Retrieval Of The Vertical Distribution Of Corn Canopy Leaf Area Index Based On The Multilayer PROSAIL Model

The vertical distribution of the leaf area of corn is uneven. Especially, when corn in grain-filling and the milking stage, the main functional leaves of the organic matter in the grain storage are the leaves above the ear and there is a close relationship between the size of these leaves and the yield. Most of the data sources used in the inversion of the leaf area index(LAI) are from the top of the canopy, and these data sources only contain the information of the top of the canopy.The result is that the leaf area index of the whole canopy and it is very difficult to get the LAI of a certain layer. In order to solve this question, we have established a model which can be calculated vertical distribution of LAI of corn, the multilayer PROSAIL model. This model is established based on the PROSAIL treated as a constituent of discrete multiple layers above a horizontal bottom soil surface and all layers are characterized by the following set of biological parameters and physico-chemical parameters. The idea comes from the conservation of energy, which means the sum of the reflected energy of the discrete multiple horizontal layers is equal to that of the whole canopy. In the multilayer PROSAIL model, PROSPECT is used to calculate the reflectance and transmittance of the leaves and SAIL is used for the calculation of the reflectance of each layer. In particular, this new model combines the terrestrial laser scanning(TLS) data to calculate the leaf angle distribution to optimize the Campbell ellipsoid distribution model of the PROSAIL model, and it improves the accuracy of the inversion results of the new model. Then we use TLS data calculate the layered LAI and corrected by measured canopy LAI values as the validation data of the multilayer PROSAIL model inversion results.This study takes three urban districts of Hebei province of China as an example, and Landsat 8 image as data source. We inverse the canopy LAI of the region by the multilayer PROSAIL model and the PROSAIL model. We take the two kinds results compare and analysis with the measured canopy LAI values and evaluate from four factors, the coefficient of determination(R2), the root mean square error(RMSE), the bias and the estimation accuracy(EA). This result of the multilayer PROSAIL(R2=0.69, RMSE=0.46, BIAS=0.04, EA=90.0%) has better agreement with the measured data than that of PROSAIL model(R2=0.63, RMSE=0.70, BIAS=0.41, EA=84.8%). Thus, the result of the multilayer PROSAIL model is better than that of PROSAIL model. Finally, we inverse the LAI above the ear(the sum of the middle and upper layer), and the result is R2=0.78, RMSE=0.21, and it shows that the credibility of the multilayer canopy PROSAIL model is relatively high.