资源环境科技发展态势分析平台

Mineral dust is a dominant aerosol type in northwestern China and can mix with other air pollutants during its long-range transport, playing an important role in Earth's energy budget and hydrological cycle on both regional and global scales. In this study, the optical properties of pure dust and polluted dust were compared by using ground-based lidar and sun photometer data at SACOL (Semi-Arid Climate and Environment Observatory of Lanzhou University) from August 2009 to August 2012. The total attenuated backscattering coefficient at 532nm, the volume depolarization ratio (VDR) and the color ratio (CR) derived from the L2S-SM-II dual-band polarization lidar and aerosol optical depth (AOD) and Angstrom exponent from sun photometer data were used to identify pure dust and polluted dust. The results showed that AOD at 440nm of pure dust was widely distributed within the range of 0.4-1.1 but the AOD of polluted dust was generally less than 0.6. The frequency distribution of VDR showed that the mean value of pure dust was greater than that of polluted dust and the peak values were 0.31 and 0.21, respectively. The frequency distribution of CR showed that the mean value of pure dust was larger than that of polluted dust, and it was mostly pure dust aerosol when color ratio was greater than 0.8. The VDR showed a positive correlation with CR. Using the frequency distribution of VDR, a threshold value of 0.25 was proposed to differentiate pure dust from polluted dust. The rates of misclassification were 20.7% and 18.6% for pure dust and polluted dust, respectively. This work will be helpful for the classification of aerosol types in satellite observations and numerical modeling.