Burn Severity Datasets for British Columbia 1985-2019
收藏NIAID Data Ecosystem2026-05-02 收录
下载链接:
https://zenodo.org/record/14811611
下载链接
链接失效反馈官方服务:
资源简介:
Burn severity datasets for wildfires in British Columbia (BC) that occurred between 1985 and 2019 were created based on methodology and JavaScript code developed by Parks et al. (2018) with minor modifications to account for BC's growing season: pre- and post-fire scene selection dates were changed to June 1-Sept 30. Floating point geoTIFFs were created in Google Earth Engine’s code development environment using JavaScript; all other processing was completed using ArcGIS 10.6 & Python 2.7.
Only wildfires >25ha are included in this repository. Fire perimeters from BC's Wildfire Perimeters - Historical spatial dataset were used to determine wildfire extents. Fire numbers correspond to the [FIRE_NUMBER] field within this dataset.
Please be aware that no field verification has been performed on this data.
Burn Severity Metrics
dnbr – differenced normalized burn ratio
rbr – relativized burn ratio
rdnbr – relativized differenced normalized burn ratio
Each metric has also been run with an offset (see “_w_offset” directories), which includes pixels from a 180m buffer outside the mapped fire perimeter. This offset may help assess unburned areas within the fire perimeter as long as the cover types outside the perimeter are similar to the cover types within the perimeter.
Available Datasets
There are 5 raster and/or vector datasets available for each fire:
1. Floating point geoTIF as exported from Google Earth Engine (GEE)
o __.tif
o ex. C10006_1985_dnbr.tif
2. Integer geoTIF
o Created from the floating point geoTIF
o ___int.tif
o ex. C10006_1985_dnbr_int.tif
3. Integer geoTIF clipped to mapped fire perimeter
o Fire perimeter from the BC Wildfire Perimeter - Historical spatial dataset , with any internal holes removed, is used to clip the integer geoTIF
o ___masked.tif
o ex. C10006_1985_dnbr_masked.tif
4. Reclassed clipped geoTIF
o Clipped geoTIF is reclassed using the thresholds in Table 1 below (from Parks et al. 2018).
o ___reclass.tif
o ex. C10006_1985_dnbr_reclass.tif
Table 1: Burn severity thresholds from Parks et al.
Low
Moderate
High
dNBR
<=185
186-417
>=418
dNBR with offset
<=159
160-392
>=393
RdNBR
<=248
249-544
>=545
RdNBR with offset
<=212
213-511
>=512
RBR
<=135
136-300
>=301
RBR with offset
<=115
116-282
>=383
5. Reclassed burn severity shapefile
o Reclassed geoTIF is converted into a polygon shapefile and the classification is converted to text (Low, Moderate, High); no smoothing is applied
o ___reclass_poly.shp
o ex. C10006_1985_dnbr_reclass_poly.shp
Known Issues
A small number of fires failed during the Google Earth Engine processing (see Table 2). These fires have no burn severity classification available.
Table 2. Fires without burn severity information available
Year
Fire Number
1985
N50175
N70035
V70088
1988
V50004
1990
C50050
K10013
R50117
V50055
VB0002
1992
K50034
1993
VB0007
创建时间:
2025-02-06



