Study On Information Extraction And The Dynamic Monitoring Of Grassland Coverage In Three River Source Area

Vegetation coverage is important eco-climatic parameter, it is also an important characterization of surface vegetation status. Vegetation cover change has great significance on alpine grassland ecosystem change, change in grassland communities, regional hydrology and the whole structure of alpine ecosystem. In this study,we estimated the vegetation coverage of alpine grassland by use improved second sub-pixel model of remote sensing and NDVI vegetation index. We propose two plans about the estimation of degree of vegetation coverage in Three River Source Area, and estimate the degree of vegetation coverage depends on second plan. we also discussed dynamic changes in vegetation cover of research area by using methods of GIS spatial statistical analysis and multivariable statistical analysis in detail. The main conclusions of the research are as follows:(1)Analyzing the fact of vegetation coverage in Three River Source Area, according to GB19377-2003 and feature of remote sensing data, select the vegetation coverage as evaluation index, established the monitoring and evaluation index system of vegetation coverage in Three River Source Area.(2)Estimate the areas of vegetation coverage in 2009 depending on improved second sub-pixel model: The area of high degree of vegetation coverage is 22625.87 km~2;The area of little-high degree of vegetation coverage is 30571.76 km~2;The area of middle degree of vegetation coverage is 32720.29 km~2;The area of low degree of vegetation coverage is 35254.8km~2;The area of lower degree of vegetation coverage is10100.26km~2;GCVI of Three River Source Area is 2.89 in 2009.The vegetation coverage of Maduo county in middle-eastern Three River Source Area is worst, GCVI is 3.3;GCVI is the highest, it is 2.34 in east 8 county, belongs to Medium grade. Estimate the areas of vegetation coverage in 2009 depending on improved second sub-pixel model: The area of high degree of vegetation coverage is 22625.87 km~2;The area of little-high degree of vegetation coverage is 30571.76 km~2;The area of middle degree of vegetation coverage is 32720.29 km~2;The area of low degree of vegetation coverage is 35254.8km~2;The area of lower degree of vegetation coverage is 10100.26km~2;GCVI of Three River Source Area is 2.89 in 2009.The vegetation coverage of Maduo county in middle-eastern Three River Source Area is worst, GCVI is 3.3;GCVI is the highest, it is 2.34 in the east eight county, belongs to Medium grade.(3) On the basis of calculating remote sensing image in Three River Source Area, analyze the changing trend of four district of five vegetation coverage type in Three River Source Area. The area of high degree of coverage reducing gradually in 25 years; The area of middle degree of coverage increasing gradually. It is a sign of slowing of dropping trend of degree of vegetation coverage; The area of low and lower degree of coverage reducing gradually, it indicated that the degradation of vegetation coverage is easing.(4) According to Markov process, we got vegetation coverage type transition probability matrices of three period in Three River Source Area, Analyze dynamic changing of vegetation coverage type, forecast spatial tribution situation of vegetation coverage in Three River Source Area. It indicated that the speed and trend of dropping of degree of coverage is slowing, but integrate dropping trend still lasting, the landscape is not optimistic.(5) The changing of degree of vegetation coverage in Three River Source Area is the result of interaction between human and nature factors; The pattern of dropping of vegetation coverage degree had formed in 80s last century.(6) Utilize NOAA GIMMS and MODIS MOD13Q1 data, probe the spatial changing of vegetation coverage, the results indicated: GIMMS and MOD13Q1 data can used for the study of large-scale changing of vegetation coverage, and effect is favorable. It is daring implore and data mining of these data for the using in grassland resource monitoring.