Adaptation of SAVI to estimate the leaf area index considering different land covers in a Brazilian atlantic forest area

Abstract

This study discusses the relationship between the spectral behavior of vegetation and vegetation indexes, using Landsat satellite images to estimate the Leaf Area Index (LAI) in a protected area in the state of São Paulo, Brazil. We calculated the Normalized Difference Vegetation Index (NDVI) for different types of land cover and found specific correction coefficients (L) from the Soil Adjusted Vegetation Index (SAVI) to mitigate LAI saturation. The areas without vegetation presented mostly negative NDVI values ​​or values ​​lower than 0.5. Therefore, we assume that these areas should correspond to a correction coefficient of L equal to 0. Based on the analysis of each year, we assume that the areas of low vegetation cover should correspond to a correction coefficient of L equal to 1. The positive correlation between SAVI and LAI indicates that SAVI can be used to estimate LAI in the study area following the proposed methodology. Therefore, by exploring the possibility of using the adjustment factor L regarding the zonal statistics of NDVI to mitigate the effects of saturation, it is possible to adopt a local model for calculating LAI. The study highlights the importance of adapting vegetation indexes to local characteristics and the need to establish specific algorithms for calculating LAI in different regions.