Using Fish Autecology Matrix &; Artificial Neural Networks to Simulate Instream Fish Habitat Conditions

• Main Authors: Huan-HsuanChang, 張桓旋
• Other Authors: Jian-Ping Suen
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/04496855971947653433

碩士 === 國立成功大學 === 水利及海洋工程學系 === 102 === Recently, more and more people realize the importance of Ecology. For river restoration, ecological engineering projects that providing more suitable habitats for fish community are being designed. To sustain fish population and maintain biodiversity, understanding the relationship between fish community and physical habitat of rivers plays an important role. This study proposes a simplified method to estimate the mesohabitat composition that would favor members of a given set of fish species. Sampling data were collected form HouKu River and WuGouShui River, Taiwan. Using an autecology matrix to identify the critical environmental factors for fish and fuzzy control theory which including depth and velocity as inputs to classify habitats as shallow pool, shallow riffle, deep pool, and deep riffle. Linear regression (LR) and artificial neural networks (ANNs) were used to run the fish habitat models which are based on fish data, abiotic data and an autecology matrix. The result shows that ANNs is an appropriate tool for modeling the relationship between fish and habitat. The models results constitute a reference condition that can be used to guide stream restoration and ecological engineering decisions aimed at maintaining the natural ecological integrity and diversity of rivers.