SCANFI: 2020 Canadian Forest Maps of Land Cover, Height, and Biomass at 30m Resolution
Updated 1mo ago
5filesTXT
Available on 1 platform
Sign in to view source links and access this dataset
Description
SCANFI is a set of 30-meter resolution raster files providing wall-to-wall maps for Canada's non-arctic landmass in 2020. The product includes broad land cover type, forest canopy height, crown closure, aboveground tree biomass, and species composition for several major tree species. It was developed by Natural Resources Canada using the National Forest Inventory photo-plot dataset, Landsat imagery, and regional k-nearest neighbours and random forest models.
Use Cases
Modeling forest carbon stocks based on aboveground tree biomass estimates.
Analyzing habitat connectivity and land cover change based on broad land cover type and canopy height maps.
Assessing regional forest composition for economic planning based on tree species cover layers.
Studying fire fuel loads and risk based on canopy closure and biomass data.
Strengths
Provides a consistent, wall-to-wall national map at 30-meter resolution.
Uses a novel imputation method based on hundreds of tile-level regional models.
Limitations
The description notes spectral disturbances from pests like the eastern spruce budworm are not fully represented, leading to imprecise estimates in affected areas.
Attributes for open stand classes (shrub, herbs, rock, bryoid) are described as potentially less reliable.
Tree species cover predictions, especially for less abundant species, have relatively high uncertainty.
Provenance
Source
Natural Resources Canada
Collection Method
Produced using the National Forest Inventory photo-plot dataset, temporally harmonized Landsat imagery, and k-nearest neighbours/random forest imputation.
Time Range
2020
Freshness
Last updated 2026-04-17 19:30:53.581865
Geography
Canada's non-arctic landmass
A newer version (SCANFI v2) exists at a separate DOI. Arctic ecozone vegetation attributes were predicted with a single model and could not be rigorously validated.