| Forest ecosystem is one of the most important terrestrial ecosystems on earth,and is one of the important components of the global carbon pool,which has an irreplaceable role in mitigating global warming and maintaining global carbon balance.As a major part of the Tibetan Plateau,changes in forest landscape patterns in Tibet have been a hot spot for research in related fields.The study of changes in forest landscape patterns in Tibet can provide a better understanding of the material cycle and energy flow in Tibetan forest ecosystems.An in-depth study of the dynamic changes of patches,types and structural functions of the Tibetan forest landscape can provide timely analysis of the dynamic changes and fragmentation of Tibetan forest landscape characteristics,and then analyze the connectivity of each type of landscape to provide important references for forestry departments to adjust forest protection policies in a timely and dynamic manner.Through an in-depth study of the dynamic changes of forest landscape patterns and changes in landscape pattern indices in Tibet,as well as combining the research results of experts in related fields,the relevant driving factors affecting the forest landscape patterns in Tibet are identified.On this basis,a geographically weighted regression model is used to quantitatively analyze the characteristics of landscape pattern changes in Tibet,to provide a scientific theory for the changes in forest landscape patterns caused by endogenous succession in Tibetan forests and the degree of productivity,species diversity,and stability resulting from their changes,and to provide a theoretical basis for land use planning decisions and sustainable management and utilization of forest resources in Tibet.In this paper,we used the ESSA land use dataset with a spatial resolution of 300 m as the data source and Arc GIS,ENVI and FRAGSTAT software to specifically analyze the forest landscape pattern distribution,landscape pattern index and dynamic change characteristics of Tibet Autonomous Region for four periods in 1992,2000,2008 and2016.A comprehensive analysis of landscape pattern indices in terms of area,patch and type,and dynamic changes in landscape is conducted to explore the landscape pattern characteristics and changes in the Tibet.Autonomous Region,and the overall landscape types are analyzed in terms of fragmentation characteristics,patch shape at landscape level and patch spatial distribution characteristics at characteristic landscape level,while the spatial heterogeneity characteristics of coniferous,broad-leaved,mixed coniferous and shrub forests in the forest landscape are analyzed in depth.The spatial heterogeneity of coniferous,broad-leaved,mixed coniferous and shrubby forest types in the forest landscape was analyzed.Geographically weighted regression models are used in combination with other data to study the changes in forest landscape patterns and driving forces in Tibet during this period.The main findings of the paper are as follows:(1)During the period of 1992-2016,the overall area of forest landscape is up to 159000km2,the area is increasing first and then decreasing,and the maximum amount of dynamic change in forest landscape area is-586km2,accounting for 0.36% of the overall forest landscape area.In the forest landscape,the size of the annual average area change is shrub forest > coniferous forest > broadleaf forest > broadleaf forest in order,among which the shrub forest has the largest annual average area change of-81.26km2,while the mixed coniferous forest has the smallest annual average area change of 0.2km2.During 1992-2016,the pattern change of each landscape type in the TAR at the landscape level is a gradual increase in NP and PD as a whole.The various landscape pattern indices indicate that the degree of forest landscape fragmentation is increasing and the complexity of patch shapes of each landscape type is slowly decreasing.The spatial and temporal changes in the forest landscape of Tibet are small,but the dynamic adjustment and changes within the forest landscape and other landscape types are constantly occurring,the diversity of the forest landscape is constantly stabilizing,the degree of fragmentation is improved,the connectivity of the landscape is better,the spatial heterogeneity is constantly being improved,and the forest resistance to disturbance is constantly being enhanced.(2)In this paper,by testing the redundancy values associated with nine main factors of topographic,climatic and socio-economic factors,such as elevation,slope,slope direction,shade,precipitation,wind speed,temperature,GDP and population density,we found that two factors,elevation and slope direction,have larger variance expansion coefficients and lower tolerance compared with other driving factors,and excluded two driving factors,elevation and slope direction,leaving slope,shade The average weight of these seven main factors is: slope > temperature >precipitation > shade > wind speed > GDP > population density,but the weight of their influence varies in different periods.The magnitudes of their influence weights vary in different periods,and slope,precipitation,temperature,GDP and population density all fluctuate only within a certain range,while the driving effects of two influence factors,wind speed and shade,on the Tibetan forest landscape range from significant driving to weak driving or even no driving with the extension of time.The results of the geographically weighted regression model analysis indicate that in Tibetan region,the driving effects of each driver are not independent,but interact through multiple drivers to jointly influence the distribution of forest landscape patterns in Tibet. |
PDF Full Text Request