Vegetation indices are mathematical combinations of spectral bands designed to enhance the vegetation signal while minimizing confounding factors like soil background, illumination variation, and atmospheric effects. The Normalized Difference Vegetation Index (NDVI) -- calculated as (NIR - Red)/(NIR + Red) -- is the most widely used, exploiting the contrast between chlorophyll absorption in the red band and strong scattering in the near-infrared by leaf mesophyll. NDVI ranges from -1 to +1, with dense healthy vegetation near 0.8-0.9, sparse vegetation around 0.2-0.4, bare soil near 0.1, and water typically negative. Other indices address specific limitations: EVI reduces atmospheric and soil effects, SAVI adjusts for varying soil brightness, and NDWI targets water content.
The spectral signature of vegetation -- low red reflectance from chlorophyll absorption, high NIR reflectance from leaf structure scattering -- is arguably the single most important pattern in remote sensing. Vegetation indices distill this pattern into a single number that can be mapped, tracked over time, and correlated with biophysical variables like leaf area index, biomass, fractional cover, and productivity.
NDVI's elegance lies in its simplicity and interpretability. By normalizing the NIR-Red contrast, it creates a dimensionless index that suppresses much of the illumination variability while amplifying the vegetation signal. Global NDVI time series from AVHRR (1981-present) and MODIS (2000-present) have revealed planetary-scale patterns: the seasonal green wave sweeping poleward each spring, drought-induced vegetation decline across the Sahel, and the global greening trend driven by CO2 fertilization and warming temperatures.
However, NDVI has well-documented limitations. It saturates over dense vegetation (LAI > 3-4), is sensitive to soil brightness in sparse canopies, and is affected by atmospheric aerosols. The Enhanced Vegetation Index (EVI) addresses these issues by incorporating a blue band for atmospheric correction and soil adjustment factors, maintaining sensitivity in high-biomass regions like tropical forests. The Soil-Adjusted Vegetation Index (SAVI) adds a soil brightness correction factor useful in arid environments. Water-related indices (NDWI, NDMI) replace the red band with SWIR to target canopy moisture content.
The practical value of vegetation indices extends far beyond ecological research. Precision agriculture uses NDVI maps to guide variable-rate fertilizer and irrigation application. Crop insurance programs use satellite NDVI to verify claims. Rangeland managers track forage production. Carbon cycle models assimilate NDVI as a proxy for photosynthetic activity. In each case, the vegetation index translates complex spectral data into actionable information.