Mapping old-growth forests using airborne lidar data and satellite images: how do plot size and rarity affect accuracy?
Fuente:
Zenodo
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Recurso digital |
| Published: |
Zenodo
2025
|
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866901278885412864 |
|---|---|
| author | Räty, Janne Myllymäki, Mari Peltoniemi, Mikko Lehtonen, Aleksi Packalen, Petteri |
| author_facet | Räty, Janne Myllymäki, Mari Peltoniemi, Mikko Lehtonen, Aleksi Packalen, Petteri |
| contents | <p>The study investigates how to map old-growth forests in boreal regions using airborne LiDAR and satellite imagery, addressing their rarity and uncertain distribution. Using field data from 176 old-growth and 1,082 managed forest plots, a Gaussian process classifier was applied to distinguish forest types. Results show that larger plot sizes improve classification accuracy by better capturing structural patterns, with the highest performance (F1-score 0.74) achieved through data augmentation. However, detection accuracy decreases as old-growth forests become rarer. The proposed method supports more effective identification and planning of field inventories for conserving old-growth forests.</p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_1139_cjfr-2024-0283 |
| institution | Zenodo |
| language | |
| publishDate | 2025 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | Mapping old-growth forests using airborne lidar data and satellite images: how do plot size and rarity affect accuracy? Räty, Janne Myllymäki, Mari Peltoniemi, Mikko Lehtonen, Aleksi Packalen, Petteri <p>The study investigates how to map old-growth forests in boreal regions using airborne LiDAR and satellite imagery, addressing their rarity and uncertain distribution. Using field data from 176 old-growth and 1,082 managed forest plots, a Gaussian process classifier was applied to distinguish forest types. Results show that larger plot sizes improve classification accuracy by better capturing structural patterns, with the highest performance (F1-score 0.74) achieved through data augmentation. However, detection accuracy decreases as old-growth forests become rarer. The proposed method supports more effective identification and planning of field inventories for conserving old-growth forests.</p> |
| title | Mapping old-growth forests using airborne lidar data and satellite images: how do plot size and rarity affect accuracy? |
| url | https://doi.org/10.1139/cjfr-2024-0283 |