NDVI Bands Explained: Red, NIR & Sentinel-2 Wavelengths
The two bands behind every NDVI map
NDVI is built from exactly two bands of light: red and near-infrared (NIR). That is it. The entire “crop health” number the world relies on comes from measuring how much of these two wavelengths your field reflects, then turning the gap between them into a single value.
NDVI = (NIR − Red) / (NIR + Red)| Band | Wavelength | Sentinel-2 band | What it does |
|---|---|---|---|
| Red | ~0.665 µm (665 nm) | Band 4 | Absorbed strongly by chlorophyll |
| Near-Infrared (NIR) | ~0.842 µm (842 nm) | Band 8 | Strongly reflected by healthy leaf structure |
Both Sentinel-2 bands 4 and 8 are captured at 10 m resolution, which is why NDVI maps show fine field detail.
Why red and NIR?
It comes down to how leaves interact with light:
- Chlorophyll absorbs red light to power photosynthesis. A healthy, leafy canopy soaks up almost all incoming red — so it reflects very little. Red reflectance stays low.
- Leaf cell structure scatters near-infrared light. A healthy leaf’s spongy mesophyll acts like a mirror to NIR, reflecting it strongly back to the satellite. NIR reflectance stays high.
A healthy canopy therefore shows a big gap between low red and high NIR. When vegetation is stressed, thin, or dead, chlorophyll drops (red reflectance rises) and leaf structure degrades (NIR reflectance falls) — the gap shrinks. NDVI measures the size of that gap.
Bare soil, by contrast, reflects red and NIR roughly equally — so its NDVI sits near zero. Water absorbs both and returns negative NDVI. This is the physics that lets a simple ratio distinguish field, soil, and water.
The NDVI formula, step by step
The ratio is normalized, meaning it divides by the sum of the two bands:
- (NIR − Red) — the reflectance gap. Big for healthy vegetation, small for soil, negative for water.
- (NIR + Red) — total brightness. Dividing by this cancels out raw illumination, so NDVI means the same thing whether the scene was captured on a bright noon or a hazy afternoon.
The result always falls between −1 and +1. Because the formula is a pure ratio, it self-corrects for brightness — that is the main reason NDVI is comparable across dates and locations.
Sentinel-2 vs Landsat bands
NDVI can be computed from any satellite that has a red and a NIR band. The most common sources for agriculture:
| Satellite | Red band | NIR band | Resolution | Revisit |
|---|---|---|---|---|
| Sentinel-2 | Band 4 (665 nm) | Band 8 (842 nm) | 10 m | 5 days |
| Landsat 8/9 | Band 4 (654 nm) | Band 5 (865 nm) | 30 m | 16 days |
Sentinel-2 is the workhorse for crop monitoring — finer resolution and more frequent revisit. The slight wavelength differences between satellites mean NDVI values are close but not perfectly identical across sensors; compare like-with-like.
What the other bands do
The red-edge, SWIR, and blue bands power the other indexes that complement NDVI:
- Red-edge (Band 5, ~705 nm) → NDRE. Penetrates the leaf to read chlorophyll and nitrogen.
- SWIR1 (Band 11, ~1.6 µm) → NDMI. Absorbed by leaf water, so it tracks moisture.
- Blue (Band 2) → EVI. Corrects NDVI for atmospheric haze.
- Shortwave bands → NBR. Burn severity mapping.
NDVI’s two bands answer “is it green and dense?” The other bands answer the questions NDVI cannot — nitrogen, water, burn. Together they cover the field.
Try it free: Get NDVI (and the bands behind it) for your field — free, no signup. Works on any land worldwide. Check my field →
Frequently asked questions
What wavelengths does NDVI use?
NDVI uses two wavelengths: red light at roughly 665 nm and near-infrared at roughly 842 nm. On Sentinel-2 these are Band 4 and Band 8. The index measures the gap between how much red the canopy absorbs and how much NIR it reflects.
Why does NDVI use near-infrared and not just visible light?
Because near-infrared is where healthy leaves reveal themselves. Chlorophyll absorbs red (visible), but a leaf’s internal cell structure strongly scatters near-infrared — and that NIR reflectance is invisible to the eye. A field can look uniformly green to a person but show clear NIR variation to a satellite, which is why NDVI detects stress the eye misses.
What is the NDVI formula and how is it normalized?
NDVI = (NIR − Red) / (NIR + Red). The numerator is the reflectance gap; the denominator is total brightness. Dividing by the sum normalizes the result, cancelling out scene brightness so the value is comparable across different images, dates, and lighting conditions. The output is always between −1 and +1.
Which Sentinel-2 bands does NDVI use and at what resolution?
NDVI uses Sentinel-2 Band 4 (red, 665 nm) and Band 8 (near-infrared, 842 nm), both captured at 10 m resolution. This is why NDVI maps show fine field-level detail. Some other indexes like NDMI use Band 11 (SWIR), which is only 20 m resolution.