Report of the Spatial Analysis of Part of Fort Benning Using Remote Imagery. Optimizing Field Sample Design from Remote Imagery.
Abstract
This final report summarises the three earlier ones and a detailed description of the second major analysis. The earlier reports included a literature review, a description of Fort Benning, and the exploratory data and variogram analyses of the spectral information from the study area (a small part of the Fort). The results showed that channels 2 and 3 are highly correlated, but neither is with channel 1. The variograms for the rows and columns of the spectral data and their averages for all three channels and NDVI (normalized difference vegetation index) show the same spatial I structure. There are two spatial scales, the short one is 9 pixels (180 m) and the long one 144 pixels (2880 m). A transect sampling was designed for reconnaissance based on the short range variation to determine ground cover. The analysis of these data showed that the structure in the variation of the image reflected that in the ground cover closely. A second sampling scheme based on a nested grid with intervals of 1500 m and. 75 m was designed for fieldwork. Our major conclusion at this stage is that it appears reasonable to design a sampling scheme for fieldwork. directly from image in the data without the need for the reconnaissance stage.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 01, 1995
- Accession Number
- ADA305642
Entities
People
- Margaret A. Oliver
- Richard Webster
Organizations
- University of Reading