akar.earthRemote sensing: Definition
What is remote sensing in forest carbon? How satellites, aircraft and drones measure forest cover and biomass, and where field data is still needed.
Remote sensing is the collection of information about the Earth’s surface from a distance, usually by satellites, aircraft or drones, without touching the ground. In forest carbon it is used to map land cover, detect deforestation and estimate tree height and biomass across large areas.
Why it matters
Field plots are accurate but cover a tiny fraction of a project. Remote sensing fills the gaps: it shows what happened everywhere, every few days, back to past decades. That makes it essential for baselines, leakage belts, deforestation monitoring and EUDR cut-off date checks. Verra’s VM0047 area-based approach builds its dynamic performance benchmark from remotely sensed project and control plots.
How it works
- Passive optical sensors such as Sentinel-2 and Landsat record reflected sunlight in visible and infrared bands. Indices like NDVI track vegetation.
- Active radar (SAR) sends microwave pulses and works through cloud and at night.
- Lidar, from aircraft, drones or the GEDI instrument in orbit, uses laser pulses to measure canopy height and structure.
Signals are converted to carbon using models trained on ground plots, and the error from that step must be reported.
Indonesia and Southeast Asia angle
Persistent cloud over Sumatra, Kalimantan and Papua limits optical imagery, especially in the wet season. Teams compensate by compositing many dates, adding radar and planning field work to match image dates. Small smallholder plots and mixed agroforestry are also hard to classify at 10–30 m resolution, so higher-resolution imagery or drones may be needed for plot-level work.
Resolution is the other trade-off. Free missions give 10–30 m pixels; commercial satellites and drones reach below one metre but cost more and cover less ground. Most projects mix free data for wide monitoring with high-resolution imagery for checks on sample sites.
Common mistakes
- Treating satellite biomass maps as ground truth without local calibration.
- Using global forest maps for legal or EUDR decisions without checking their definitions.
- Ignoring geolocation error when overlaying small plots on coarse pixels.
Learn more in remote sensing for biomass and field plots vs remote sensing.
Frequently asked questions
Can remote sensing measure carbon directly?
No. Sensors measure reflected light, radar backscatter or laser returns. Carbon is estimated from those signals with models calibrated on field plots.
What is the best remote sensing data for Indonesia?
There is no single best source. Optical data such as Sentinel-2 is free and detailed but blocked by cloud; radar sees through cloud; lidar gives canopy height. Most projects combine them.