Forest Monitoring Data From Lidar and Remote Sensing

Lidar sensors and remote sensing can measure a wide range of forest attributes, including tree height, canopy cover, biomass, species composition, and structural complexity, across scales from a single stand to entire continents. You get these data by combining laser-based 3D point clouds with optical or radar imagery, then calibrating the measurements with field plots. This article explains what each technology measures, how the data are processed, and how to choose the right approach for your monitoring goal.

Formula

The calculator above uses a simplified allometric equation to estimate timber volume from lidar-derived canopy height and point density. The formula is:

Volume per hectare (m³/ha) = 0.5 × Canopy height (m) × Point density (pts/m²)

Where:

• Canopy height (m) – the mean height of the dominant tree layer, typically taken as the 90th or 95th percentile of lidar returns.
• Point density (pts/m²) – the number of laser pulses that hit a square meter of ground; higher density captures more fine branches and understory.
• 0.5 – a form factor that accounts for stem taper and branch volume; real values range from 0.3 (open-grown pines) to 0.6 (dense hardwoods).

This is a teaching approximation. Real forest inventories use species-specific equations that include diameter at breast height (DBH) and wood density, which lidar cannot measure directly.

Worked Example

Using the default values from the calculator: canopy height = 25 m, point density = 5 pts/m², area = 10 ha.

1. Volume per hectare = 0.5 × 25 × 5 = 62.5 m³/ha
2. Total volume = 62.5 × 10 = 625 m³

The calculator shows 62.5 m³/ha and 625 m³ total. In a real lidar survey, you would replace the 0.5 factor with a species-specific equation and add a correction for slope and canopy density.

What Lidar Directly Measures

Lidar (Light Detection and Ranging) sends laser pulses toward the ground and records the time each pulse takes to return. The result is a 3D point cloud with X, Y, and Z coordinates. From that cloud, you can extract:

• Canopy height – the difference between the first return (treetop) and the last return (ground). This is the most reliable lidar product.
• Canopy cover – the proportion of laser pulses that hit vegetation versus ground. This differs from canopy closure, which is measured from a single point on the ground.
• Vertical structure – the distribution of returns at different heights. A multi-layered forest shows returns at several elevations; a uniform plantation shows a narrow band.
• Digital elevation model (DEM) – the bare ground surface after removing vegetation returns. This reveals microtopography, drainage channels, and slope.
• Tree density – by detecting individual tree crowns in high-density data (usually >10 pts/m²), you can count stems and measure crown diameter.

Lidar does not measure tree diameter, species, or wood density directly. Those require field plots or additional sensors.

What Optical Remote Sensing Measures

Optical satellites (Landsat, Sentinel-2, MODIS) and aerial imagery measure reflected sunlight in visible and infrared bands. They provide:

• Spectral reflectance – different species and health conditions absorb and reflect light differently. For example, healthy vegetation reflects strongly in the near-infrared, while stressed trees show reduced reflectance.
• Vegetation indices – combinations of bands like NDVI (Normalized Difference Vegetation Index) correlate with greenness and leaf area, but saturate in dense canopies.
• Land cover classification – algorithms assign each pixel to a class (forest, grassland, water) based on spectral signatures. This is how global forest maps are made.
• Change detection – comparing images from different dates reveals deforestation, fire scars, or seasonal leaf-on/leaf-off patterns.

Optical sensors see only the top of the canopy. They cannot see understory or ground structure, and they fail in cloudy or smoky conditions.

What Radar (SAR) Measures

Synthetic Aperture Radar (SAR) sends microwave pulses and records the backscatter. Unlike optical sensors, radar penetrates clouds and works at night. It measures:

• Biomass – longer wavelengths (L-band or P-band) penetrate deeper into the canopy and correlate with woody volume. The ESA Biomass mission (launched 2024) uses P-band for this purpose.
• Structure – radar backscatter varies with branch orientation, size, and moisture content. Interferometric SAR (InSAR) can produce canopy height models, though less precisely than lidar.
• Moisture – the dielectric constant of wood changes with water content, affecting the signal. This is useful for fuel moisture estimation.

Radar is less intuitive than lidar or optical data, and its interpretation requires specialized processing. But it is the only option for tropical regions with persistent cloud cover.

