Evaluation and Validation of The MODIS LAI Algorithm with Digital Hemispherical Photography at Bhitar Kanika Mangrove Forest, India

Estimation of LAI is really a challenge and researchers are developing new methods to get more accurate estimation of LAI. MODIS continuously provides globally temporal LAI product and it is easily accessible for application and development purpose. But main problem is the spatial resolution (500m) where the main target features has affected due to the presence of unwanted features like clouds and others. It was concluded by previous researchers that MODIS MOD15A2H products give the overestimation or underestimation of LAI. From the Choudhury's method [1] of view, using the available MODIS products like vegetation indices, view zenith angle, sun azimuthal angle information LAI estimation can be improved. In this study an algorithm was developed on the basis of this Choudhury method to extract the improved LAI estimation by indirectly retrieving of canopy leaf angle distribution function, extinction coefficient from the MODIS MODIS13Q1 product. This modified Choudhury method (ModcModLAIE) applied in the Bhitar Kanika mangrove forest of Odisha, India for estimating the LAI. It was proved and validated with the help of Digital Hemispherical Photography (DHP) that from the available MODIS products itself it is possible to extract finer resolution (250m) improved LAI product. DHP analysis works on the basic principle of Beer's law, gap fraction calculation. At the time of comparison, it has shown that DHP extracted LAI has the higher correlation with ModcModLAIE output than the MOD15A2H LAI product. So, this research has proved the overestimation and underestimation fact of MODIS LAI product as well as it has shown a way to extract an improved LAI estimation from available course resolution MODIS products.