Satellite Verification for EUDR Deforestation-Free Claims

Updated 2026-11-02 · 4 min read · By the akar.earth specialist team

How EUDR satellite monitoring works for Indonesian plots: imagery sources, forest maps, alert systems, cloud cover problems and when field checks are needed.

Satellite verification for the EUDR means checking each plot’s imagery from around 31 December 2020 to the present to show that no forest was cleared or, for wood, degraded after the cut-off date. It is the main evidence behind a deforestation-free claim, but it only works with accurate plot geolocation and with field checks for the plots that imagery cannot resolve.

What the check has to answer

For every plot of production, the operator needs to show:

1. Was the plot forest on 31 December 2020, under the EUDR forest definition?
2. If it was, has it been converted to agricultural use since?
3. For wood, has forest been degraded, for example natural forest converted to plantation?

A “no” to question 1 is the most common answer for established farmland, and it is the one most worth documenting well.

Data sources

See Sentinel-2 vs PlanetScope and our guide to Sentinel-2 for forest monitoring.

Maps and alert systems

Ready-made products speed up screening:

• The EU global forest cover map for 2020, produced by the Commission’s Joint Research Centre for the EU Observatory on deforestation. It is a non-binding reference. Use it as a first screen, not as a verdict.
• Global tree cover loss datasets, which show annual loss since 2000 but do not separate forest from plantations.
• Near-real-time alert systems, such as optical and radar-based deforestation alerts, which flag new disturbance. See deforestation alerts.
• Indonesian national data, including the government’s land cover maps and forest estate boundaries, which help with both deforestation and legality screening.

Each product has errors. Agroforests, rubber and oil palm are often confused with forest, and small plots fall between pixels.

A practical screening workflow

1. Validate geometry. Check every point and polygon. Bad coordinates make good analysis useless.
2. Automated screen. Overlay plots on forest maps and loss data for 2021 onwards. Apply a buffer to points.
3. Classify results. Clear (no forest in 2020, or no loss since), flagged (possible loss), or uncertain (cloud, mixed pixels).
4. Visual review. An analyst checks flagged plots on time-series and higher-resolution imagery.
5. Field check. Send teams to plots still unresolved. Record crop age, land use and photos with coordinates.
6. Decide and document. Keep the evidence for each decision. Exclude non-compliant plots from EU-bound volume.
7. Monitor. Re-screen the supply base regularly so new clearing is caught before it enters a shipment.

Why field data still matters

Satellites cannot tell the age of a cocoa tree under coconut, or whether a clearing in 2022 was a rubber replant or a new garden in forest. Ground-truth plots also calibrate the models used to estimate canopy height and biomass, which helps both deforestation screening and carbon accounting. See ground truth vs satellite.

Indonesia-specific pitfalls

• Cloud. Use multi-month composites and radar for Sumatra, Kalimantan and Papua.
• Smallholder plot size. Many plots are under 1 hectare. A 30 m pixel covers almost a tenth of a hectare.
• Peat and fire. Burn scars can look like clearing. Check fire history.
• Agroforestry. See EUDR and agroforestry for false positives.

What a plot record should contain

For each plot, keep a short record that a buyer or authority can understand:

• Plot ID and geometry.
• Screening result and the data sources used, with dates.
• Images or links showing land cover around 31 December 2020 and the latest image.
• For flagged plots: the analyst’s interpretation, field visit date, photos and conclusion.
• The final decision: included or excluded from EU-bound volume.

Records should be reproducible. If an authority asks in three years why a plot was accepted, you should be able to show the same evidence. See the due diligence statement for record keeping duties.

How we approach it

We combine RTK-surveyed plot boundaries, time-series imagery and field verification, and hand buyers a plot-by-plot record. Our AKAR dMRV platform runs the same pipeline used for carbon monitoring. See the EUDR hub for the full topic map, or message us on WhatsApp through our EUDR compliance services.