Estimating the leaf area index (LAI) of black wattle from Landsat ETM+ satellite imagery: scientific letter

Remote sensing techniques have the potential to provide resource managers with a rapid and economical method of acquiring information related to forest productivity and water use. This study evaluated the utility of Landsat ETM+ satellite imagery to predict canopy attributes of Black Wattle (Acacia mearnsii). The study encompassed ground-based measurements of leaf area index (LAI) and plant area index (PAI) using destructive sampling and LI-COR LAI-2000 plant canopy analyzer, respectively. Vegetation indices (VIs) were estimated from Landsat ETM+ images covering four study sites of pure stands of A. mearnsii located in the KwaZulu-Natal Midlands. The indices included: normalized difference vegetation index (NDVI), ratio vegetation index (RVI), transformed vegetation index (TVI) and vegetation index 3 (VI3). Relationships between the various vegetation indices, SLA, actual LAI and PAI values were tested using correlation and regression analyses. Results showed strong correlations between LAI and PAI (to calculate LAI), LAI and NDVI, and between PAI and NDVI. No significant correlations were found between VI3 and either PAI or actual LAI. Regression analysis revealed that actual LAI had significant relationships with PAI and NDVI. The results indicate the potential of the Landsat ETM+ satellite imageries to estimate values of important canopy attributes of A. mearnsii that are related to stand productivity that may be used as inputs into process-based models such as 3-PGS which attempt to estimate stand productivity and water use of commercial plantation tree species.Key Words: Remote sensing, Acacia mearnsii, Leaf area index, LAI, 3-PG, 3-PGS, Process-based modelsSouthern African Forestry Journal Vol.201, 2004: 3-12