Hyper spectral remote sensing is widely used to identify ground objects as a result of the advantages of ground radiation intensity characteristics and spectral position characteristics, in which inversion of vegetation components is the difficult point and hotspot. In this study, Huma county of Heilongjiang Province was selected as the study area, the canopy spectra of four types of typical vegetation were measured in situ firstly, including mongolian oak, cotton grass, lespedeza and white birch. Then, on the basis of analyzing the canopy spectral characteristics and their parameterization, the spectral differences of different vegetations were located, and the parameterization method of characteristics identification was determined. Finally, Hyperion data were used to calculate the canopy albedos based on the bidirectional reflectance model of vegetation canopies, and to map the vegetation components in the study area by use of linear spectral mixture model. The results showed that inversion of vegetation components in high vegetation-covered area was accurate using the canopy albedos and liner spectral mixture model, and was identical with the field sampling, which validated the feasibility of canopy albedos and liner spectral mixture model for the inversion of vegetation components. 相似文献
The watershed flow concentration scheme in the distributed hydrology-soil- vegetation model (DHSVM) is coupled with the mesoscale atmospheric model MM5 version 3.5, in which the Oregen States University land surface model (OSULSM) was involved. The flood event which happened in July 2002 in the upper reaches of Heihe river basin is simulated and the surface flow convergence process is shown with this coupled model. It has been concluded that times water head reaches each place of the basin are different. Water amount at each point is split-flow proportionally as the drops in elevation between it and neighbor points. Large part of the water amount pass away in greater slope direction and small part pass away in smaller slope one. Adding of the slope convergence makes the atmospheric model redistributes the surface water laterally.