Remote sensing has become increasingly important for agriculture because it has the potential to provide information about agricultural crops quantitatively, instantaneously, and non-destructively. The measurements and modeling of the canopy reflectance can help gain a better understanding and lead to a better interpretation of remotely sensed data. Studies were conducted on subjects related to canopy reflectance, including leaf orientation, canopy bidirectional reflectance and modeling, and the assessment on the background contribution to the canopy reflectance. Direct measurements of the leaf angle distributions of 14 types of plant canopies were taken using a Spatial Coordinate Apparatus. Four categories of zenith angle distributions were found. These were planophile, plagiophile, erectophile, and uniform. Some canopies were found to have non-uniform leaf azimuth angle distributions and there were differences between the upper and lower parts of the canopies for some species. Field measurements were conducted to examine the reflectance characteristics of selected vegetation canopies as affected by the view and solar angles. Strong relationships were observed between the reflectance factor and the view and solar angles for each canopy although differences were found among canopies. A canopy reflectance model based on radiative transport was examined to assess the effects of the major model input parameters on the canopy reflectance. Comparisons were made between measured and the model generated canopy reflectance. The model indicated that the leaf normal distribution can have large affect on the canopy reflectance in the near-infrared spectral region, as well as the red spectral region when leaf area index is small. Large differences in reflectance were observed by using different leaf scattering functions. The bi-Lambertian leaf scattering model showed the best agreement with the measured soybean canopy reflectance data. The second derivatives of canopy reflectance were examined and used a spectral indices to eliminate soil background contribution to measured reflectance. Two "windows" of second derivatives, {dollar}rm rhosp{lcub}primeprime{rcub}sb{lcub}0.69mu m{rcub}{dollar} and {dollar}rm rhosp{lcub}primeprime{rcub}sb{lcub}0.74mu m{rcub}{dollar}, exist which can be used to estimate plant growth parameters such as LAI. The second derivatives showed an advantage over the commonly used near-infrared:red ratio for eliminating the soil background contributions to the composite of soil/canopy spectrum.
展开▼