How to Choose Between Sensors

Your choice depends on the question you are asking and your budget. Here is a practical guide:

• Need tree height and terrain? Use airborne lidar. It is the gold standard for these measurements.
• Need to map species or detect disease? Use optical imagery with high spectral resolution (hyperspectral if possible).
• Need biomass across a large region? Combine lidar samples (to calibrate) with radar or optical imagery (to extrapolate).
• Need to monitor monthly changes? Use free satellite data like Sentinel-2 (10 m resolution, 5-day revisit) or Landsat (30 m, 16-day).
• Need understory or ground fuel? Lidar with high point density (>10 pts/m²) can see through gaps, but very dense understory will block the ground.

In practice, most professional forest monitoring programs use a combination: lidar for structure, optical for species and health, and field plots for calibration.

Common Problems and Misconceptions

One common misconception is that lidar can directly measure biomass. It cannot. Lidar measures volume and height, which are then converted to biomass using wood density and allometric equations. The error in these equations often exceeds the sensor error.

Another issue is point density. Low-density lidar (1-2 pts/m²) is fine for terrain mapping but misses small trees and understory. High-density data costs more and requires more storage and processing time.

Optical data has a saturation problem: NDVI and similar indices stop responding in dense, high-biomass forests. This is why optical-only biomass maps underestimate tropical forests.

Finally, terrain affects lidar accuracy. On steep slopes, the ground return can be delayed, causing height errors. Most processing software includes slope correction, but it is not perfect.

Costs and Practical Considerations

Costs vary widely. Free satellite data (Landsat, Sentinel-2) is available for any region. Airborne lidar surveys typically cost between $1 and $5 per hectare for low-density coverage, and $10-$30 per hectare for high-density, depending on the contractor and terrain. Drone-based lidar is cheaper for small areas but requires a pilot license and good weather.

Processing time is another cost. A 10,000 ha lidar survey produces billions of points that require specialized software and computing power. Cloud-based platforms like Google Earth Engine can process optical data quickly, but lidar point clouds are still too large for most cloud platforms.

For most users, the practical workflow is:

1. Download free optical satellite data (e.g., Sentinel-2).
2. Create a vegetation index or land cover map.
3. If you need height or biomass, contract a small lidar survey over a subset of your area.
4. Use field plots to calibrate the lidar-derived metrics.
5. Extrapolate to the full area using the optical data.

Frequently Asked Questions

Can lidar measure tree diameter?

No. Lidar cannot see through bark to measure diameter at breast height (DBH). However, in high-density scans, it can measure crown width, which correlates with DBH. You still need field measurements to establish the relationship.

What is the difference between canopy cover and canopy closure?

Canopy cover is the proportion of ground covered by the vertical projection of tree crowns, typically measured from above (e.g., by lidar). Canopy closure is the proportion of sky blocked by vegetation when viewed from a point on the ground. They are related but not identical; lidar measures cover, while hemispherical photos measure closure.

How accurate is satellite-based forest height?

Satellite lidar (like GEDI on the ISS) provides height estimates with an error of about 2-5 meters, depending on slope and canopy density. Radar-based height models are less accurate, typically 5-10 meters. Airborne lidar is the most accurate, with errors under 0.5 meters.

Do I need to do field work if I have lidar?

Yes. Lidar gives you structural metrics, but you need field plots to convert those metrics into biomass, volume, or species composition. Field data also validates the lidar measurements. A typical design uses 20-50 plots per forest type.

Can remote sensing detect forest fires?

Yes. Thermal infrared sensors on satellites like MODIS and VIIRS detect active fires in near real-time. Optical sensors detect burn scars days to weeks after the fire. Lidar can measure fuel loads before a fire and the severity of the burn after, but it is rarely used for active fire detection due to cost and revisit time.

Related

• AI Data Observability: Anomaly Detection Limits
• How to Level Up on Steam: Fast XP Guide
• Gene Silencing Fat: What It Means in Research

in t | Canopy height (m) | m | 25 | 0 | 120
in d | Point density (pts/m²) | pts/m² | 5 | 0.1 | 50
in a | Area (ha) | ha | 10 | 0 | 10000
out vol | Estimated volume (m³/ha) | m³/ha | 0.5 * t * d | 0
out tot | Total volume (m³) | m³ | 0.5 * t * d * a | 0