Research On The Relationship Of Forest Fires In Heilongjiang Based On Spatial Data Analysis

The forest resources are the most abundant in Heilongjiang Province, in which the forest fires also frequently happen, which have caused great harm to the forest resources in Heilongjiang Province, and the annual forest burned area is the first in the country. The occurrence of forest fire has its inevitability, and suddenness as well. With the accurate grasp of the basic situation of forest fire and scientific detection and forecast in time, forest fires can be effectively prevented and fought in time, and then the losses caused by forest fires can be avoided and reduced.Meteorological factors have a certain impact on the occurrence of forest fires, such as temperature, humidity, wind speed, precipitation, pressure etc. Focusing on the influence of meteorological factors and employing spatial data analysis, the paper mainly studies spatial clustering, that is whether there is a correlation between the occurrence of forest fires in Heilongjiang in 2005. Meanwhile, the influence of meteorological factors on fire is analyzed by regression analysis.Spatial data analysis refers to the analysis of the data with spatial location attributes. Through attention to spatial characteristics of the attribute data, in the analysis of spatial relations based on numerical analysis, the spatial interaction is highlighted and then spatial agglomeration and abnormity is analyzed to find out the spatial interaction mechanism of the studying objects.The following methods can find their ways in this thesis:principal component analysis, spatial clustering analysis, multiple regression analysis and geographical weighted regression analysis. Principal component analysis helps calculate the impact of forest fire on the meteorological impact index, spatial cluster analysis contributes to analyze and discuss whether the occurrence of forest fire is related to the existence of global and local space, finally, through multiple regression analysis and geographical weighted regression analysis, the effect of meteorological factors on forest fire intensity gets tested and compared with the two regression methods.The result shows that the forest fires in Heilongjiang province in 2005 and April, May and June in 2005 are all less than zero under the meteorological influence index Moran’s I, which suggests that, under the effects of meteorological factors, the occurrence of forest fires has negative spatial correlation on the overall situation. Through the statistical significance test, there presents dispersion state between the high and low forest fire areas. Through the scatter plot and the saliency map, the conclusion is drawn as the following:in the local space, the forest fire is more and more distributed, and the forest fire is in accordance with overall judgment. Using multiple regression analysis and geographically weighted regression analysis, the more influential meteorological factors of fire are found out:temperature, latitude, humidity and wind speed, and the direction and degree of the impact are significant. At the same time, through the comparison of the two regression tests, the geographical weighted regression test is found to be superior to the multiple regression analysis.