| Forest is a renewable natural resource as well as an intangible environmental resource and potential "green energy".The status and changes of forest resources not only affect regional and global environmental changes,but also have a profound impact on the sustainable development of social economy.Therefore,the significance of conducting forest resource survey and monitoring is very important.Since the 1970 s,aerospace remote sensing technology has made great progress.The spatial resolution and time resolution of sensors have been greatly improved.The emergence of multi-sensor,multi-temporal,and multi-spectrum is also for studying the status and development of forest resources.Provides the ideal tool.This study uses Landsat images and MODIS time series data launched by the National Aeronautics and Space Administration(NASA)as the basic data source,supplemented by elevation and slope information.Use the object-oriented remote sensing classification method of forest vegetation to achieve high-precision information extraction and classification of the spatial distribution of forest resources in the Russian Far East,and compare the selected points based on the results of random and historical classification of images for accuracy verification;use the landscape pattern index method to analyze the forest landscape Spatial pattern;At the same time,the maximum value synthesis method is used to obtain the overall NDVI value and FVC value,and the forest resources in the Russian Far East in 1990 and 2015 are analyzed and compared through the three aspects of forest area in and out conversion,landscape dynamics,and vegetation growth status.The main research conclusions are as follows:1.Combining DEM data,multi-temporal information data and time series data can extract feature information in medium-resolution TM remote sensing images faster and more effectively,so as to reflect the true surface conditions;and it is more intuitive and The real-time nature can accurately reflect the distribution pattern of the secondary forest types on the ground.2.Through the verification of randomly selected points on Google Earth,the overall accuracy of this forest classification method in this study is relatively good.In1990 and 2015,the forest classification accuracy of the Russian Far East was 90.13%and 91.27%,respectively,which met the accuracy requirements.3.In the past 25 years,the total forest area in the Russian Far East has decreased by 14,600 km~2,and the area of coniferous forests,broad-leaved forests,and shrubs have increased.From the perspective of landscape pattern,the number of patches(NP)and patches of mixed coniferous and broad-leaved forests The increasing density(PD)indicates that this landscape has become more fragmented;in 1990,the coniferous and broad-leaved mixed forest was the dominant forest landscape type in the Russian Far East.After 25 years of conversion,shrubs became the dominant forest landscape type in the Russian Far East in 2015.4.According to the spatial slope distribution of NDVI and FVC,the overall vegetation status in the Russian Far East has not changed drastically.The forest stock volume is generally increasing.In 2015,the total forest stock volume in Russia was81.488 billion m~3,the above-ground biomass increased by about 153 million tons every five years,and the underground biomass increased by about 82 million tons every five years.. |
PDF Full Text Request