Abstract
An attempt was made to classify plant cover from a datafile developed by the Japan Environment Agency. This datafile has two kinds of information, codes of plant communities and the degree of influences of human activity on primeval vegetation in the center of a cell. A lookup-table for plant communities and corresponding plant cover types was constructed after alpha-numeric conversion of the plant community code and sorting. These plant cover codes in the table were applied to plant community codes in the above file, and a plant cover datafile was developed. To overlay the plant cover on other elements, precise spatial coordination of data sampling points was examined using two datafiles with different cell sizes, such as a 1km interval for the plant cover and a 250m grid system for a digital terrain model (DEM) in which altitudes above sea level were sampled at 250m intervals. This study confirmed that the plant cover derived from the above datafile is sufficiently useful for ground truth in satellite remote sensing, and for careful overlaying of other environmental elements.
