Research On Long Time Series Forest Disturbance Monitoring And Driving Forces In Guangdong Province Based On Landsat Images

As an important part of the earth’s biosphere,forests can regulate the circulation of ecosystems and play the role of ecological barriers.However,with the rapid development of society,forest resources are subject to various disturbances,resulting in a large degree of destruction of vegetation,posing a great threat to the ecosystem.Guangdong Province is not only rich in forest resources,but also has rapid economic development.Therefore,the study of forest disturbance in Guangdong Province is of great significance for the sustainable management of forest resources and the study of ecosystems.Remote sensing technology can obtain ground information in a large area,quickly and repeatedly,so it provides strong data guarantee and technical support for the research of forest resource interference monitoring.After the forest is disturbed,the degree of forest landscape fragmentation is aggravated.However,there are few studies on the forest landscape fragmentation after forest disturbance,especially the spatial process of landscape fragmentation.In addition,the driving factors of forest disturbance are complex,and there are interactions between the driving factors.However,many current methods of studying driving mechanisms ignore the impact of this interaction.Therefore,if we want to obtain accurate forest disturbance information,on the basis of extracting forest disturbance,we need to further study the impact of forest disturbance on forest landscape and analyze the driving mechanism of forest disturbance.Based on Landsat time series data,this paper constructs an analysis and research framework for forest disturbance and driving factors based on long-term remote sensing images by extracting forest disturbance areas and disturbance types in Guangdong Province from 1990 to 2019.The main research contents and innovations of this paper are as follows:1.Combining the LandTrendr algorithm and the random forest algorithm to extract the forest interference area and interference type in Guangdong Province from 1990 to 2019,and analyzing the spatio-temporal change characteristics of forest interference from a multi-scale perspective.Studies show that the area of forest disturbance in Guangdong Province showed a fluctuating upward trend from 1990 to 2019;the average value of the area of forest disturbance showed a trend of "clustering".In different stages and different regions,each interference type presents different distribution characteristics.2.Based on the forest fragmentation process model to quantify the spatial process of forest landscape fragmentation,discovering the change characteristics of the spatial process of forest landscape fragmentation,and further analyzing the relationship between the type of disturbance and the spatial process of forest landscape fragmentation.The results show that the forest disturbance patches are mainly contracted,and the distribution trend of the spatial process of forest fragmentation migrates inland as a whole;the integrated cumulative counting method and the main approach counting method analyze the relationship between the type of disturbance and the spatial process of forest landscape fragmentation.3.Combining multiple driving factors to explore the driving mechanism of forest disturbance and the spatio-temporal changes of disturbance types from the perspectives of spatial distribution and dynamic changes,which can analyze the relationship between forest disturbance and various influencing factors more comprehensively.This paper studies the spatial distribution driving mechanism of forest interference based on geographic detectors,and at the same time studies the driving mechanism of the spatial and temporal dynamic changes of the area of forest disturbance area based on the correlation coefficient method.The study found that population density has the greatest impact on the spatial pattern of forest disturbances,and pairwise interactive detection of driving factors enhances the influence;The dynamic driving factors of different cities have different degrees of correlation with the area of forest disturbance,which are related to the economic development direction and regional characteristics of each region;And the degree of influence of driving factors on each type of forest disturbance varies in different time periods.