Study On Forest Fire Driving Factors And Fire Risk Zoning In Central Yunnan

Forest fire is one of the most important and active ecological disturbance factors in the global forest ecosystem,which continuously affects the landscape structure and function of the forest.Xinping County,a typical representative area in Central Yunnan,is one of the typical distribution areas of warm temperate evergreen coniferous forest and pinus yunnanensis forest in China.This area has a long fire history and fire culture.Forest fires with high intensity and high frequency often occur,which has caused many adverse effects on social and economic benefits and the sustainable development of forest ecosystem.Under the background of increasing extreme fire risk weather,understanding the temporal and spatial differentiation characteristics of forest fire in this region,quantitatively analyzing the occurrence law of forest fire,identifying the high incidence areas of forest fire,revealing the impact of various factors on the temporal and spatial distribution pattern of forest fire,developing the county forest fire risk prediction model and accurately dividing the forest fire risk grade are the basis of scientifically formulating forest fire management policies and strengthening the regional fire management system,It is of great significance to maintain the stability of forest ecosystem structure and function and ensure the sustainable development of forest resources.It can provide scientific guidance for forest fire management and fire prevention resource allocation of relevant departments,help to strengthen fire prevention project construction and fire prevention publicity in key fire prevention areas,and arrange fire prevention work scientifically and efficiently.Therefore,Xinping County,a typical representative area in Central Yunnan,was selected in this study.Based on MODIS Fire Product data(mcd14dl)from 2001 to 2020,the temporal and spatial distribution characteristics of forest fire in Xinping County from 2001 to 2020 were studied by using mathematical statistics,spatial analysis,multi distance spatial cluster analysis and nuclear density analysis;Based on the data of climate,vegetation,topography and social infrastructure,the contribution rate of various factors to forest fire and the relative importance of the interaction of various forest fire driving factors to forest fire were quantitatively analyzed by mathematical statistical analysis;The maximum entropy model is used to predict the forest fire risk in Xinping County,Central Yunnan,and the fire risk level of the county is divided.At the same time,the accuracy of the forest fire prediction model is evaluated by using the working characteristics of subjects(ROC curve).The results show that:(1)Forest fires are unevenly distributed in time and space.From 2001 to 2020,the number of forest fires in Xinping County in Central Yunnan fluctuated greatly,with high peaks in 2005 and 2012 respectively;The seasonal variation of fire is obvious in spring and winter;Under different temporal and spatial scales,forest fires in this area show aggregation distribution mode,and the degree of aggregation is high;There are two significant forest fire hot spots,which are located in the northeast of Mosha town and at the junction of Gucheng street Pingdian Township Yangwu town.(2)The relationship between forest fire occurrence times and driving factors is complex.There is not a simple linear relationship between the number of forest fires and the driving factors of forest fires in Xinping County,Central Yunnan.Among them,with the increase of meteorological factors(annual precipitation,average annual wind speed,average annual temperature and relative humidity),the number of forest fires increases first and then decreases;In addition,topographic factors(altitude,slope and aspect)also have a profound impact on the number of forest fires.In the area with an altitude of 500-1500 m,the number of forest fires is the most;Forest fires mainly occur in forest areas with low slopes and are evenly distributed on all slopes;The influence of distance from roads,residential areas and rivers on forest fire showed a downward trend.The farther the distance,the lower the number of forest fires;Evergreen coniferous forest(Pinus yunnanensis)is the vegetation type with the largest number of forest fires in Xinping County.In short,the main influencing factors of forest fire in Xinping County are human factors,and other factors(meteorological factors and topographic factors)have low influence on forest fire.In addition,the driving force of the interaction of each forest fire driving factor is higher than that of a single factor,and the interaction type is nonlinear,and the interaction is enhanced.Among them,the distance from residential areas,annual temperature,altitude and other forest fire impact factors have the most significant impact on the number of forest fires in Xinping County,but the interaction between annual temperature and residential areas is the strongest.(3)The maximum entropy model has good simulation accuracy and can effectively classify the fire risk level.The AUC values of 75% of the training data and 25% of the validation data are 0.841 and 0.711 respectively.The simulation accuracy of the maximum entropy model for forest fire risk is in the middle level,and the prediction results are more accurate.In addition,the boundary of fire risk level in Xinping county is obvious.Taking the water system of Yuanjiang River as the boundary,the overall fire risk level of the region east of Yuanjiang River is generally higher than that of the region west of Yuanjiang River.Among them,the extremely high fire risk areas are mainly concentrated in Yangwu town;According to the statistical calculation of five types of fire insurance grades(very low fire insurance,low fire insurance,medium fire insurance,high fire insurance and extremely high fire insurance),the five types of forest fire insurance grades account for 27.10%,35.81%,28.02%,8.73% and 0.34% of the area of Xinping County respectively.