Skip to content
Open Data Finder
BC Data Catalogue

Fire Burn Severity - Historical

Published by Forest Analysis and Inventory Branch

About this dataset

This layer is the one-year-later burn severity classification for large fires (greater than 100 ha). Burn severity mapping is conducted using best available pre- and post-fire satellite multispectral imagery acquired by the MultiSpectral Instrument (MSI) aboard the Sentinel-2 satellite or the Operational Land Imager (OLI) sensor aboard the Landsat-8 and 9 satellites. The post-fire imagery is acquired during the subsequent growing season. Mapping conducted during the subsequent growing season benefits from greater post-fire image availability and is expected to be more representative of tree mortality. Every attempt is made to use cloud, smoke, shadow and snow-free imagery that was acquired prior to September 30th. Please note, this layer is 1-year-later burn severity dataset. The same-year burn severity mapping dataset (WHSE_FOREST_VEGETATION.VEG_BURN_SEVERITY_SAME_YR_SP) is considered an interim product to this layer. #### Methodology: • Select suitable pre- and post-fire imagery or create a cloud/snow/smoke-free composite from multiple images scenes • Calculate normalized burn severity ratio (NBR) for pre- and post-fire images • Calculate difference NBR (dNBR) where dNBR = pre NBR – post NBR • Apply a scaling equation (dNBR_scaled = dNBR*1000 + 275)/5) • Apply BARC thresholds (76, 110, 187) to create a 4-class image (unburned, low severity, medium severity, and high severity) • Apply region-based filters to reduce noise • Confirm burn severity analysis results through visual quality control • Produce a vector dataset and apply Euclidian distance smoothing

Fire Burn Severity - Historical · Open Data Finder