利用卫星遥感资料反演感热和潜热通量的研究综述

区域平均感热和潜热通量是气象、水文、生态模式中的关键物理因子，卫星遥感反演为观测区域平均感热和潜热通量提供了可能。对利用卫星遥感资料反演地气通量的方法进行了总结和评述。首先描述了现在常用的反演方法，分析了方法中的各种假定对反演结果的影响，并对不同的模式反演结果进行了比较。还指出了评价卫星反演通量的精度时需要注意的问题。最后对该领域内现存的问题与发展方向进行探讨。

Using Satellite Remotely Sensed Data to Retrieve Sensible and Latent Heat Fluxes: A Review

Regional sensible and latent heat fluxes are the key physical parameters of the meteorological, hydrological and ecological models. However it is very difficult to obtain these fluxes from conventional ground measurement. Using satellite remotely sensed data to retrieve these fluxes supplies one possible way to solve this problem. However, the gradient measurement is needed when calculating the fluxes are calculated, while the measurements from only one layer can be obtained from satellite. To solve this problem, many studies have been carried out. There are two main ways that can be seen: the physical model and empirical model. The physical models have two directions: the gradient model and the thermal inertia method. Gradient models combine the satellite remotely sensed data with the ground measurements, and use the difference of surface temperature and air temperature at reference height to calculate sensible heat flux. The latent heat flux is obtained as the residual. The thermal inertia method uses the response of soil to the absorption of solar radiation to calculate the sensible and latent heat flux. The empirical method calculates empirical regression between the measurements of the fluxes and the satellite remotely sensed data, and then extends this relationship to calculate the fluxes. Here the daily-average fluxes are often used. 
    Four gradient models are reviewed, including one one-source model and three two-source models. When use the one source model is used for to partly vegetation-covered surface, the difference of the air dynamic temperature and the thermal radioactive temperature hinds it usage. Two-source models can solve this problem. However, all the gradient models are sensitive to the error of the difference between the satellite retrieved surface temperature and the measurements of the air temperature. Another shortcoming of the gradient model is that they need to interpolate ground measurements, such as the air temperature and wind speed. These interpolations always are of low quality with unacceptable errors. The thermal inertia method calculates sensible and latent heat fluxes only using the satellite remotely sensed data, which will have a wider usage in the near future. However, up to today, this method only succeeded in the bare soil surface. More attention should be paid to it in the future. 
    At last, the methods used to evaluate the accuracy of the retrieval of sensible and latent heat fluxes are reviewed.