Leaf Area Index from Landsat-8: review and comparison of existing algorithms applied to mixed agricultural and forest areas.

The Leaf Area Index is an important vegetation biophysical parameter, defined as a ratio of leaf area to unit ground surface area (Watson, 1947). This index is related to several vegetation exchange processes, providing information on changes in productivity or climate impacts on ecosystems. In literature it is possible to find many algorithms to its retrieval (Viña, 2011; Ganguly, 2012). The choice of the model to use become thus crucial for any kind of application. The following research aims to compare different models of LAI, undergoing the following steps: first, through the USGS archive we selected a sample of images acquired by Landsat-8 satellite, from 2013 to 2016, with 30m of spatial resolution. Subsequently, we carried out a classification of soil based on different uses, which led to the identification of five land use classes. Then, images were preprocessed through Envi and Matlab softwares, in order to isolate a particular sub-region and apply correction of cloudiness and radiometric calibration. Therefore data processing consisted of vegetation indices calculation: NDVI (Normalized Difference Vegetation Index) (Rouse et Al., 1974), WDVI (Weighted Difference Vegetation Index) (Clevers, 1988), and EVI (Enhanced Vegetation Index) (Liu and Huete, 1995). Then, LAI algorithms were chosen and applied. Finally, multi-temporal statistical analysis was carried out to evaluate the most performing models for every land cover category, according to existing experimental data.