GEDI Spaceborne Lidar for Forest Carbon

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

GEDI lidar biomass explained: how ~25 m footprints measure canopy structure, which data products to use, and its limits for carbon projects in Indonesia.

GEDI is a NASA spaceborne lidar that measures forest height and vertical structure in footprints about 25 m wide, and its footprint-level biomass product is one of the most useful global references for tropical forest carbon. It samples rather than maps: you get millions of scattered measurements, which are most valuable when combined with field plots and wall-to-wall imagery.

How GEDI works

GEDI (Global Ecosystem Dynamics Investigation) was installed on the International Space Station in late 2018. It fires laser pulses at the ground and records the full returned waveform: energy reflected from the top of the canopy, from leaves and branches inside it, and from the ground. The time difference between the first return and the ground gives canopy height; the shape of the waveform describes how foliage is distributed vertically.

Key characteristics:

• Footprint: about 25 m diameter.
• Sampling: along-track footprints about 60 m apart, in several parallel tracks; large gaps remain between orbits.
• Coverage: roughly 51.6°N to 51.6°S, the ISS orbit, which covers all of Indonesia and Southeast Asia.
• Clouds: lidar cannot see through cloud, so many shots over Indonesia are unusable.

The mission has had operational interruptions linked to ISS scheduling. Check current mission status and data availability on the GEDI or NASA Earthdata pages before you design a workflow that depends on new acquisitions.

GEDI data products

L4A uses models calibrated by region and plant functional type against field and airborne lidar data. In tropical Asia the calibration data is thinner than in some other regions, so treat L4A as an informative estimate, not ground truth.

Using GEDI in a carbon project

1. Download footprints for your area and buffer.
2. Filter quality. Use the quality flags, drop degraded shots, and prefer power-beam and night-time acquisitions where you have enough data. Remove shots on steep slopes, where ground detection is less reliable.
3. Check geolocation. Footprint position error can be several metres. Avoid matching GEDI shots to small plots or narrow strata edges.
4. Use as predictors or references. Common uses are training a canopy height or biomass model with Sentinel-2 and Sentinel-1 predictors; stratifying the project area by height class; or comparing your plot-based estimates with an independent source.
5. Combine with field plots. Your own plots, measured with good design and appropriate allometry, calibrate the final model.

Strengths and limits

Strengths. GEDI measures structure directly, so it does not saturate the way optical reflectance does in dense forest. It is free, global within its latitude range, and provides an independent check on canopy height models and biomass maps.

Limits.

• Sampling gaps mean small project areas may have few usable shots.
• Cloud cover in humid Indonesia removes a large share of data.
• Ground detection is harder under very dense, tall canopy and on steep terrain.
• Footprint biomass inherits the uncertainty of the allometric and calibration models behind it.
• Data is time-bound: changes after the last valid acquisition are not captured.

Planning around GEDI coverage

Before building a workflow on GEDI, check how much usable data actually falls on your site. A practical pre-check:

1. Download all footprints for the project area plus a buffer of a few kilometres.
2. Apply your quality filters and count the shots left per stratum.
3. Map them. Large unsampled patches or strata with only a handful of shots are a warning sign.
4. Decide whether GEDI will be a main predictor, a secondary check, or only a regional reference.

For small ARR or agroforestry projects of a few hundred hectares, GEDI usually plays the third role. For large REDD+ or IFM areas, it can be a central input. Either way, it complements rather than replaces airborne or drone data where fine detail is needed.

Common mistakes

• Treating L4A biomass as measured biomass and skipping field plots.
• Ignoring quality flags and using every shot.
• Matching footprints to plots without accounting for geolocation error.
• Comparing GEDI heights (RH98) directly with field “top height” without checking definitions.
• Assuming coverage: some small project areas in cloudy provinces have very few valid footprints.

What good evidence looks like

For a verifier, document the GEDI product versions, the filtering rules and the number of footprints kept versus dropped, the matching rule between footprints and plots or pixels, and how GEDI-derived predictors entered the model. Show validation against independent field plots.

GEDI is one input to remote sensing biomass estimation. For a comparison of airborne and drone options, see lidar vs photogrammetry. The wider monitoring stack is on the dMRV hub, and our carbon modelling service builds GEDI-informed models calibrated with field plots.