Forestland Classification for The Experimental Forest of National Taiwan University

• Main Authors: CHANG-CHING LIN, 林長青
• Other Authors: Yeong-Kuan Chen
• Format: Others
• Language: zh-TW
• Published: 2002
• Online Access: http://ndltd.ncl.edu.tw/handle/65033443620786358690

碩士 === 國立臺灣大學 === 森林學研究所 === 90 === Forestland classification could be the regarded as the foundation of forest resource management of forestry ecosystem management. The purpose of this study is to establish a multi-scale forestland classification decision support system and further to provide Nation Taiwan University Experimental Forest Management Department a multi-scale forestland classification following ecology principles, under the integration of a hierarchical forestland classification. The Ecosystem Management Decision Support (EMDS) system, which was developed by the Forest Service, USDA was applied as a tool in this study. To conform with ecological environment characters, this study use watershed generated by Digital Terrain Model(DTM) as the classification units, and combine with selected variations and soil data to establish forestland classification database and to build a knowledge base for the forestry management. Multivariate analysis is applied for the establishment of multi-scale hierarchical structure. Knowledge base follows fuzzy logic approach and hypothesize classification standard to judge the classification criteria. It does not only conform to ecological characters, but also could amend the knowledge base according to different environments and demands to ecosystem environment variation features. The multi-scale forestland classification decision support system of NTU Experimental Forest established by this study could proceed to forestland suitability analysis and forestland classification of NTU Experimental Forest. This is a practicable model could be adjusted and revised by arbitrarily demanding, and could evaluate forestland suitability by different scale. Therefore, this model could combine forestland classification with different manage level models to demand of different hierarchical manages and different management purposes.