Ramsey & Chatteris Fens · EnMAP ·

Fen Hyperspectral

One image from the German EnMAP satellite: narrow colours from violet to shortwave infrared, where Sentinel-2 has 13. Fields are outlined in the crop Defra's crop map recorded. Click one to compare its spectrum with others.

Image

Fields

Can the colours tell crops apart?

Every field in this image with a clear view

Can the colours tell crops apart?

Spectra by crop

Line: the typical field. Shaded: the middle 80% of fields. Where two crops' shading overlaps, a field could be either. Pick up to four crops.

Crop guessed from this image alone

Most common mix-ups

Crops are made of the same few things. Every green canopy is mostly water, chlorophyll, cellulose and protein, in proportions that shift more with growth stage, variety, fertiliser and weather than between crops. Wheat and barley in June are chemically very close.
A 30 m pixel isn't a lab sample. It mixes leaves at every angle, shadow, soil between rows, weeds and field edges, seen through the atmosphere. Canopy structure drives the signal more than chemistry, and water in living leaves swamps the fine protein and cellulose features lab instruments rely on.
Timing tells crops apart better than colour. Winter wheat is green in November, spring barley is bare soil until April, oilseed rape flowers yellow in April and sugar beet closes its canopy in July. Sentinel-2 and Sentinel-1 record that calendar every few days for free; EnMAP takes images only on request, and cloud spoils most of them. This was the only usable EnMAP image over the Fens in the 2025 season.
Where hyperspectral earns its place is crop condition rather than crop type: nitrogen status, water stress, disease, cover-crop mixes. That is worth testing on fields a Sentinel-based model has already flagged.