| On the basis of summarizing the relative research results of home and abroad,the degradations of Ruoergai grassland was diagnosed by avail of Landsat TM resource.The main contents include:explore the "Indexes of classification system of sub-alpine meadow degradation" through analyzing soil and vegetation changes;study the distributions of degraded grassland by supervised classification of TM image;find the most suitable model to simulate the above ground biomass of sub-alpine meadow by analyzing the vegetation indexes and field survey data and finally discussed the driving forces of grassland degradation.Through the above study,the following conclusions were drawn:(1) The "Indexes of classification system of sub-alpine meadow degradation" was established by analyzing the vegetation characters and soil components of zero degraded,light degraded,moderate degrade and heavy degrade grassland.(2) Under the ENVI and ARCGIS platform,it is found that the optimal band combination for Ruoergai grassland diagnosis is Band4,Band 3 and Band 2 combination.The Maximum Likelihood Supervised Classification was carried out with the TM432 to interpret the grassland degradations.The results showed that:the area of moderate, light,heavy degraded grassland are 3219.15km2,597.98km2,768.52km2 respectively, which take up 42.21%,7.84%,10.08%of the total targeted area respectively.The overall classification precision is 82.78%and the Kappa coefficient is 83.72%.(3) The statistical analysis was conducted between seven vegetation indexes(NDⅥ,RⅥ,DⅥ,SAⅥ,MSAⅥ,PⅥ,GⅥ)extracted from the Landsat TM which covers Ruoergai area in July 2008 and the field survey data.Lineal and non-lineal(second-degree polynomial,Cubic polynomial,Logarithm,Exponential) models were established between the seven indexes and the above ground biomass.The results showed that,the correlations between the 7 indexes and the aboveground biomass were highly significant which indicated that use the vegetation indexes to simulate the grassland biomass is a simple and feasible method,and eachⅥ-biomass regression model was significant at the 0.01 level except the Logarithm model.For the 6 vegetation indexes (NDⅥ,DⅥ,SAⅥ,MSAⅥ,PⅥ,GⅥ)-biomass models,the cubic polynomial model was the best,followed by the second-degree polynomial model,lineal model, exponential model;For the RⅥ-biomass model,the exponential model was the best, followed by the cubic polynomial model,second-degree polynomial model,and lineal model.Analysis showed that the polynomial model based on RⅥ-Biomass was the best model,with the multiple correlation coefficient R2 being 0.8177;the precision test of the model indicated that the average error was 6.8000%;the rating precision was 93.2000%.The simulated biomass based on this model showed that the Ruoergai grassland biomass was higher in the southeast than in the northwest area,which is consist with the results of Supervised Classification.(4) Through analyzing the natural and human activities,the result showed that,during the past 30 years,the temperature moved up year after year,the seasonal and 5-year movable average temperature ranged between 0.49℃and 0.55℃.The precipitation declined with the lineal rate up to -16.993mm·(10a)-1,and the evaporation increased with the lineal rate up to 7.621mm°(10a)-1.The desertification,livestock and population increased 22.76%,78%,80%respectively.Human activity and natural factors interwoven during the process of grassland degradation,the former take a dominant position and the latter couldn't be neglected.
|
PDF Full Text Request