Font Size: a A A

Derivation of hyperspectral signatures for selected vegetation, soil, and water features located within the Longhorn Army Ammunition areas using A VIRIS data

Posted on:1999-11-12Degree:M.S.FType:Thesis
University:Stephen F. Austin State UniversityCandidate:Phillips, Matthew EdwardFull Text:PDF
GTID:2463390014967655Subject:Physical geography
Abstract/Summary:
Until recently, remote sensing data with hyperspectral resolution was not available to research. Acquisition of Airborne Visual/Infrared Imaging Spectrometer (AVIRIS) data taken over the Longhorn Army Ammunition Plant (LHAAP) and Caddo Lake gave an opportunity to observe, with greater spectral resolution, differences between pine and hardwood vegetation as well as soil and water. This study documents spectral profiles of these various coverages and closely analyzes them with the use of second and first derivatives to examine subtle changes in reflectance. LHAAP authorities and researchers from the Arthur Temple College of Forestry at Stephen F. Austin State University provided ground truthing information for the study. This information was in the format of timber cruise data, soil maps, aerial photos, and wetland hardwood community classification maps. Determining the spectral signature for pine in the LHAAP was an important goal. This was not attained due to infrared reflectance from intermixed hardwood species. Though no definite pine spectral signature was established, the hyperspectral data was still able to distinguish between vegetated hardwood and pine. Spectral values derived from "bare" soil plots were not attained due to grass vegetation coverage, however, the spectral profiles of the three plots did show subtle reflectance patterns characteristic of soil. Analysis of open water areas and Bald cypress (Taxodium distichum, L.) stands were made as well.
Keywords/Search Tags:Spectral, Soil, Data, Water, Vegetation
Related items