FY-3D MERSI-II地表温度遥感反演与验证

地表温度是是决定地表辐射能量收支的重要变量，在岩石圈、水圈、生物圈和大气圈的能量平衡和水量平衡研究中起着重要作用。利用热红外遥感技术可实现区域和全球尺度地表温度的快速获取，其受到了研究者的广泛关注。目前，FY-3D是国内光谱分辨率最高的对地观测卫星，极大的提高了对地观测能力，其搭载的中分辨率光谱成像仪（MERSI-II）经过大幅升级改进，性能有了显著提升，热红外数据的空间分辨率达到了250 m。本文使用大气辐射传输模型MODTRAN 5模拟了MERSI-II传感器热红外通道星上观测数据。在此基础上，构建了通用劈窗地表温度反演模型，结合ASTER GED全球地表发射率产品以及MERSI-II自身大气水汽反演算法，发展了地表温度遥感反演方法。最后，利用2019-08内蒙古乌海沙漠地区及美国SURFRAD多个站点的实测地表温度数据对本文提出的方法进行了验证。研究结果表明，相较地表实测数据，构建的劈窗算法反演的地表温度RMSE在1.6—2.6 K，反演精度达到了预期目标，还具有较高的空间分辨率，可以用于业务化的地表温度的反演，同时也说明其辐射定标精度有了一定保证，有效满足了区域和全球尺度地表温度遥感监测应用需求。

Retrieval and validation of the land surface temperature from FY-3D MERSI-II

Surface temperature is an important variable that determines surface radiation energy budget, and plays an important role in the study of energy balance and water balance in the lithosphere, hydrosphere, biosphere and atmosphere. Thermal infrared remote sensing technology can be used to achieve rapid acquisition of regional and global land surface temperature, which has been widely concerned by researchers. At present, FY-3D is the earth observation satellite with the highest spectral resolution in China, which has greatly improved the earth observation capability. The Moderate Resolution Spectral Imager (MERSI-II) carried by FY-3D has been greatly upgraded and improved, and its performance has been significantly improved. The spatial resolution of thermal infrared data has reached 250 m. In this paper, the atmospheric radiation transfer model MODTRAN 5 is used to simulate the MERSI-II thermal infrared channel satellite observations. On this basis, construct the land surface temperature inversion model of the generalized split-window. Combined with ASTER GED global surface emissivity product and MERSI-II atmospheric water vapor content inversion algorithm, the technical process system of remote sensing inversion of land surface temperature is developed. Finally, the proposed method was verified by using the measured surface temperature data from several stations in the Wuhai of Inner Mongolia and SURFRAD stations in the United States in August 2019. The results show that compared with the measured surface temperature data, the RMSE of the land surface temperature inversion based on the generalized split-window algorithm is between 1.6 K and 2.6 K for all sites. The accuracy of inversion has reached the expected target, and it has a high spatial resolution, which can be used in the inversion of operational surface temperature. At the same time, it also shows that the accuracy of radiometric calibration is guaranteed to a certain extent, which effectively meets the application requirements of regional and global surface temperature remote sensing monitoring.