Spatiotemporal distribution of aerosols over the Tibet Plateau and Tarim Basin (1980–2020)

Abstract

Aerosol Optical Depth (AOD) represents the optical attenuation of aerosol, which, to a certain extent, can reflect regional air pollution levels with obvious spatiotemporal characteristics. However, the currently limited monitoring technology cannot provide long-time sequential and detailed AOD information. Therefore, in this paper, Modern Era Retrospective-Analysis for Research and Applications, Version 2 (MERRA-2) data, and Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) data from 1980 to 2020 were used to analyze the characteristics of both the spatiotemporal distribution of aerosols and dust transportation in the Tibet Plateau and Tarim Basin. Moreover, Random Forest (RF) Model was employed for ranking the importance of multiple natural factors (evaporation, mean surface temperature, daily precipitation, mean air pressure, mean relative humidity, mean temperature and mean wind speed) on AOD, Wavelet Analysis Method was utilized for the scale analysis between key natural factors and AOD, and Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) Model was applied to the simulation of the transportation trajectories of dust aerosols. The results show that high AOD values in the Tibet Plateau and Tarim Basin appeared in spring. Dust aerosols in the northern slope of the Tibet Plateau dispersed and decreased from the north to the middle of Plateau. Dust Aerosol Optical Depth (DAOD) reached its peak in spring and summer with the maximum transportation range of 34° N or so. Besides, Tarim Basin transported dust to the northern slope of the Tibet Plateau significantly, which was then transported to the main body of the Plateau along its northern slope (37°N) with a dust particle proportion of 40%–80%. Moreover, precipitation was the most significant factor (32.18%) influencing the spatiotemporal distribution of AOD. The former lagged behind the latter by 5/12 periods at most and showed a negative correlation with it.