Remotely Sensed Monitoring Of Grassland Aboveground Biomass In Northern Xinjiang Area

Xinjiang is one of China's three major pastoral areas.The north of Xinjiang Uygur autonomous region(Northern Xinjiang,hereafter)is the most important grassland areas in Xinjiang.There are masses of grassland,various herb types and enriched nutritive value.It is an important basis for animal husbandry.Therefore,Effective and accurate monitoring of the grassland vegetation biomass changes,will not only directly related to the decision-making of grassland animal husbandry management,but also indirectly affect the sustainable development of economy and society that is mainly inhabited by minorities.In order to rapidly monitor the growth condition of grassland vegetation,estimate aboveground biomass,and provide the basis for the balance between forage supply and animal demand,by means of the new generation satellite remote sensing data of EOS-MODIS(Earth Observing System-Moderate Resolution Imaging Spectroradiometer),'3S' technologies(Remote Ssensing,Geographical Information System and Global Positioning System)and statistical analysis approaches in the dissertation,based on science of grassland the grassland biomass remote sensing model is created,and the optimized biomass model is discussed,the biomass models for eight major types of grasslands are comparied and analyzed for providing an effective way of grassland monitoring on a large areas in the future.The main findings and conclusions in the dissertation are as follows:1.The value of EVI(Enhanced Vegetation Index)in the 5 types of alpine meadow, alpine steppe,temperature desert steppe,artificial grassland and swamp is less than NDVI (Normalized Difference Vegetation Index).The variation coefficient of EVI in Alpine grassland,alpine meadow and temperature desert steppe is less than that of NDVI,but it is greater than that of NDVI.in artificial grassland and swamp in pastoral areas.2.There exist obvious correlationships between NDVI,EVI values and aboveground biomass in Northern Xinjiang region.The exponent function model and linear function model of EVI is better than that of NDVI.EVI exponent function model can better estimate aboveground biomass(kg / hm~2)than EVI linear function.EVI exponent function model is y = 568.22e4.1008x(r = 0.792).3.In the 8 types of grasslands of low-land meadow,alpine meadow,mountain meadow,temperature meadow steppe,temperate desert steppe,temperate steppe, temperature desert and temperature desert steppe in the northern Xinjiang region,the EVI and NDVI biomass linear models are better than non-linear models for lowland meadow, alpine meadow,temperate steppe desert and temperate desert.For linear model,the correlationship between EVI and biomass is better than that of NDVI except for low-land meadow,alpine meadow and temperate steppe desert.For exponent model,the correlationship between NDVI and biomass is better than that of EVI except for low-land meadow,alpine meadow and the temperate desert steppe.4.In growing season of northern Xinjiang the aboveground biomass for every 16 days in 2004 are higher than that of the average values from 2004 to 2006.The monitoring value of aboveground biomass in 2006 is less than that in 2004.The aboveground biomass varies greatly in 2005,which the major reason is because there is a great difference due to the different inter-annual precipitation and temperature,and other climate condition.5.The maximum aboveground biomass for different types of grasslands in northern Xinjiang region occurs in different times,and mainly concentrates in the end of June to August.The maximum aboveground biomass from June to August can be used to calculate the grassland resources classification and its spatial distribution maps to provide a scientific basis for balance between forage supply and livestock demand and comprehensive evaluation.