Study On Changes Of Forest Line In Changbai Mountain Based On Spectral Indexs And SFA Algorithm

The forest line ecotone is an important indicator of climate change in high altitude areas and is highly sensitive to global climate change.Therefore,the analysis results of forest line migration in high altitude regions can be used as an important basis for evaluating global climate change in alpine regions.In the previous typical studies,forest lines were divided by isotherm,tundra and birch forest alpine tundra classification methods,as well as tree ecology,so as to further explore the migration of forest lines.However,due to the different criteria used by different researchers in different regions and the high mountain sampling needs a lot of manpower,material and financial resources,etc.This study proposes a method to explore the range of forest line migration by directly detecting the change of vegetation cover type(tundra to birch)in the forest line migration area of Changbai Mountain.When the birch forest invades the tundra vegetation at higher elevations,it is manifested by the upward migration of the forest line.Compared with the changes of different types of ground features,the change of the upward movement of the forest line is slow and narrow under the influence of climate factors and topographic factors.Therefore,SFA algorithm is used to minimize the difference of invariant pixels in the transformation space,so that the changing region can be better highlighted.In this study,Landsat image data,DEM data and Google Earth images were used in two periods.On the basis of studying the migration of forest lines with normalized vegetation index(NDVI),the difference between the migration area of forest lines and the unchanged area of other vegetation types was highlighted by slow feature analysis(SFA)change detection algorithm.Based on the input data characteristics of the algorithm,the NDVI index threshold was "dynamically" divided by blackbody surface temperature(BBST)to study the forest line migration of the whole Changbai Mountain(including the Chinese side and the Korean side)in the past 30 years.In this study,vegetation index and change detection method were combined to obtain the results of forest line migration in the whole Changbai Mountain,and to realize the large-scale mapping and analysis of the climate ecological community in Changbai Mountain.The results show that the forest lines on the west and south sides of Changbai Mountain have obviously migrated in the past 30 years based on Landsat images,but the distribution is irregular rather than banding.In this study,the validity of the proposed method is verified by using the previous sample plot data.It is detected that the forest line of Changbai Mountain is located at an elevation of 2100 meters on the western slope,and there is no obvious forest line migration on the eastern slope.At the same time,continuous forest line migration pixels were detected near the elevation of 2000 meters on the southern slope of Changbai Mountain,and the forest line migration pixels were tested by using Google image and pixel similarity measure.In addition,the application of this method avoids replacing the whole mountain forest line migration with the regional sample forest line,which has a guiding role for the study of other mountain forest lines.