作者: S. H. Peterson , D. A. Stow
DOI: 10.1080/0143116031000082415
关键词: Vegetation type 、 Remote sensing 、 Thematic Mapper 、 Chaparral 、 Soil water 、 Normalized Difference Vegetation Index 、 Environmental science 、 Multispectral pattern recognition 、 Endmember 、 Vegetation
摘要: This research tested the ability of a multiple endmember (EM) spectral mixture analysis (SMA) approach, applied to multi-temporal Landsat Thematic Mapper (TM) data, produce realistic and meaningful EM fractions for study post-fire regrowth in southern California chaparral landscape. Eight different image EMs were used, two types each class interest (green vegetation (GV), non-photosynthetic (NPV), soil, shade); best combination was selected pixel. These validated with derived from 1 m Airborne Data Acquisition Registration multi-spectral data. The datasets similar (r=0.873, 0.776, 0.790 GV, NPV, respectively). Chaparral stands delineated using type, fire history slope aspect GIS layers. Mean calculated stand, variance performed determine if age. Short-...