akar.earthLidar vs Photogrammetry for Forest Structure
Lidar vs photogrammetry for forest structure in Indonesia: canopy height, terrain under canopy, accuracy and when drones or lidar suit carbon projects.
Choose lidar when you need canopy height and structure under closed tropical canopy, including the ground elevation beneath it, for example in REDD+ forests, mature agroforests or peat swamp forest. Choose photogrammetry (structure-from-motion from drone photos) when canopy is open or young, budgets are tight, and you also want high-resolution orthophotos for counting trees or checking plantings. Both need field plots to turn structure into biomass and carbon.
Side-by-side comparison
| Attribute | Lidar | Photogrammetry (SfM) |
|---|---|---|
| How it works | Laser pulses measure distance; multiple returns per pulse | Overlapping photos reconstructed into 3D points |
| Sees under canopy | Yes, some pulses reach the ground through gaps | No, only visible surfaces |
| Terrain model in dense forest | Reliable digital terrain model (DTM) | Poor unless ground is visible or a DTM exists |
| Canopy height | Accurate canopy height model from DSM minus DTM | Top-of-canopy surface; height depends on external DTM |
| Vertical structure | Profiles of understorey and layers | Outer canopy surface only |
| Imagery | Not by itself; often paired with a camera | Produces orthophotos as a by-product |
| Sensitivity to light and wind | Works in low light; less affected by wind | Needs good light; wind blurs crowns and harms matching |
| Equipment cost | Higher (sensor plus capable drone or aircraft) | Lower (consumer or prosumer drone) |
| Processing | Point cloud classification, specialist software | SfM software; heavy compute for large areas |
| Best Indonesian uses | REDD+ biomass mapping, peat swamp forest, mature agroforest | Young ARR, nurseries, smallholder plot mapping, tree counts |
Key differences
The ground problem. Canopy height equals canopy surface minus ground elevation. Lidar measures both in one flight because some pulses pass through gaps. Photogrammetry only reconstructs what the camera sees, so under closed canopy it has no ground. In Indonesian lowland rainforest, that gap is decisive. Teams sometimes pair drone photogrammetry with an existing lidar DTM or with surveyed ground points.
Structure versus surface. Lidar returns describe vertical structure, which correlates with biomass and degradation, for example selective logging that removes mid-canopy trees but leaves the top surface largely intact. Photogrammetry captures crowns and gaps from above, which is useful for counting and crown delineation in open stands.
Field conditions. Photogrammetry struggles in wind, with uniform textures such as dense young acacia canopy, and in low sun or haze. Lidar is more robust but has its own limits: battery life, data volume and permits for flying.
Scale. Neither covers a whole province cheaply. Spaceborne GEDI lidar samples forest structure across the tropics and helps build regional models; drones and aircraft cover project-scale samples. See canopy height models.
When each fits in Indonesia and Southeast Asia
- REDD+ in lowland dipterocarp or peat swamp forest: lidar, flown on sample transects, calibrated with plots and extrapolated with satellite data.
- ARR in the first years: photogrammetry for survival counts and planting audits; mobile tree measurement for DBH and height on the ground.
- Coffee, cocoa and rubber agroforestry: photogrammetry maps shade trees and plot boundaries; also useful for EUDR polygons.
- Mangroves: lidar handles complex root and canopy structure better; photogrammetry suffers over water and mud reflections.
- Permits: check current Indonesian drone flight rules and any local permissions before campaigns.
Cost and effort considerations
We do not quote prices. Drivers:
- Hardware: lidar sensors cost substantially more than cameras; renting or contracting flights is common.
- Flight time: both are limited by battery and terrain access. Remote Papua or Kalimantan sites add logistics cost.
- Processing and storage: lidar point clouds and SfM projects are both large; plan compute and archiving.
- Calibration plots: the field plot campaign often costs as much effort as the flights. Precise plot location with RTK GNSS is essential to link plot biomass with point clouds.
Our field data campaigns combine drone flights with RTK-surveyed plots, and carbon modelling converts structure into biomass estimates.
Questions to ask before you choose
- Is the canopy closed? If you cannot see the ground from above, photogrammetry alone will not give reliable tree heights.
- What output do you actually need? Tree counts and planting audits favour photogrammetry; biomass and degradation mapping favour lidar.
- How often will you fly? Repeat surveys need consistent sensors, altitudes and seasons, so plan the whole monitoring series up front.
- Who will process the data? Both methods produce large datasets; budget for skilled analysts, not just flights.
Common misconceptions
- “Lidar measures carbon.” It measures structure. Biomass and carbon come from models calibrated with field allometry; see allometric equations.
- “Drone photogrammetry is a cheap lidar.” It is a different measurement and fails where ground is hidden.
- “One flight is enough.” Change detection needs repeated, comparable flights with consistent settings.
- “Remote sensing replaces plots.” Plots remain the reference; see field plots vs remote sensing.
More in the dMRV hub.
Frequently asked questions
Can drone photogrammetry measure tree height in dense tropical forest?
It measures the top of the canopy well, but it usually cannot see the ground under closed canopy. Without an accurate terrain model from lidar or field survey, height estimates in dense forest are unreliable.
Is lidar required by carbon methodologies?
No major forest methodology requires lidar. It is used to improve biomass estimates and reduce uncertainty, always calibrated with field plots.
What is the difference between airborne and spaceborne lidar?
Airborne or drone lidar produces dense 3D point clouds over a project area. Spaceborne lidar such as GEDI samples footprints along tracks across the tropics, useful for regional models but not wall-to-wall mapping.
Which is better for young ARR plantations?
Photogrammetry often works well where canopy is open and the ground is visible, as in young plantings and agroforestry. Lidar becomes more valuable as canopy closes.