A study of long-term trends in mineral dust aerosol distributions in Asia using a general circulation model

Dust events have been observed in Japan with high frequency since 2000. On the other hand, the frequency of dust storms is said to have decreased in the desert regions of China since about the middle of the 1970s. This study simulates dust storms and transportation of mineral dust aerosols in the east Asia region from 1981 to 2001 using an aerosol transport model, Spectral Radiation-Transport Model for Aerosol Species (SPRINTARS), implemented in the Center for Climate System Research/National Institute for Environmental Studies atmospheric global circulation model, in order to investigate the main factors that control a dust event and its long-term variation. The model was forced to simulate a real atmospheric condition by a nudging technique using European Centre for Medium-Range Weather Forecasts reanalysis data on wind velocities, temperature, specific humidity, soil wetness, and snow depth. From a comparison between the long-term change in the dust emission and model parameters, it is found that the wind speed near the surface level had a significant influence on the dust emission, and snow is also an important factor in the early spring dust emission. The simulated results suggested that dust emissions from northeast China have a great impact on dust mass concentration in downwind regions, such as the cities of northeastern China, Korea, and Japan. When the frequency of dust events was high in Japan, a low-pressure system tended to develop over the northeast China region that caused strong winds. From 2000 to 2001 the simulated dust emission flux decreased in the Taklimakan desert and the northwestern part of China, while it increased in the Gobi desert and the northeastern part of China. Consequently, dust particles seem to be transported more from the latter region by prevailing westerlies in the springtime to downwind areas as actually observed. In spite of the similarity, however, there is still a large disagreement between observed and simulated dust frequencies and concentrations. A more realistic land surface and uplift mechanism of dust particles should be modeled to improve the model simulation. Desertification of the northeastern China region may be another reason for this disagreement.