Study On Forest Map And Parameters Inversion Using Multi-temporal ENVISAT ASAR Data

Forest is the ecosystem over land that has the largest area, prolific materials, most complicated structure and is distributed most widely.The remote sensing technologies are most hopeful to provide quantitative information for forest resources inventory and monitoring .But the visible light remote sensing technology is limited to receive the data because of the weather or light condition. With its all-weather, all-time capabilities, radar remote sensing has great potential for forestry management. The application study of radar remote sensing data on the forestry is a hot problem, but the development in our country is still nripe.In this study, multi-temporal ENVISAT ASAR was used for forest mapping and stem volume estimate. Unifies the SAR polarization characteristic,the backscattering characteristic of multi-temporal ENVISAT ASAR data of different land use type and different forest type were analyzed; The one timeporal and HH, HV and dual-polarization of multi-temporal ASAR data were used for land use classification research in decision tree classification method. The result from this study shows that the improvement of water classification accuracy owes to the summer data, and the classification accuracy will be improved when we used the multi-temporal data, HV polarization data is more advantageous for forest map.Using the statistics method, the scatter plot between the backscatter coefficient of HH polarization and HV polarization and HV/HH polarization and forest volume were protracted, the regression equation was made, and estimated forest volume. Its RMSE is 16.526m~3/ha.