The Application of Satellite Imagery on Monitoring the Spectral Change of the Vegetation in the Dadu Mountain Area

碩士 === 國立臺灣大學 === 森林學研究所 === 92 === The purpose of this thesis was to find out the fire frequency and climatic characteristics of Dadu Mountain Area by the fire records; accordingly, the grassland was frequently burnt in this area through spatial analysis. For that reason, we investigated fuel loadi...

Full description

Bibliographic Details
Main Authors: Jen-Chien Tseng, 曾仁鍵
Other Authors: Chyi-Rong Chiou
Format: Others
Language:zh-TW
Published: 2004
Online Access:http://ndltd.ncl.edu.tw/handle/41007932080573780946
Description
Summary:碩士 === 國立臺灣大學 === 森林學研究所 === 92 === The purpose of this thesis was to find out the fire frequency and climatic characteristics of Dadu Mountain Area by the fire records; accordingly, the grassland was frequently burnt in this area through spatial analysis. For that reason, we investigated fuel loading in long-term and focused on burnt and un-burnt grasslands to detect the growth tendency of vegetation. In addition, we analyzed the spectrum change of different seasons in the main three vegetation types—Acacia confusa, Panicum maximum Jacq, and Casuarina equisetifolia. Moreover, we implemented the analysis of variance by the independent variables, including red, near-IR, BRI, and NDVI. Based on the analysis of bi-spectral plot, the vegetation spectral change in different seasons by both the NDVI, relation to the spectral angle, and the BRI, relation to the spectral vector length could be explained. Furthermore, the bi-spectral plot variations with some vegetation spectral change could be caused in counter-clockwise direction during different seasons. In summary, the bi-spectral plot could be apply in monitoring the vegetation spectral change. The burnt vegetation was significant different on spectral reflection on the bi-spectral plot and the seasonal change. Finally, there was significant relation between the green cover percentage of fuel and spectral reflection. However, it was unreasonable to represent this relationship using simple regression analysis by the spectral reflection and the curve of fuel loading.