Discriminating Vegetation and Soils Using Landsat MSS and Thematic Mapper Bands and Band Ratios,
Abstract
Characterizing arid region soils and vegetation conditions from remotely sensed imagery is often limited by the low brightness contrast between soil and vegetation and between different plant communities. The purpose of this study was to determine those Landsat multi-spectral scanner and Thematic Mapper bands and band ratios useful for discriminating soil and vegetation in a semiarid region. Ground level reflectance spectra, 400 to 1100nm, were taken of 35 soil and 127 vegetation samples. The sample's mean brightness values were calculated for the four MSS bands and Thematic Mapper bands 1, 2, 3 and 4. Band:bnad ratios were calculated. The soils were usually brighter than the vegetation in the visible region, although some gravelled soils and some gray colored species had similar alues. In the visible region, brightness values ranging from 15 to 20 separated most vegetation and soils. Many soil and vegetation surfaces had similar near infrared (NIR) brightness values. The NIR:visible band ratios discriminated most soils from the vegetation samples. The soils had NIR:visible ratios less than 2.3, gray and yellow-green colored species had ratios of 2.0 to 7.5 and green vegetation had ratios . 7.5.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 1984
- Accession Number
- ADA140198
Entities
People
- M. B. Satterwhite
Organizations
- Geospatial Research Laboratory