예측 할 수 있는 강수 예측 모형을 개발하였다. 장마와 태풍과 같이 전선형강수와 선풍형강수 등 강수 양상의 차이를 고려하기 위하여 6월, 7월과 8월, 9월 자료를 구분하여 신경망을 구축하였으며, 자료가용성에 기초하여 2006년에서 2008년 기간 동안에 대하여 모형을 학습하고 2009년에 대하여 모형의 적용성을 검증한 결과, 단시간 강수예측에 대한 모형의 적용 가능성을 보여주었으나 다양한 광역 자료와 인공신경망을 사용함에도 불구하고 단시간 강수예측의 정량적 정도향상을 위한 여지가 많음을 보여준다. In this study, satellite data (MTSAT-1R), a numerical weather prediction model, RDAPS (Regional Data Assimilation and Prediction System) output, ground weather station data, and artificial neural networks were used to improve the accuracy of summer rainfall forecasts. The developed model was applied to the Seoul station to forecast the rainfall at 3, 6, 9, and 12-hour lead times. Also to reflect the different weather conditions during the summer season which is related to the frontal precipitation and the cyclonic precipitation such as Jangma and Typhoon, the neural network models were formed for two different periods of June-July and August-September respectively. The rainfall forecast model was trained during the summer season of 2006 and 2008 and was verified for that of 2009 based on the data availability. The results demonstrated that the model allows us to get the improved rainfall forecasts until lead time of 6 hour, but there is still a large room to improve the rainfall forecast skill.

RDAPS 모델의 강수량과 실측강수량의 비교를 통한 적용성 검토

최근 우리나라뿐만 아니라 전 세계적으로 기후변화에 따른 국지성 호우의 증가로 피해가 증가하고 있고 그에 따른 대책으로 단기예보의 중요성이 높아지고 있다. 우리나라의 기상청에서는 전구 모형인 GDAPS와 지역 모델인 RDAPS를 이용하여 정량적 예보인 수치 예보를 하고 있다. RDAPS 모델을 이용하여 생성된 자료는 3시간의 누적 강수량으로 48시간에 대한 예측 자료를 12시간 간격으로 00UTC와 12UTC를 생성한다. 본 연구에서는 2005년의 RDAPS의 결과를 9가지의 경우에 대하여 일강수량으로 변환하였으며, 금강유역을 대상으로 면적평균강수량 (MAP)을 산정하였다. 또한, 기상청 강우관측소의 관측강우량과 절대상대오차평균 (AARE)을 산정하여 가장 정확한 변환의 방법을 제시하고 적용성을 검토하였다. The climate change has been observed in Korea as well as in the entire world recently. The rainstorm has been gradually increased and then the damage has been grown. It is getting important to predict short-term rainfall. The Korea Meteorological Administration (KMA) generates numerical model outputs which are computed by Global Data Assimilation and Prediction System (GDAPS) and Regional Data Assimilation and Prediction System (RDAPS). The KMA predicts rainfall using RDAPS results. RDAPS model generates 48 hours data which is organized 3 hours data accumulated at 00UTC and 12UTC. RDAPS results which are organized 3 hours time scale are converted into daily rainfall to compare observed daily rainfall. In this study, 9 cases are applied to convert RDAPS results to daily rainfall data. The MAP (mean areal precipitation) in Geum river basin are computed by using KMA which are 2005 are used. Finally, the best case which gives the close value to the observed rainfall data is obtained using the average absolute relative error (AARE) especially for the Geum River basin.

Evaluations of Upper-Level Turbulence Diagnostics Performance Using the Graphical Turbulence Guidance (GTG) System and Pilot Reports (PIREPs) over East Asia

AbstractThe forecast skill of upper-level turbulence diagnostics is evaluated using available turbulence observations [viz., pilot reports (PIREPs)] over East Asia. The six years (2003–08) of PIREPs used in this study include null, light, and moderate-or-greater intensity categories. The turbulence diagnostics used are a subset of indices in the Graphical Turbulence Guidance (GTG) system. To investigate the optimal performance of the component GTG diagnostics and GTG combinations over East Asia, various statistical evaluations and sensitivity tests are performed. To examine the dependency of the GTG system on the operational numerical weather prediction (NWP) model, the GTG system is applied to both the Regional Data Assimilation and Prediction System (RDAPS) analysis data and Global Forecasting System (GFS) analysis and forecast data with 30-km and 0.3125° (T382) horizontal grid spacings. The dependency of the temporal variation in the PIREP and GFS data and the forecast lead time of the GFS-based GTG combination are also investigated. It is found that the forecasting performance of the GTG system varies with year and season according to the annual and seasonal variations in the large-scale atmospheric conditions over the East Asia region. The wintertime GTG skill is the highest, because most GTG component diagnostics are related to jet streams and upper-level fronts. The GTG skill improves as the number of PIREP samples and the vertical resolution of the underlying NWP analysis data increase, and the GTG performance decreases as the forecast lead time increases from 0 to 12 h.

