LiDAR-Based Forest Thinning Design and Long-Term Impact Assessment in Wetzin’kwa Community

LiDAR data achieved a 0.91 F1-score for tree detection in a pine forest within the Wetzin'kwa Community Forest, British Columbia. This dataset, created by Ciyuan Jin and hosted on Borealis Harvested Dataverse, models commercial thinning paths and simulates their long-term impact on merchantable volume and stand health from 2025 to 2125 using the Forest Vegetation Simulator. The DBH estimation model from the study had limited explanatory power, with an R² of 0.21.

Use Cases

• Compare thinning path designs based on the least-cost path method described.
• Simulate long-term forest growth and merchantable volume based on the Forest Vegetation Simulator (FVS) outputs.
• Evaluate tree detection and DBH estimation accuracy using the provided LiDAR and field survey data.
• Assess the impact of variable-density versus fixed 40% thinning intensity on stand health recovery.

Strengths

• Includes a 100-year simulation horizon (2025-2125) for long-term impact assessment.
• Tree detection using LiDAR achieved a promising accuracy with an F1-score of 0.91.
• Compares multiple thinning approaches (variable-density and fixed 40% intensity) across three plots.

Limitations

• The DBH estimation model had limited explanatory power, with an R² of 0.21.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and specific file formats are unknown, which may limit suitability assessment.

Provenance

Source

Borealis Harvested Dataverse

Collection Method

LiDAR remote sensing combined with field survey data and orthophotos, analyzed with least-cost path method and Forest Vegetation Simulator.

Time Range

Simulation covers 2025 to 2125.

Freshness

Last updated 2026-05-02 04:10:35; freshness should be verified.

Geography

Wetzin'kwa Community Forest, British Columbia, Canada.