Application of Analytic Network Process onto Land Use Suitability Assessment

• Main Authors: Yu-Jyun Lai, 賴俞均
• Other Authors: Chen-Fa Wu
• Format: Others
• Language: zh-TW
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/7yzjb6

碩士 === 國立中興大學 === 園藝學系所 === 106 === The Analytical Network Procedure (ANP) has been widely used in the field of sociology and management science; however, it has not been fully applied in studying land use change. This thesis attempts to use the ANP method to identify and evaluate the driving factors of land use change and to determine appropriate management strategies. First, the impact factors of land use change in central Taiwan were identified by literature review and expert interviews. Based on this, a draft ANP expert questionnaire for evaluating land use management was developed. The formal version questionnaire was finalized after pilot studies and revision. A total of 14 scholars were selected as interviewees. The background and expertise of them includes urban planning, landscape change, land planning and design, and applications of geographic information system, etc. After the first-round survey, the consistency of answers would be checked. The questionnaire with inconsistent responses would be sent for a second-round review in order to performing revision.. Totalvalid questionnaires were 10 samples. The Super Decision analysis software was applied to analyze data collected by questionnaires. The results showed that among the 11 driving factors of land use change, the main factors affecting the growing patterns of rice paddy field include soil organic matter (0.16521), land subsidence (0.14465) and soil fertility (0.14452). The location of dryland is mainly affected by soil organic matter (0.17099), land subsidence (0.14317) and distance to river or irrigation system (0.12822). The main influence on the distribution of forest land is the slope (0.16764), elevation (0.15237), and distance from the river irrigation channel (0.15003). How the water distributes could be influence by following factors, which are distance to river or irrigation system (0.18800), land subsidence (0.15515) and soil organic matter (0.15031). Vacant land is commonly affected by three factors: land subsidence (0.16931), slope (0.14006), and distance to urban planning area (0.10880). Regarding to the proper management strategies for the six land use types, 10 experts highly recommended following strategies, which are: (1) it is necessary to preserve prime farmlands, prime and fine dry fields; (2) development of newly buildings should be restricted no matter in urbans, industrial zones, and rural areas; (3) Land use strategies such as conserving native forests and national forest lands should be promoted; (4) Also, aqua culture area should avoid future expansion. Results of applying weights coefficients of different drivers onto the suitability assessment of six land use types showed that the suitability ditribution for paddy field (95.45%), dry land (96.46%) and vacant land (99.02%) were the highest, followed by land construction (89.79%) and forest land (81.41%). Water (46.51%) displayed the lowest suitability. Finally, the overall land use distribution simulation was carried out. The results of the comparison with the 2015 land use distribution map showed that the simulation of paddy field, dry land, forest land and idle land was higher, and the overall accuracy was 74.31%.