Estimating wetland vegetation biomass in the Poyang Lake of central China from Landsat ETM data

The estimation of the biomass of wetland vegetation is very important for wetland environment research. A method of rapidly estimating the biomass is discussed in this paper by using the combination of Remote Sensing, Geographic Information System (GIS) and Global Position System (GPS). By using Landsat ETM data of April 16,2000 and the field data investigated contemporaneously, the authors estimated the biomass of wetland vegetation in the Poyang Lake, the largest fresh-water lake in China. Firstly, a reasonable sampling route for a measure of the biomass was designed by utilizing the false color composite derived from the ETM data as one of the main references, and then the measurement was carried out directed by the portable GPS and the imagery during April 18-28, 2000. Secondly, both the sampling data and the ETM data were geometrically corrected to an equal-area projection of Albers. Thirdly, several vegetation indices and the principal component are derived from the Landsat ETM data. Finally, The correlation between the biomass, the Landsat ETM and its derived data was analyzed. The results showed that the sampling biomass data was best relative to the ETM 4 data with the highest coefficient of 0.86 at the significant level of 0.05, and the following was the DVI and NDVI with 0.83 and 0.80 respectively. Therefore, band 4 data of ETM was used to estimate the total biomass of entire Poyang Lake. This study also provided a case in that a single band data of near infrared could be used to estimate the wetland vegetation biomass