Lightning-caused Forest Fires Risk Assessment Based on Historical Fires Events in DaXingAn Mountains of China

Lightning is a dangerous natural source of ignition, which has been the main cause of forest fires in some countries. DaXingAn Mountains area is the largest natural frigid-temperate zone coniferous forest ecosystem, but it is also one of the serious areas under the influence of lightning-caused fires in China, Therefore, it is very useful to predict the lightning-caused fires occurrence. In this paper, 238 records of lightning-caused forest fires and MODIS 8-day combined reflectance product and LST product were collected. LST, EWT; NDVI, TCI, VCI and WCI were derived to express the weather and vegetation condition from the MODIS data, meanwhile DEM, Slop, Relief and RPI were also calculated as topographic parameters. With these ten fire environment condition parameters, a logistic regression model was built to predict the probability of lightning-caused forest fire occurrence. The result showed that the accuracy of prediction was 63.4%, which was much helpful for the forest managers to prevent fires.