在对不同类型土壤进行室内光谱二向反射率测定的基础上,分析了可见光波段及TM4近红外波段土壤二向反射率随观测角度变化规律,得出以下结论:在不同的观测方位角,土壤二向反射率都随着观测天顶角的增加而增加,土壤的二向反射率在垂直主平面方向是对称的;反射率在后向散射方向达到最高,在前向散射方向最低。并利用基于辐射传输理论的Hapke二向反射模型对不同类型土壤在可见光波段及Land-sat TM4近红外波段的二向反射率进行了模拟,得到了很好的结果;其在模拟可见光波段时的RMSE值分别为0.003,0.002,0.004,相关系数分别达到0.995,0.998及0.998;在模拟TM4近红外波段时的RMSE值分别为0.004,0.006,0.005,相关系数分别达到0.997,0.996及0.998,这表明通过逐波段的模拟可以获取整条二向反射光谱曲线。
Knowledge of radiative transfer over bare soils is a prerequisite to addressing vegetation canopies and predicting soil properties by remote sensing. In the present study, the change in the spectral reflectance for three soils (i. e. red soil, paddy soil and coastal saline soil) with different view zenith and azimuth angles in the visible band (620 nm) and Landsat TM4 near in- frared wavebands (760-900 nm) was measured in laboratory. The results showed that soil reflectance increased with increasing off-nadir view angle for all azimuth directions and soil bidirectional reflectance was azimuthally symmetric. The reflectance was highest in baekseattering direction and lowest in forward-scattering direction. The bidireetional reflectance was simulated well using the Hapke model derived from the radiative transfer theory. The root mean square errors (RMSE) were 0. 003, 0. 002 and 0. 004 and the correlation coefficients were 0. 995, 0. 998 and 0. 998 in simulating bidirectional reflectance in visible wavebands for red soil, paddy soil and coastal saline soil, respectively; RMSE were 0.004, 0.006 and 0.005 and the correlation coefficients were 0. 997, 0. 996 and 0. 998 in simulating bidirectional reflectance in NIR wavebands for these soils, respectively. It is stated clearly that Hapke model could be used to simulate the whole spectra curve, then to retrieve the soil surface characteristics.