Chen Xiang-Yue, Ding Jian-Li, Wang Jing-Zhe, Ge Xiang-Yu, Liang Jing
Key Laboratory of Smart City and Environment Modelling of Higher Education Institute, College of Resources & Environmental Science, Xinjiang University, Urumqi 830046, China.
Key Laboratory of Oasis Ecology, Xinjiang University, Urumqi 830046, China.
Huan Jing Ke Xue. 2019 Nov 8;40(11):4824-4832. doi: 10.13227/j.hjkx.201904200.
Aerosol optical depth (AOD) describes the attenuation of light by aerosols and reflects the degree of regional air pollution to some extent. This study was based on the data from the long-term sequence MOD09A1 from 2000 to 2015 and the generation of a lookup table using the deep blue algorithm (DB) to perform AOD remote sensing estimation on the Landsat TM/ETM+/OLI data from the Ebinur Lake Basin to analyze the temporal and spatial variation characteristics of AOD in the Ebinur Lake Basin and to perform an AOD prediction and factor contribution ranking using the random forest model (RF) combined with environmental variables. The results showed that:① AOD of Ebinur Lake Basin has significant seasonal variation characteristics, and the AOD values were spring (0.414) > summer (0.390) > autumn (0.287), with the largest variation in spring. ② The AOD average of the Ebinur Lake Basin was 0.374, and the interannual variation as a whole showed an upward trend. However, the AOD increased rapidly during 2010-2015, with an interannual increase of 32.32%, which indicated increasing air pollution in the basin over the past 15 years, especially the past five years. ③ The spatial distribution of AOD in the Ebinur Lake Basin was stepped up from the north to the south of Lake Ebinur. In this area, the pollution in Jinghe County was the most prominent, and the AOD value reached 0.483. ④ The RF model had a good predictive effect on AOD, =0.866, RMSE=0.042, and evapotranspiration had the most significant effect on AOD in the Ebinur Lake basin.
气溶胶光学厚度（AOD）描述了气溶胶对光的衰减程度，在一定程度上反映了区域空气污染状况。本研究基于2000年至2015年长期序列MOD09A1数据，利用深蓝算法（DB）生成查找表，对艾比湖流域的陆地卫星TM/ETM+/OLI数据进行AOD遥感估算，分析艾比湖流域AOD的时空变化特征，并结合环境变量利用随机森林模型（RF）进行AOD预测和因子贡献排序。结果表明：①艾比湖流域AOD具有显著的季节变化特征，AOD值为春季（0.414）＞夏季（0.390）＞秋季（0.287），春季变化最大。②艾比湖流域AOD平均值为0.374，年际变化总体呈上升趋势。然而，2010 - 2015年期间AOD增长迅速，年际增长32.32%，表明该流域过去15年空气污染加剧，尤其是过去五年。③艾比湖流域AOD的空间分布从艾比湖北部向南部呈阶梯状升高。该区域精河县污染最为突出，AOD值达到0.483。④RF模型对AOD具有良好的预测效果， =0.866，RMSE =0.042，且蒸散量对艾比湖流域AOD的影响最为显著。
