LARS Tech Report Number
062283
Abstract
The primary aim of an analysis of Landsat data is to obtain as much useful and reliable information as possible about the various earth-surface materials present in a scene. To assist the analyst in accomplishing this objective, quantitative data analysis techniques have been developed and successfully applied in computer-aided mapping, inventorying, and monitoring of earth resources. To increase the effectiveness of these techniques, the analyst must have a thorough understanding of the spectral behavior of earth-surface features, the radiometric characteristics of the digital Landsat MSS data, and the fundamental concepts supporting each numerical analysis function used.
The purpose of this paper is to show how calibration of Landsat data can be a useful analysis tool, particularly when adequate reference data are not available. In the approach described, the Landsat digital counts, along with the minimum and maximum radiance values from the internal calibration lamp, are used to derive calibrated in-band radiance values which, when plotted, are similar in shape to spectral reflectance curves, such as those obtained by spectroradiometers. Included in this paper for comparison is a series of examples of calibrated Landsat mean plots for a variety of earth-surface materials over different geographic locations at different time periods. Data used includes examples from all four Landsat MSS systems.
Date of this Version
January 1983