Determining vegetation indices from solar and photosynthetically active radiation fluxes

The objective of this study was to quantify the seasonal variability of vegetation spectral indices to deduce leaf area index (LAI) for use in soil-vegetation-atmosphere exchange models using near-real-time and archived flux tower radiation data. The 30-min data from 11 flux tower locations in 5 vegetation types (desert grassland, temperate grasslands, crops, deciduous forests, and pine forest) were collected across the United States. Vegetation indices were derived using solar radiation and photosynthetically active radiation (PAR) measured above the vegetation canopy throughout the year. The normalized-difference vegetation index (NDVI) estimated using the 30-min data was then used to quantify the LAI of the vegetation types at the various sites. The exponential function between LAI and NDVI indicated a non-linear relationship with the maximum tower-derived NDVI/LAI about 0.82/4.5 for corn, 0.85/6 for soybean, 0.6/2-0.8/4 for grasslands, and 0.81/7 for forest. Each vegetation type and environment exhibited unique seasonal and annual signatures of NDVI/LAI. The NDVI/LAI from the flux towers compared well with the Moderate Resolution Imaging Spectroradiometer (MODIS) data derived at 1-km resolution and derived LAI showed excellent agreement with measurements in corn/soybean crops. These results encourage the use of real-time single point measurements of vegetation spectral indices in characterizing vegetation for routine plant-environment models. (c) 2007 Elsevier B.V. All rights reserved.