Pemodelan Spasial Resiko Kebakaran Hutan (Studi Kasus Provinsi Jambi, Sumatera)

Abstract

The lack of information and references about forest fires risk has caused forest fires controlling in Jambi Province could not be done effectively and efficiently. Therefore, a spatial model of forest fires risk based on predictor variables which cause forest fires must be established as one of the alternative innovation in supporting on forest fire controlling. Linear regression analysis on the relation between predictor variables (biophysics and human behavior/anthropogenic) and the occurrences of forest fires in Jambi Province showed significant influence 80.5%, with the form of the equation: Y = 2.380 to 0.02 X1 + 2.45 X2 - 0.67 X3 - 1 , 07 X4 - 0.79 X5 - X6 0,32 - 0,18 1,17 X7 + X8. Analysis of confusion matrix confirmed the accuracy of the model with accuracy rate 80.24% and spatially most areas in Jambi province belongs to the category of high risk of forest fires.