NDVI (Normalized Difference Vegetation Index) is a remote-sensing index that quantifies live green vegetation using the contrast between near-infrared and red reflectance. Healthy vegetation strongly reflects near-infrared light while absorbing red light for photosynthesis, so the difference between the two bands is a robust proxy for vegetation vigour and cover.
The formula
NDVI = (NIR − Red) / (NIR + Red)
Values range from −1 to +1. Dense, healthy vegetation typically scores roughly 0.6–0.9; sparse vegetation and bare soil fall around 0.1–0.3; water, snow, and clouds are near zero or negative.
A concrete example
For Sentinel-2, NIR is Band 8 (~842 nm) and Red is Band 4 (~665 nm), so NDVI = (B8 − B4) / (B8 + B4). For Landsat 8/9 OLI, NIR is Band 5 and Red is Band 4. In QGIS or the raster calculator in GDAL, you compute it directly from the surface-reflectance bands; in Google Earth Engine the normalizedDifference method does the same.
Why it matters in geology
Vegetation hides the rocks geologists want to see. NDVI is useful for the inverse: low-NDVI areas highlight bare ground, outcrop, and disturbed terrain suitable for spectral mineral mapping, and changes in NDVI can flag drought stress, contamination, or subtle lithological control on vegetation patterns.
Common pitfall
NDVI says nothing directly about rock type—it maps vegetation, not geology. It also saturates at high biomass (losing sensitivity in dense canopy) and is affected by soil background and atmosphere, so use atmospherically corrected reflectance, not raw digital numbers. Treat NDVI as a vegetation mask or screening layer, never as a lithology classifier.