Views on the application of MODIS image in forest mapping and forest type identification

The Moderate-resolution Imaging Spectroradiometer (MODIS) is an important sensor on board the Aqua satellite and the Terra satellite in the Earth Observation System (EOS). In remotely sensed mapping of the complex forest vegetation with low-resolution satellite data as MODIS, how to improve the accuracy of information extraction and enhance the effectiveness of data mining have become aroused general interest in forestry. Based on analyzing the current application status of MODIS data in forest mapping, this paper suggests several spectral bands and products which are used relatively rare at present but valuable for future research, and proposes some critical issues in MODIS forest mapping and forest type identification, namely, mixed pixel decomposition, reconstruction of vegetation index time-series dataset, auxiliary application of biophysical parameters, forest type identification based on the bidirectional reflectance distribution function (BRDF) model and coupling or process imitation of hyperspectral remote sensing data. Holistic and integrated utilization of spectral, temporal, biophysical and angle information has great potential in vegetation mapping.