Wildfire Knowledge Graph · Fire Investigation
A structured category assigned to fire cause based on the investigation framework and strength of available evidence.
In plain language
A structured category assigned to fire cause based on the investigation framework and strength of available evidence.
Why it matters
Cause Classification is useful because it separates one specific part of wildfire behaviour, exposure, operations or evidence from nearby concepts that can otherwise be mixed together.
Cause classification: preserve the source’s coding scheme
Agencies group fire causes into categories to support investigation, statistics and management. Categories such as natural, human-related and undetermined may be further subdivided. Classification schemes are not perfectly identical between jurisdictions, datasets or reporting years.
Data example: a regional dashboard may combine equipment use, recreation and debris burning within human-related causes while another dataset records them separately. Summing these two tables as if the labels were identical produces a misleading comparison, even if both use official records.
Before aggregating causes, preserve the raw category, agency, reporting year, definition, count unit and whether data describe incidents, investigated ignitions or final determinations. Report missing, unresolved and uninvestigated cases transparently; do not redistribute them among known causes to make percentages add to 100%.
Incident-specific attribution should rely on investigation findings, not category prevalence. A statistical pattern can motivate a research question but does not establish the cause of any named fire.
Compare Fire Cause, Undetermined Cause and the Source Library.