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

[1] 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 mug 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.