Related Experiment Video
Updated: Jul 24, 2026
09:33
A Technical Perspective in Modern Tree-ring Research - How to Overcome Dendroecological and Wood Anatomical Challenges
Published on: March 5, 2015
29.2K
A novel post-fire method to estimate individual tree crown scorch height and volume using simple RPAS-derived data
Jeremy Arkin1, Nicholas C Coops1, Lori D Daniels2
1Integrated Remote Sensing Studio, Department of Forest Resources Management, University of British Columbia, Vancouver, BC V6T 1Z4 Canada.
Fire Ecology
|March 28, 2023
Summary
A new method uses drone imagery to accurately assess individual tree fire damage, estimating crown scorch height and volume. This approach provides a scalable solution for post-wildfire analysis, improving upon traditional methods.
Area of Science:
- Forestry
- Remote Sensing
- Ecology
Background:
- Wildfire impact assessment is crucial for post-fire management.
- Satellite imagery has limitations in detecting individual tree damage.
- Remotely Piloted Aerial Systems (RPAS) offer higher resolution fire severity mapping.
Purpose of the Study:
- To develop a novel methodology for analyzing individual tree fire impacts.
- To utilize synergistic data from RPAS-acquired orthomosaics and Digital Aerial Photogrammetric (DAP) point clouds.
- To enable scalable estimation of crown scorch heights and volumes.
Main Methods:
- Classifying trees as burned or unburned using RPAS orthomosaics.
- Extracting crown scorch heights and volumes from DAP point clouds of fire-damaged trees.
- Comparing RPAS-derived measurements with field data.
Main Results:
- The methodology correctly classified 92.1% of trees as burned or unburned.
- Field-validated crown scorch height extraction showed high accuracy (R-squared = 0.78, RMSE = 2.2 m).
- 83.3% of estimated crown volumes scorched were within ±10% of field measurements.
Conclusions:
- A novel, cost-effective RPAS-based methodology accurately characterizes individual tree fire severity.
- The approach estimates crown scorch height and volume at broader spatial scales than field data.
- Future improvements include shadow removal, spectral data calibration, and cost-benefit analysis.