Statistics and Possible Sources of Aviation Turbulence over South Korea

Abstract The characteristics of aviation turbulence over South Korea during the recent five years (2003–08, excluding 2005) are investigated using pilot reports (PIREPs) accumulated by the Korea Aviation Meteorological Agency (KAMA). Among the total of 8449 PIREPs, 4607 (54.53%), 1646 (19.48%), 248 (2.94%), 7 (0.08%), and 1941 (22.97%) correspond to the turbulence categories of null, light, moderate, severe, and missing, respectively. In terms of temporal variations, the annual total number of turbulence events increased from 2003 to 2008, and the seasonal frequency is the highest in the spring. With regard to spatial distributions, reported turbulence encounters are dominant along the prevailing flight routes, but are locally higher over the west coast, Jeju Island, and the Sobaek and Taebaek mountains. The turbulence events in these regions vary by season. To examine the regional differences and possible sources of the observed turbulence, lightning flash data, Regional Data Assimilation and Prediction System (RDAPS) analysis data with a 30-km horizontal grid spacing provided by the Korean Meteorological Administration (KMA), and a digital elevation model (DEM) dataset with a 30-s resolution, are additionally used. Convectively induced turbulence (CIT) and clear-air turbulence (CAT) events comprised 11% and 89% of the total 255 moderate or greater (MOG)-level turbulence events, respectively. CAT events are classified as tropopause/jet stream–induced CAT (TJCAT) and mountain-wave-induced CAT (MWCAT) events. The MOG-level TJCAT and MWCAT events are responsible for 41.2% and 19.6% of the total MOG-level turbulence events, respectively. The CIT events in summer and the TRCAT and MWCAT events in spring occur most frequently over the previously mentioned regions of South Korea, associated with specific generation mechanisms.

Development of updateable model output statistics (UMOS) system for air temperature over South Korea

In this study, a Updateable Model Output Statistics (UMOS) system has been developed for the forecast of 3-h temperature over South Korea using two significantly different models’ (Regional Data Assimilation and Prediction System (RDAPS) and Korea Meteorological Administration (KMA) Weather Research and Forecasting (WRF) model (KWRF)) outputs based on the Canadian UMOS system (Wilson and Vallee, 2002; 2003). The UMOS system is designed to consider the local climatology and the model’s forecasting skills. The 20 most frequently selected potential predictors for each season, station, and forecast projection time from the 67 potential predictors of the Model Output Statistics (MOS) system, were used as potential predictors of the UMOS system. The UMOS equations are developed by a weighted blending of the new and old model data, with weights chosen to emphasize the new model data while including enough old model data in the development to ensure stable equations and a smooth transition to dependency on the new model. The UMOS equations were updated regularly at a predefined time interval to consider the changes of covariance structure between the new model output and observations as the new model data increase. The validation results showed that seasonal mean bias, Root Mean Square Error (RMSE), and correlation coefficients for the total forecast projection times are −0.379∼0.055°C, 1.951∼2.078°C, and 0.741∼0.965, respectively. Although, the forecasting skills of UMOS system are very consistent without regard to the season and geographic location, the performance is slightly better in autumn and winter than in spring and summer, and better in coastal regions than in inland region. When we take into account the significant differences of the RDAPS and KWRF, the UMOS system can be used as a supplementary forecasting tool of the MOS system for 3-h temperature over South Korea. However, the UMOS system is very sensitive to the selected number and/or types of predictors. Therefore, more work is needed to enable the use of the UMOS system in operation, including tuning of the number and types of potential predictors and automation of the updating processes of the UMOS equations.

