最近 濟州地域에서 觀測된 極甚한 黃砂 事例 特性分析 및 數値模擬

Abstract

The purpose of this study is to analyze the generation and transport process of Asian dust and the cause of high concentration PM10 appeared in Jeju Island with the use of the three-dimensional numerical modeling (WRF and CMAQ) in relation to the episode day (March 20, 2010) of the most severe Asian dust in the Korean peninsula in recent years. In regard to the characteristics of research episode day, PM10 concentration started increasing rapidly from 7 PM on March 20 (increased by approximately 700 to 2,000 ㎍ /m3). The maximum concentration (approximately 2,500 to 3,000 ㎍/m3) was reached around 8 and 9 PM. The concentration then started falling gradually. According to the weather chart and satellite image, the sand dust appeared in the desert located at the boundary between Tuva Republic and Mongolia and Gobi Desert around noon on March 18 and noon on March 19, respectively. However, the weather chart showed that the sand dust, which had appeared around noon on March 18, soon disappeared, whereas the satellite image continuously showed the signal of Asian dust. This difference implies that the sand dust generated from Gobi Desert around noon on March 19 was not the only cause of Asian dust on the episode day. That is to say, the sand dust, which appeared at noon on March 18, might influence Korea directly or indirectly. In addition, it generally takes approximately 2 or 3 days for Asian dust to arrive to Korea from Gobi Desert. However, it took only about 1 to 2 days in the case of research episode day. This indicates that Asian dust enters Korea at a very fast pace. The numerical simulation on Asian dust was conducted through CMAQ modeling. The emission quantity of Asian dust was calculated in consideration of the critical friction velocity, vegetation indices and attenuation factors. In regard to the emission quantity of Asian dust, the emission appeared at the time and location that were similar to that of the weather chart and satellite image. The emission quantity of PM10, which appeared in the Chinese industrial complex, was approximately 10 times lower than Asian dust. Therefore, it is assumed that only Asian dust made a significant impact on the high concentration of observed PM10. As a result of the analysis on the spatial distribution and wind field of Asian dust, high concentration Asian dust appeared in the desert located at the boundary between Tuva Republic and Mongolia around 6 PM on March 18. It then moved southeast along the wind field. Since then, Asian dust was transported up to the vicinity of Gobi Desert around noon on March 19. In the picture taken at 6 PM, the concentration of Asian dust transported to Gobi Desert became high again. Similarly to the result of the satellite image analysis, the result of the picture taken at 6 PM implies the possibility that the concentration of Asian dust was again increased because the sand dust of Gobi Desert floated due to the strong wind while Asian dust generated on March 18 was being transported. It then entered Korea around 8 PM on March 20 along the wind. Its concentration was approximately 600 to 3,000 ㎍/m3. That is, its concentration was almost identical to the highest observed concentration of PM10 in Jeju island. It then exited gradually toward Japan. In regard to the horizontal field in Jeju Island, the time when the PM10 concentration increased and the increased concentration of PM10, which were derived from the observed values of Jeju island, were almost identical. Also, the central area of Asian dust was divided when it passed by Jeju island. This is probably because Halla Mountain, which is located at the center of Jeju island, blocked Asian dust. Similarly to the horizontal field for each time slot, the vertical cross-section analysis shows in detail how Asian dust floats and sinks while entering Jeju island. Overall, Asian dust is transported at an altitude of approximately 5 to 8 km, which is the upper part of the troposphere. Some portion of Asian dust is transported at an altitude of higher than 8 km. This fact implies that a very strong cyclone might contribute to the transport of Asian dust. As a result of the comparative analysis on the results of CMAQ model and the observed values, there was almost no difference in the time when PM10 concentration increased. Overall, the observed values tended to be smaller than the results of CMAQ model. For the quantitative analysis, the statistical analysis was conducted in this study. In the case of RMSE, it reached even 190 to 500 ㎍/m3. In contrast, the model showed a low value of PM10 because MB represents a negative value at all the places. IOA shows a high value of approximately 0.7 to 0.9 in all the places with the exception of Gwangju. Therefore, it is concluded that the results of CMAQ model are at reasonable levels for reproducing Asian dust. The Asian dust analysis based on the above-mentioned numerical simulation complements the limitations of synoptic meteorology, such as weather chart. In particular, this analysis enables us to better understand Asian dust in terms of how it is generated and transported from its origin, what altitude it travels at and why its concentration increases. In general, a majority of Asian dust, which come to Korea, are originated Gobi Desert, Inner Mongolia and yellow earth plateau. Hence, it is necessary to consider not only the effects of the aforementioned regions but also the effects resulting from many unknown deserts located in the region where the windward of westerlies. It is required to analyze the generation process of Asian dust and the causes of high concentration PM10 more accurately through various numerical simulation analyses on those episode days of high concentration Asian dust, which is different from Korea.