Wetland Leaf Area Index modelling with field and satellite hyperspectral data

Leaf Area Index (LAI) is an important variable in environmental processes modelling. Already several approaches were proposed to model wetlands LAI with remote sensing data. However, none of these methods was based on upscaling the field spectral reflectance measurements, which is a matter of this paper. In this study, we used combined measurements of spectral reflectance (350-2500 nm) and LAI to establish a regression model of LAI. The spectral reflectance was resampled to the spectral resolution of a satellite hyperspectral sensor (CHRIS-PROBA) beforehand and then used to calculate NDVI-based spectral indices. From the set of spectral indices the one with the strongest correlation with LAI was chosen for the regression. Finally, the regression was applied to the CHRIS satellite images and the results were analysed within the scope of different wetland communities of the study area. The results show that the optimal regression model gives statistically different LAI values for the majority of different ground truth plant communities, rivers and urban areas.