Development of the Korean Mid- and Upper-Level Aviation Turbulence Guidance (KTG) System Using the Regional Unified Model

Abstract Korean mid- and upper-level aviation turbulence guidance (KTG) system isdeveloped using the unified model (UM)-based regional data assimilation and prediction system(RDAPS) of the Korea Meteorological Administration. The KTG system includes three steps.First, the KTG system calculates a suite of diagnostics in the UM-RDAPS domain. Second,component diagnostics that have different units and numerical magnitudes are normalized intothe values between 0 and 1, according to their own thresholds in the KTG system. Finally,normalized diagnostics are combined into one KTG predictor by measuring the weighting scoresbased on the probability of detection, which is calculated using the observed pilot reports(PIREPs) exclusively of moderate-or-greater (MOG) and null (NIL) intensities. To investigatethe optimal performance of the KTG system, two types (RD-KTG and UM-KTG) of the KTGsystems are developed and evaluated using the PIREPs over Korea and East Asia. Componentdiagnostics and their thresholds in the RD-KTG are founded on the 8-yrs (2002.12-2010.11)MM5-based RDAPS (previous version of the RDAPS; ∆x = 30 km) and PIREPs data, whilethose in the UM-KTG are based on the 6 months (2010.12-2011.5) UM-based RDAPS (∆x =12km)and PIREPs data. In comparison between the RD-KTG and UM-KTG, overall performance ofthe UM-KTG (0.815) is better than that of the RD-KTG (0.79) during the recent 6 months,because forecasting skill for the upper-level wind is higher in the UM-RDAPS than in the MM5-RDAPS. It is also found that the UM-KTG is more efficient than the RD-KTG according to thestatistical evaluations and sensitivity tests to the number of component diagnostics. Keywords: mid- and upper-level turbulence, PIREPs, KTG system, forecasting skill

A Numerical Study of Clear-Air Turbulence (CAT) Encounters over South Korea on 2 April 2007

Abstract On 2 April 2007, nine cases of moderate-or-greater-level clear-air turbulence (CAT) were observed from pilot reports over South Korea during the 6.5 h from 0200 to 0830 UTC. Those CAT events occurred in three different regions of South Korea: the west coast, Jeju Island, and the eastern mountain areas. The characteristics and possible mechanisms of the CAT events in the different regions are investigated using the Weather Research and Forecasting model. The simulation consists of six nested domains focused on the Korean Peninsula, with the finest horizontal grid spacing of 0.37 km. The simulated wind and temperature fields in a 30-km coarse domain are in good agreement with those of the Regional Data Assimilation and Prediction System (RDAPS) analysis data of the Korean Meteorological Administration and observed soundings of operational radiosondes over South Korea. In synoptic features, an upper-level front associated with strong meridional temperature gradients is intensified, and the jet stream passing through the central part of the Korean Peninsula exceeds 70 m s−1. Location and timing of the observed CAT events are reproduced in the finest domains of the simulated results in three different regions. Generation mechanisms of the CAT events revealed in the model results are somewhat different in the three regions. In the west coast area, the tropopause is deeply folded down to about z = 4 km because of the strengthening of an upper-level front, and the maximized vertical wind shear below the jet core produces localized turbulence. In the Jeju Island area, localized mixing and turbulence are generated on the anticyclonic shear side of the enhanced jet, where inertial instability and ageostrophic flow are intensified in the lee side of the convective system. In the eastern mountain area, large-amplitude gravity waves induced by complex terrain propagate vertically and subsequently break down over the lee side of topography, causing localized turbulence. For most of the CAT processes considered, except for the mountain-wave breaking, standard NWP resolutions of tens of kilometers are adequate to capture the CAT events.

수도권지역에서의 권역간 대기오염물질 상호영향 연구

3) , Il-Su Park Abstract This study were to simulate major criteria air pollutants and estimate regional source-receptor relationship using air quality prediction model (TAPM ; The Air Pollution Model) in the Seoul Metropolitan area. Source-receptor relationship was estimated by contribution of each region to other regions and region itself through dividing the Seoul metropolitan area into five regions. According to administrative boundary, region Ⅰ and region Ⅱ were Seoul and Incheon in order. Gyeonggi was divided into three regions by directions like southern(region Ⅲ), northern(Ⅳ) and eastern(Ⅴ) area. Gridded emissions (1km×1km) by Clean Air Pollicy Support System (CAPSS) of National Institute of Environmental Research (NIER) was prepared for TAPM simulation. The operational weather prediction system, Regional Data Assimilation and Prediction System (RDAPS) operated by the Korean Meteorology Administration (KMA) was used for the regional weather forecasting with 30km grid resolution. Modeling period was 5 continuous days for each season with non-precipitation . The results showed that region Ⅰ was the most air-polluted area and it was 3~4 times more polluted region than other regions for NO2, SO2 and PM10. Contributions of SO2 NO2 and PM10 to region Ⅰ, Ⅱ and Ⅲ were more than 50 percent for their own sources. However region Ⅳ and Ⅴ were mostly affected by sources of region Ⅰ, Ⅱ and Ⅲ. When emissions of all regions were assumed to reduce 10 and 20 percent separately, air pollution of each region was reduced linearly and the contributions of reduction scenario were similar to those of base case. As input emissions were reduced according to different ratio - region Ⅰ 40 percent, region Ⅱ and Ⅲ 20 percent, region Ⅳ and Ⅴ 10 percent, air pollutions of region Ⅰ and Ⅲ were decreased remarkably. The contributions to regionⅠ, Ⅱ, Ⅲ were also reduced for their own sources. However, region Ⅰ, Ⅱ and Ⅲ affected more regions Ⅳ and Ⅴ. Shortly, graded reduction of emission could be more effective to control air pollution in emission imbalanced area.

