Abstract: The land surface temperature (LST) plays an important role in the course of interaction between land and atmosphere. The distribution of land surface temperature with two dimensions can only be offered by remote sensing. The retrieval of sea surface temperature (SST) has been in operational use for a long time with split windows method. Is it possible to solve LST problem by split-windows data? People argue on it. Many people hope to solve missing equations by adding middle infrared information, but new problems will be brought in. For example, how to eliminate the effect of sun direct radiation and how to solve scattering problem by atmospheric aerosol. This paper proves the feasibility of retrieval of LST with accuracy less than 1 ^oC by digital simulation approach only using split windows data. Results had been tested by real time NOAA-AVHRR data.

Key words: land surface temperature, remote sensing

摘要： 陆面温度在地气相互作用过程中扮演着十分重要的角色，唯有遥感可以提供二维陆面温度分布信息，鉴于“分裂窗口”方法求取海面温度已进入业务运行，那么能否依赖“分裂窗口”数据求解陆面温度呢?长期以来人们为此争论不休。为了使方程组封闭，许多人寄希望于中红外窗区能提供更多的信息，然而中红外的使用必然会引发新问题，其一是太阳直接辐射的影响如何消除；其二是气溶胶的散射作用不可忽略。本文在数值实验的基础上论证了应用“分裂窗口”方法实现陆面温度反演精度在1℃以内的可行性，并用实测NOAA-AVHRR资料初步验证了该方法的稳定性。

关键词: 陆面温度, 遥感