Spruce stand age estimation based on the successional spectral trajectory

Awaya Y.

doi:10.11519/jjfs1953.79.1_22

A forest age classification method, which was based on the minimum distance classifier using training data derived from age-spectral curves, was proposed selecting a test site, which included spruce stands (Picea jezoensis Carr., Picea glehnii Mast.), in Tomakomai, Hokkaido. This method made it possible to classify images almost free of the effects of seasonal vegetation changes and the interpreters' subjectivity. Classification results of Landsat Thematic Mapper images, which were taken in March 1985, August 1985, March 1993 and July 1993, using the method showed the following features. 1) Correlation coefficients between the classified images were high except the cases of using the image in March 1993, and thus very similar results were derived from images, which were observed under very different vegetation conditions in winter and summer. 2) The regression analysis was applied to any two images to check their relationship. The intercepts of regression lines on Y axis were close to the expected values from the elapsed time between two images, except the cases of using the image in March 1993. 3) The regression slope between classification results of March 1985 or August 1983 and July 1993 was less than 1. The smaller annual spectral changes in mature stands than those in younger stands would caused it.

Spruce stand age estimation based on the successional spectral trajectory

Bibliographic Information

Search this article

Description

Journal

Citations (2)*help

References(17)*help

Keywords

Details 詳細情報について

Export

Report a problem