상호정보량 기법을 적용한 인공신경망 입력자료의 선정

Input variable selection is one of the various techniques for improving the performance of artificial neural network. In this study, mutual information is applied for input variable selection technique instead of correlation coefficient that is widely used. Among 152 variables of RDAPS (Regional Data Assimilation and Prediction System) output results, input variables for artificial neural network are chosen by computing mutual information between rainfall records and RDAPS` variables. At first the rainfall forecast variable of RDAPS result, namely APCP, is included as input variable and the other input variables are selected according to the rank of mutual information and correlation coefficient. The input variables using mutual information are usually those variables about wind velocity such as D300, U925, etc. Several statistical error estimates show that the result from mutual information is generally more accurate than those from the previous research and correlation coefficient. In addition, the artificial neural network using input variables computed by mutual information can effectively reduce the relative errors corresponding to the high rainfall events.

Multistage stochastic linear programming model for daily coordinated multi-reservoir operation

Operation planning for a coordinated multi-reservoir is a complex and challenging task due to the inherent uncertainty in inflow. In this study, we suggest the use of a new, multi-stage and scenario-based stochastic linear program with a recourse model incorporating the meteorological weather prediction information for daily, coordinated, multi-reservoir operation planning. Stages are defined as prediction lead-time spans of the weather prediction system. The multi-stage scenarios of the stochastic model are formed considering the reliability of rainfall prediction for each lead-time span. Future inflow scenarios are generated by a rainfall–runoff model based on the rainfall forecast. For short-term stage (2 days) scenarios, the regional data assimilation and prediction system (RDAPS) information is employed, and for mid-term stage (more than 2 days) scenarios, precipitation from the global data assimilation and prediction system (GDAPS) is used as an input for the rainfall–runoff model. After the 10th day (third stage), the daily historical rainfall data are used following the ensemble streamflow prediction (ESP) procedure. The model is applied to simulate the daily reservoir operation of the Nakdong River basin in Korea in a real-time operational environment. The expected benefit of the stochastic model is markedly superior to that of the deterministic model with average rainfall information. Our study results confirm the effectiveness of the stochastic model in real-time operation with meteorological forecasts and the presence of inflow uncertainty.

Parameterization of dust emission for the simulation of the yellow sand (Asian dust) event observed in March 2002 in Korea

An extraordinarily severe yellow sand event was observed in Korea for the period of 21–23 March 2002. This event has been simulated using the three‐dimensional eulerian transport model with meteorological outputs of the Regional Data Assimilation and Prediction System (RDAPS) together with the statistically derived dust emission conditions from WMO 3 hourly synoptic reporting data for seven spring seasons (March–May) from 1996 to 2002 in East Asia. The statistically derived dust‐rise conditions with the help of the Chinese soil map in the source regions are found to be the threshold wind speed of 9.5, 7.5, 6.0 and 9.2 m s−1 and the upper limit of relative humidity of 60, 35, 30% and 40% in the Gobi, Sand, Loess and Mixed soil regions, respectively. These dust‐rise conditions are implemented to the eulerian long‐range transport model to simulate the extraordinarily intensive yellow sand event observed in Korea in terms of temporal and spatial variations of dust concentrations, starting and ending times of the yellow sand event over the Korean peninsula. It is found that the model simulates quite well the observed concentrations of more than 1,000 μg m−3 within the same order of magnitude and the starting and ending times of yellow sand in Korea within an hour. The spatial distribution of vertically integrated dust concentration in the model is quite well coincided with that of aerosol index obtained by total ozone mapping spectrometer (TOMS). The simulated dust particle sizes indicate that most of mass concentrations of dust in the source regions are mainly contributed by dust particles larger than PM10 whereas those at distant regions from the sources are largely attributed to the PM10 particles, suggesting the usefulness of the particle size spectrum in identifying the origin of a dust storm. The presently derived emission conditions enable to forecast the yellow sand event in Korea more reasonably than previously used conditions of threshold friction velocities.