| Wetland is one of the most valuable ecosystems in the world, with a rich variety of animal and plant resources and unique features. Wetland vegetation biomass is one of the important basic data to describe the wetland, which is a reflection of the health of wetland ecosystem and an important significance of studying the ecological environment.The correlation between the remote factors and measured biomass is analyzed in this paper. The Honghe Wetland Nature Reserve is the study area and the Landsat-5 TM data and measured biomass is the data source. The linear and curve models are established with the remote factors and the total biomass. After analyzing, the perfect model is achieved, which is improved by spatial analysis. Finally, the total biomass spatial distribution maps of 2004, 2006 and 2008 are made, and the trend of the biomass is analyzed in this paper. The conclusions of the research are got.1. Using the factor extracted from the TM images as independent variables to estimate the aboveground biomass of the study area can achieve a higher accuracy. 6 factors (TM3, TM5, TM7, NDVI,RVI and WET) as independent variables of the linear model are the best, and the accuracy is 84.8168% and 83.0644%.2. In this paper, the correlation between the wet weight and the biomass and the correlation between the dry weight and the biomass are analyzed, and the latter is higher than the former.3. By analyzing the total biomass spatial distribution maps of 2004, 2006 and 2008, we can know that the dry biomass vary from 300 g/m2 to 900 g/m2 and the weight biomass vary from 600 g/m~2 to 1800 g/m~2. High biomass and low biomass are relatively small. By analyzing the total biomass of the three years, the total biomass shows a downward trend.
|
PDF Full Text Request