| Wuyi Mountain is a world heritage site for both nature and culture.It is also a forest fire prone and high-risk area.This paper takes Wuyishan City as the research area,and uses statistical analysis method and multi-distance spatial clustering method(Ripley’s k function)based on statistics,meteorological factors,topographical factors,vegetation factors,and social environmental factors of forest fires in Wuyishan City from 1980 to 2013.The temporal and spatial pattern of forest fires was studied.The linear regression was used to analyze the correlations between forest fires(number of fires,burned areas),annual and monthly meteorological factors,and the model was fitted with significant influence factors;Logistic model and GWLR model were used to construct the forest fire forecasting model,and the models were fitted and compared.The optimal model was selected to further classify the forest fire risk rating.The main results are as follows:(1)In Wuyishan City,the number of forest fires and the over-fired area changed greatly from 1980 to 2013.The peak value and the valley value alternated.In the 1980s,both the number of fires and the burned area were high.Since the 1990s,the forest fires were at a low value.Overall,the number of forest fires and the number of burned areas showed a downward trend.The monthly distribution pattern of forest fires in Wuyishan City from 1990 to 2013 shows that the number of forest fires and the burned area are concentrated in the winter and spring seasons,especially from December to April of the following year.Among them,forest fires are the most serious in March.This is related to the entire Fujian Province.The critical period of forest fire protection coincides.The law of distribution of forest fires in Wuyishan City from 1999 to 2013 shows that the distribution rules of forest fires are obvious during the day.Forest fires occur frequently in the afternoon(13 to 17 o’clock).(2)Judging from the layout of administrative districts,the forest fires in Wuyishan City from 1999 to 2013 were concentrated in the central and southern regions,and were mainly distributed in the jurisdiction forest areas such as Xingtian,Chong’an,Xinfeng,and Wuyi.From the standpoint of spatial distance,the forest fire points are clustered on the spatial scale,and the fire clusters have the highest degree of clustering around 6km.The Moran’s I coefficient was used to analyze the spatial distribution of the burned area and the damaged forest area in Wuyishan City.The Moran’s I coefficient was less than 0,and P>0.05 failed the significant test,which showed that the forest fires in Wuyishan City had agglomeration and distribution.However,the area affected by forest fires does not constitute spatial autocorrelation but tends to be randomly distributed.(3)Forest fires in Wuyishan City are dominated by man-made fires,of which forest fires caused by agricultural production activities such as burning and burning fields account for 58.1%;and forest fires caused by natural causes(lightning)only account for 4.1%.Therefore,relevant departments should strengthen universal education and raise people’s awareness of forest fire protection.(4)The number-of forest fires in Wuyishan City from 1980 to 2013 was significantly negatively correlated with the annual average precipitation(r=-0.456,P<0.01),and the burned area was significantly negatively correlated with the annual average precipitation(correlation coefficient r=-0.445,P<0.01),a significant positive correlation with annual average wind speed(correlation coefficient r=0.346,P<0.05).However,the number of forest fire occurrences and burned areas in Wuyishan City did not form a linear correlation with the monthly meteorological factors from 1990 to 2013.Through the principal component analysis,it was found that the number of fires per month and the burned area of the moon were all influenced by multiple meteorological factors.As a result,different months are affected by different meteorological factors.(5)The logistic model of forest fire forecast was established.From the model,it can be seen that the daily average relative humidity,sunshine hours,altitude,and railway The distance and other fire risk factors have a significant impact on the occurrence of forest fires in Wuyishan.The probability of forest fires decreases with the increase of relative humidity and altitude,increases with the increase of sunshine hours and railway density.The study is in line with local conditions.The prediction accuracy of the model is relatively high 84.4%.According to the GWLR model fitting,the fire risk factor coefficient is non-stationary in space,and there are both positive and negative correlations between sunshine hours,distance to railroad and altitude and forest fire occurrence.The relative humidity and forest fire The occurrence of negative correlation in space indicates that the fire risk factor in Wuyi Mountain has spatial heterogeneity.Comparing the results of the two models,the accuracy of GWLR model is 85.3%,and the fitting effect is better than the traditional binary Logistic model.(6)According to the forecasting probability of forest fires in Wuyishan City,the risk classification of forest fires in the study area was determined.It was found that the forest fire risk grades in the middle and southern parts of Wuyishan City are high,and the high fire risk area is more concentrated,with Wuyi Street,Xinfeng Street and Chongan Street.The forest areas under the jurisdiction of Xingtai Town are the most obvious,and other regions are classified as middle and low forest fire risk,and the probability of forest fires is relatively small.Therefore,special attention should be paid to the forest fire prevention work in high fire risk areas in central and southern areas of Wuyishan City. |
PDF Full Text Request