Individual Garry Oak Tree Crowns Mapped via LiDAR and Imagery in Greater Victoria

A 2026 study by Kate Mcintyre produced a map of individual Garry oak tree crowns across the Greater Victoria region. The dataset was created using an automated workflow that fused airborne laser scanning and four-band multispectral imagery to train a Random Forest classifier. The model achieved an overall accuracy of 89.7% and a specificity of 96.2%.

Use Cases

• Establishing a baseline for conservation planning based on the high-specificity tree crown map.
• Informing targeted habitat restoration strategies based on the location of individual Garry oaks.
• Developing scalable, non-invasive monitoring frameworks for endangered ecosystems based on the described methodology.
• Training or validating other tree species classification models based on the structural and spectral metrics used.

Strengths

• Model performance is quantified with an overall accuracy of 89.7% and a specificity of 96.2%.
• The methodology provides a non-invasive and scalable approach for monitoring endangered ecosystems.
• The dataset addresses a noted knowledge gap by providing high-resolution spatial data for a critically endangered ecosystem.

Limitations

• The model's sensitivity was 61.0%, indicating challenges in discriminating interlocking crowns and non-native species in urban areas.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and file formats are unknown, which may limit suitability assessment.

Provenance

Source

Borealis Harvested Dataverse

Collection Method

Automated workflow using a fusion of airborne laser scanning and multispectral imagery, with a Random Forest classifier.

Freshness

Last updated 2026-05-02 04:11:25; freshness should be verified.

Geography

Greater Victoria region