Leaf Area Index And Biophysical Parameter Retrieval Of GF-1 Data Based On PROSAIL

Leaf area index is an important parameter in crop monitoring. With the change of corn policy and the launch of a series of autonomous high resolution satellites, precision farming using remote sensing will be of great necessity and possible at the same time.The PROSAIL model combines the leaf model PROSPECT and the RT model SAIL, which both have physical meaning, are invertible and have been widely used respectively. PROSAIL has been proved to be applicable to precision farming.This study uses a version of PROSAIL in which PROSPECT-5 is coupled with 4SAIL to retrieve leaf area index and leaf biochemical concentrations of GF-1 data. Affected by hail, corn growth in the experiment area is not uniform. This study found out that the reflectance of leaves in different growing conditions could be expressed by a linear model of reflectance measured from a healthy leaf and a yellow leaf. Using iterative numerical optimization, a look-up table of leaf biochemical concentrations for different mixing coefficients was established with PROSPECT. LAI was divided into several grades and directional reflectance of the canopy was calculated with SAIL for each combination of the mixing coefficient and LAI. In the end, a look-up table consisting of leaf biochemical concentrations, leaf area index and band reflectance of GF-1 was generated through the mixing coefficient. The leaf area index of GF-1 was then retrieved. Validation shows that the average precision of the inversion is 60.59% and that there is a linear relationship between inversed and measured leaf area index. After adding the panchromatic band, LAI inversion shows certain improvement. Inversed leaf biochemical concentrations are reasonable, but there are difficulties with yellow leaves.