基于GOCI数据的霾天气气溶胶辐射强迫的日内变化——以长江三角洲为例

Diurnal variations of aerosol radiative effect under haze weather condition using GOCI data——A case study of Yangtze River Delta

以长江三角洲(以下简称长三角)为研究区,基于2017年10～12月内共16d 94景的晴空静止卫星GOCI L1B图像和6S模型,采用深蓝算法反演长三角地区气溶胶光学厚度(AOD),并利用2个实测站数据进行验证.再通过反演得到的AOD结果进一步模拟计算该区域地表和大气层顶的气溶胶直接辐射效应(ADRE),并结合典型的雾霾天气进行分析.结果表明:利用GOCI数据反演的AOD具有较高拟合精度,北辰楼站点拟合相关性R^2为0.68,太湖站点为0.67.空间上,由于气溶胶存在制冷效应,AOD和地表面、大气层顶气溶胶直接辐射强度均呈现出显著的线性关系.时间上,由于气溶胶扩散和风向等因素,早上和下午辐射效应强度较高,中午较低.其中10:00和11:00影像可以较好地模拟日均ADRE的特征,利用ADRE日内变化成功捕捉到一次雾霾爆发并消失的现象,对气象部门监测近地表气温和分析灰霾的形成等具有重要意义.

Based on the Second Simulation of the Satellite Signal in the Solar Spectrum(6S)model and the Geostationary Ocean Color Imager(GOCI)L1B data of the clear-sky geostationary satellites with a total of 94 images observed in 16 days selected from October to December in 2017,the Aerosol Optical Depth(AOD)in the Yangtze River Delta was inversed with the deep blue algorithm,which was validated by in-suit observation data at two stations.The Aerosol Direct Radiation Effect(ADRE)on the surface and the top layer of the atmosphere were further calculated and analyzed under typical haze weather.The results showed that the AOD retrieved from GOCI data had high fitting precision.In addition,the fitting correlation R2 at Beichen building site and Taihu site were 0.68 and 0.67,respectively.Due to the refrigeration effect of aerosols,the AOD had a significant linear relationship with the direct radiant intensity of aerosols on the surface and at the top of the atmosphere spatially.Due to aerosol diffusion and wind direction,the intensity of radiation effects was high in the morning and afternoon and low at noon temporally.The images observed at 10:00 LT and 11:00 LT could better simulate the characteristics of the daily average ADRE.The process of an outburst and disappearance of the haze was successfully captured by using the ADRE diurnal variations,which was greatly significant to measure the near-surface temperature and analyze the formation of haze for the meteorological agency.