Premium
Simulation of Landsat Multispectral Scanner Response of Soils using Laboratory Reflectance Measurements
Author(s) -
Thompson D. R.,
Pitts D. E.,
Henderson K. E.
Publication year - 1983
Publication title -
soil science society of america journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.836
H-Index - 168
eISSN - 1435-0661
pISSN - 0361-5995
DOI - 10.2136/sssaj1983.03615995004700030030x
Subject(s) - multispectral scanner , soil water , remote sensing , multispectral image , environmental science , multispectral pattern recognition , reflectivity , vegetation (pathology) , soil science , normalized difference vegetation index , spectral bands , geology , climate change , physics , optics , medicine , oceanography , pathology
Laboratory soil reflectance measurements from surface soil samples representing 246 soil series within 39 of the 48 contiguous states of the USA were used to simulate Landsat MSS (multispectral scanner) digital counts for two atmospheric conditions. The resulting Landsat MSS digital counts were within the range of values for soils seen in Landsat data. The importance of accounting for the soil background in various spectral vegetation indexes is demonstrated. Spectral curve forms for different soils are shown to be generally separable within greenness and brightness vector space for the various reflectance values and simulated Landsat MSS values and atmospheres. Organic matter content could be stratified into 0 to 2 and > 2% with > 80% accuracy. The ability to convert reflectance data from controlled experiments to simulated Landsat MSS digital counts will allow researchers to account for the effect of the soil on crop‐spectral relationships, conduct sensitivity analyses of the effect of soil on spectral models, and to develop a better understanding of the relationship of spectral and physical‐chemical properties of soils and advance the application of remote sensing for soil survey and classification.