Retrieval Of Leaf Area Index By Using Remote Sensing Data MISR

Leaf area index (LAI) is one of the most important parameters of land surface process. While taking the Multi-angle data MISR along with MODIS data of land cover types as data base, this paper estimated LAI of xiaoxing’anling forest by5-Scale geometric optical model combined with statistical model. The results show that at the study site the simple ratio vegetation index(SR) has the highest relationship with measured LAI, which could be used as the best predictor of LAI.It provides an effective means of estimating LAI by remote sensing for large areas. Results says:1The errors are within acceptable ranges with that the accuracy of inversion is75.7%,and the mean-square deviation is0.34.A high accuracy and desirable result were obtained.The LAI at study site decreased with increasing the elevation with a mean value of LAI is1.21,the maximum value is9.28, and the minimum value is0.83.2Compare results before and after atmospheric correction ratio vegetation index SR inversion LAI SR is very sensitive to atmospheric effects before, during inversion, remote sensing images should strictly atmospheric correction processing to ensure that the inversion results accuracy.3In the study, the simulation results for different regional differences, should be based on the measured plot data effectively adjust the parameters of the model runs, in the application of5-Scale geometrical optics model to simulate vegetation canopy reflectance annoying step settings for the model, you should first determined the step threshold angle of the image, and then adjust each time you run the step value.4For the forest vegetation growth direction is complicated, and subject to a variety of external light conditions, so the growth direction has a certain randomness, so whichever average of9angles as the study area LAI final inversion results with the actual situation more in line.5This study due to the presence of LAI plots measured distribution is not uniform and did not present a normal distribution and a series of other factors, making the results of the study by a certain degree of adverse effects.