What vegetation indices tell you about tree health

Satellite tree health maps are built from vegetation indices. They are useful, but only if you know what they measure and what they do not.

NDVI in one paragraph

NDVI compares how much red and near-infrared light a canopy reflects. Healthy leaves absorb red light and reflect near-infrared, so a high NDVI generally means a dense, active canopy. A low value can mean sparse leaves, stress, bare soil or a dormant tree.

Why NDRE is often added

In dense canopies NDVI tends to flatten out and stop separating good from very good. NDRE uses a band at the red edge, which stays sensitive to differences in leaf chlorophyll at high canopy cover. Used together, the two give a better picture than either alone.

Where indices mislead

• Young orchards: bare soil and weeds between trees dilute the signal of small canopies.
• Cover crops and grass in alleys can look green when the trees are not.
• Deciduous and flowering crops change naturally through the year, so a drop may be phenology rather than stress.
• Cloud, haze and shadow can distort a single date.
• An index shows that something differs. It does not say whether the cause is water, nutrition, pests, disease or damage.

How to use them well

• Compare a block with itself over time and with similar blocks of the same crop, age and stage.
• Look at individual trees, not only block averages, so a weak patch does not hide.
• Treat a low reading as a reason to scout, then record what you find.
• Keep thresholds per crop and stage. One band for every crop gives false alarms.

That is how Plantation Monitoring approaches it: tree-level scores, thresholds per crop and stage, a visible reason for every alert, and scouting tasks that close the loop. Scores are indicative by design.