Applying neural network technology to human-caused wildfire occurrence prediction
Document Type: Journal Article
Author(s): C. Vega-Garcia; B. S. Lee; P. M. Woodard; S. J. Titus
Publication Year: 1996

Cataloging Information

  • Alberta
  • Canada
  • disturbance
  • elevation
  • fire danger rating
  • fire frequency
  • fire management
  • forest management
  • fuel types
  • GIS
  • human caused fires
  • land management
  • statistical analysis
  • topography
  • wildfires
  • wildlife habitat management
Record Maintained By:
Record Last Modified: June 1, 2018
FRAMES Record Number: 36049
Tall Timbers Record Number: 10386
TTRS Location Status: In-file
TTRS Call Number: Fire File
TTRS Abstract Status: Okay, Fair use, Reproduced by permission

This bibliographic record was either created or modified by the Tall Timbers Research Station and Land Conservancy and is provided without charge to promote research and education in Fire Ecology. The E.V. Komarek Fire Ecology Database is the intellectual property of the Tall Timbers Research Station and Land Conservancy.


Human-caused forest fires are a serious problem throughout the world. Believing that there are predictable characteristics common to all fires, we analyzed the historical human-caused fire occurrence data for the Whitecourt Provincial Forest of Alberta using artificial neural network and geographic information system (ARC/INFO) technology. These data were also analyzed using logistic regression analysis (the binary legit model), which served as the 'domain expert' to identify the im portant input variables. A 314 fire and no-fire data set for the period 1986-1990 was used for training. The observations were whether at least one fire occurred, on a certain day, in one of the eight geographic zones defined within the study area. The models developed were tested using data from the 1991-1992 fire seasons, which had 58 fire obser- vations. Using as input variables the Canadian Fire Weather Index for the day, area in km2 of the geographic zone, and district (a 0/1 dummy vari- able from the logistic regression model, which accounts for observa- tions within a forest district where human use is higher), the resultant model had four input nodes and two output nodes, and correctly predicted 85 percent of the no-fire observations and 78 percent of the fire obsenmtions.

Vega-Garcia, C., B. S. Lee, P. M. Woodard, and S. J. Titus. 1996. Applying neural network technology to human-caused wildfire occurrence prediction. AI Applications, v. 10, no. 3, p. 9-18.