协同遥感数据与气象资料的高分辨率地表净辐射估算

High Resolution Surface Net Radiation (HRSNR) is of great importance in fields such as ecological environment and evapotranspiration at the ecosystem level. But restricted by actual sensor technique, it is very difficult to accurately estimate HRSNR from single satellite-based remote sensing data. In addition, most of existing methods were developed for the moderate or low resolution surface net radiation estimation without giving sufficient thought to the effect of terrain and surface heterogeneity. Thus, this project proposes a research on methods of the HRSNR estimation by combining MODIS, Landsat-8, DEM and near surface air temperature from meteorological station, and focuses on three studies: 1) the surface descending short-wave radiation estimation with considering the terrain effect; 2) the calculation of surface albedo and broadband emissivity with the consideration of the land cover information; and 3) the near surface air temperature retrieval restricted by non-stationary spatial relationship. The accuracy of HRSNR estimation could be improved by taking full advantage of the multi-source data. This project can improve the theory and method of HRSNR estimation from satellite-based remote sensing data, and also provide data and information for building “eco-city” and developing “wisdom agriculture”.

高分辨率地表净辐射对生态系统尺度上的生态环境和地表蒸散等研究有极其重要的意义。由于当前传感器技术的限制，单一星载遥感数据难以精确估算高分辨率地表净辐射；现有的研究方法大多集中于中低分辨率地表净辐射估算，未充分考虑地形效应和地表高异质性的影响。为此，本课题开展协同MODIS、Landsat-8、DEM与站点气温数据的高分辨率地表净辐射估算研究，重点研究：1）顾及地形效应的地表下行短波辐射估算，2）考虑地表类型信息的地表反照率与宽波段比辐射率估计，3）空间非稳定关系约束的近地表气温反演；充分发挥多源数据的优势，提升高分辨率地表净辐射估算的精度。本课题成果将完善高分辨率地表净辐射遥感估算的理论和方法，为“生态城市”建设和“智慧农业”发展提供重要的数据和信息支持。

地表净辐射是地表通过短波、长波辐射过程积累的净辐射能量，控制着光合作用和地表蒸散，在地气系统相互作用中起着关键作用。高分辨率地表净辐射对城市环境保护和精细农业应用有着重要意义。围绕着高分辨率地表净辐射估算中复杂地形区下行短波辐射的大气和地形效应协同校正和近地表气温与环境变量之间的空间非稳定关系表达两个关键科学问题，本课题在实施过程中，发展了实用单通道算法、考虑地形效应的地表下行辐射估算方法、考虑地表覆盖类型的地表反照率和宽波段发射率估算方法和基于地理加权模型的近地表气温估算方法等，实现了高分辨率地表净辐射估计；经过模拟数据和实测数据的验证，地表净辐射及中间变量的精度得到较明显的提升。完成了课题申请书中的全部内容。项目研究成果在国内外期刊公开发表SCI论文6篇。相关成果可为“生态城市”建设和“智慧农业”发展提供重要的数据和信息支持。