| Forest has functions of ecosystem services such as climate regulation,water conservation,and flood prevention and disaster reduction.Natural or man-made forest disturbances cannot be avoided.Therefore,it is of great significance to grasp the temporal dynamic changes and spatial distribution laws of forest disturbances,scientifically carry out regional forest management,and monitor forest disturbances reasonably and effectively.In this study,Hunan Province was selected as the research area,using Landsat long-term sequence image data sets from 1988 to 2018 as the main data source,based on the Google Earth Engine(GEE)platform,using LandTrendr algorithm,supplemented by land-use data and DEM elevation data,The global forest change data set and other data,based on extracting forest interference information in the study area,analyzed the spatial-temporal distribution characteristics of forest interference,and realized the historical reconstruction of forest interference in Hunan Province.The main research work and results of this paper are as follows:(1)Historical reconstruction and accuracy verification of forest disturbance in the study area.Based on the in-depth discussion of the LandTrendr algorithm,taking Landsat images from 1988 to 2018 as the main data,with the help of the GEE platform,through cloud removal,synthesis and other technologies to establish a reasonable and complete long-time sequence image set to carry out forest disturbance history Reconstruction has achieved dynamic monitoring of long-term and large-scale forest changes,and has filled the gaps in the study of the long historical period and the spatial pattern of forest disturbances in Hunan Province.Based on the 30m global forest change data set,a visual interpretation method was used to construct an independent verification data set to evaluate the accuracy of forest disturbance monitoring results.The results showed that the average recognition rate of forest disturbance patches was 68%,of which,more than The recognition rate of 29.5%of the sample plaques is between 70% and 80%,with high accuracy.(2)Analysis of the time-varying law of forest disturbance.Forest disturbances in the study area showed an obvious increasing trend year by year(Mann-Kendall test)as a whole,and occurred frequently in the past ten years.In the early period(1988-2001),the forest disturbance in the study area was relatively stable,and the annual interference area was less than 252.03 km2.However,in the following 18 years,the forest disturbance intensity increased significantly and had strong volatility,and reached its maximum value(676.96km2)in 2010.The study of forest disturbances in different cities showed the same change pattern:the changes in the early stage were gentle,and the subsequent changes were frequent.(3)Analysis of the spatial distribution pattern of forest disturbance.The forest disturbances in the study area are mainly distributed in the southern and eastern regions,with a wide distribution and distributed in fragments.Studies have shown that the distribution of forest disturbances is closely related to topography and landforms,mainly in low-altitude areas,with an average altitude of less than 421.7m.For example,cities with lower elevations such as Changde,Yueyang,and Yiyang,and cities with higher elevations Less distributed. |