Correlation Analysis of Topographic Complexity and Fish Abundance around Artificial Reefs – A Case Study of Artificial Reefs at Kaohsiung, Taiwan
| Autor: | Kuan-Chun Huang, 黃冠淳 |
|---|---|
| Rok vydání: | 2017 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 105 Artificial Reefs (ARs) had been deployed more than one hundred sites around Taiwan coast since 1960. One of important designing factors for ARs are suitable site selection and configuration in horizontal and vertical aspects. Topographic complexity is correlated with fish and species abundance according to reports from a lot of papers. The indices used for topographic complexity in this study have (1) surface rugosity, (2) relief, (3) volume, (4) 3D fractal dimension, (5) standard deviation of depths. Some indices from landscape ecology such as: (1) patch density, (2) percentage of landscape, (3) largest patch index, (4) mean patch area, (5) area-weighted average patch fractal dimension calculated by Fragstats software were also explored. In this study, Reson SeaBat 7125 MultiBeam Echo Sounder (MBES) was used to collect water column data around four AR sites at Yungan coast, Kaohsiung, Taiwan, and to analyze the fish abundance by Echoview software. Above mentioned four AR sites have different forms of piling and arrangements, so this study tries to find the correlation between topographic complexity and fish abundance. Water column data were collected four time periods in a day, such as: dawn, noon, dusk, midnight, in each trip last for one or two continuous days. It was found most of midnight water column data were very noisy, and some of the dusk data had similar problem. After removing those noisy data, 14 water column data sets of 2014-2016 were analyzed by Echoview to estimate amount of fishes and their position. It was found site #3 had detected more fishes than rest of the three sites. The water depths surveyed by Reson 7125 Multibeam Echo sounding system for AR sites were gridded into sizes of 10m and 30m. Then we calculated the topographic complexity indices and landscape indices. After integrating amount of fishes and landscape indices and topographic complexity indices information, correlation analysis was performed to analyze which indices are good for topographic complexity and having higher correlation coefficients. There are four conclusions we can make: (1) among landscape indices, percentage of landscape and largest patch indices have significant higher positive correlation with fish abundance. That means the higher the percentage of area or largest patch of an AR site have, the richer fish abundance may be detected, (2) all of the topographic complexity indices have significant positive correlation with fish abundance, but relief, standard deviation of depths and volume have higher correlation coefficients. That means if the ARs were piled up to 2-3 layers or had larger surface area, we may have higher amount of fishes, (3) it was found the detected fishes not only located right above the ARs, they also appeared around the edge of ARs. This phenomenon conforms to the theory of edge effect. The correlation coefficient for grid size of 30m is higher than that of 10m, this may be because 10m is too small to demonstrate the characteristics of fish wandering, (4) it is also found amount of fishes detected in the dawn is higher than in the noon and dusk, and most of them were found at outside of the ARs, We may infer fishes likes to leave AR site in the dawn. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|