Using GIS Spatial Analysis and Logistic Regression to Predict the Probabilities of Human-caused Grassland Fires
Journal of Arid Environments, Volume 74, Issue 3, March 2010, Pages 386-393
Z.X. Zhang, H.Y. Zhang, and D.W. Zhou
“This study created a method to examine the probability of human-caused ignition in grassland in the east of Inner Mongolia, China. Spatial analysis was used to locate the human-caused ignition of grassland fires that occurred in the study areas during 1976–1996. The area was divided into grid cells, and for each grid cell the distances to various human-built infrastructure and weather factors were determined. Logistic regression was used to build a predictive model of the probability of human-caused ignition of grassland fires in each of the grid cells, using topography, distances, and weather factors as predictor variables.
“The probability of a grid cell ignition was positively correlated with temperature, and elevation; and negatively correlated with precipitation, relative humidity and distances to villages, dirt roads, railroads and isolated buildings. Grid cells were grouped into five classes of ignition probabilities, i.e. <0.2, 0.2–0.4, 0.4–0.6, 0.6–0.8 and >0.8. These methods and results should be useful to local administrative managers in assessing initial ignition risk for grassland.”
Trackbacks