Comparison of different vegetation indices for the remote assessment of green leaf area index of crops

Many algorithms have been developed for the remote estimation of biophysical characteristics of vegetation, in terms of combinations of spectral bands, derivatives of reflectance spectra, neural networks, inversion of radiative transfer models, and several multi-spectral statistical approaches. However, the most widespread type of algorithm used is the mathematical combination of visible and near-infrared reflectance bands, in the form of spectral vegetation indices. Applications of such vegetation indices have ranged from leaves to the entire globe, but in many instances, their applicability is specific to species, vegetation types or local conditions. The general objective of this study is to evaluate different vegetation indices for the remote estimation of the green leaf area index (Green LAI) of two crop types (maize and soybean) with contrasting canopy architectures and leaf structures. Among the indices tested, the chlorophyll Indices (the CIGreen, the CIRed-edge and the MERIS Terrestrial Chlorophyll Index, MTCI) exhibited strong and significant linear relationships with Green LAI and thus were sensitive across the entire range of Green LAI evaluated (i.e., 0.0 to more than 6.0 m(2)/m(2)). However, the CIRed-edge was the only index insensitive to crop type and produced the most accurate estimations of Green LAI in both crops (RMSE = 0.577 m2/m2). These results were obtained using data acquired with close range sensors (i.e., field spectroradiometers mounted 6 m above the canopy) and an aircraft-mounted hyperspectral imaging spectroradiometer (AISA). As the CIRed-edge also exhibited low sensitivity to soil background effects, it constitutes a simple, yet robust tool for the remote and synoptic estimation of Green LAI. Algorithms based on this index may not require re-parameterization when applied to crops with different canopy architectures and leaf structures, but further studies are required for assessing its applicability in other vegetation types (e.g., forests, grasslands). (C) 2011 Elsevier Inc. All rights reserved.