碩士 / 國立臺灣大學 / 海洋研究所 / 101 / The objectives of this research were: (1) to evaluate short-term larval fish variations and discuss the sampling representatives, through high frequency sampling; (2) to discuss sampling relations between species parameters (species richness and density) and the sampling cost. Larval fish were collected from 3 stations in costal waters of Tamsui river estuary, Northern Taiwan, from June to July, 2012, depending on different sampling frequencies (2~43 days intervals, and 2~12 hours intervals crossing day and night) by using the round mouth ichthyoplankton net (90 cm in diameter and 330 μm in mesh size). Larval fish were identified by morphology and mitochondrial DNA COI sequence barcoding. We evaluated larval fish assemblage variations under set factors including station, sampling date, sampling hour, day/ night and tidal phase, and taxa that contributed to assemblage variations most by using ANOSIM and SIMPER analysis, respectively. Total of 5302 larval fish were collected. There were 189 taxa belonging to 49 famalies, 42 genera and 23 species by morphological identifications, but there were 131 taxa belonging to 52 families, 91 gerera and 78 species with the aid of DNA barcoding, indicating that larval fish identification ability could elevate by using DNA barcoding. No significant differences in larval fish assemblages were found among stations, because of the short distance between each and the same substrate. There were significant larval fish assemblage differences among sampling dates, mainly resulted from different recruitment patterns of larval fish species. Significant differences in larval fish assemblages were found in different sampling hours and day/ night factors, but chiefly resulted from the latter, comparisons between sampling hours belonging to daytime or nighttime. The main typifying species in daytime group were Sillago sp. 2, Alepes djedaba and Ambassius sp., and the main typifying species in nighttime group were Engraulidae sp. 1, Secutor ruconius and Engraulidae sp. 2. There was significant difference in larval fish assemblages between flood tide and ebb tide only on 9, July, but the statistical significance mainly came from high larval fish density for each species during flood tide. The subsampling results showed that the cumulative curve of species richness percentage reached an asymptote with the elevated sampling number. Moreover, the higher the larval fish density was, the lower the number of hauls needed.
The results suggested that, to elevate sampling representatives and correctly reflect larval fish assemblages in the Tamsui river estuary with the most economical sampling effort invested in the future larval fish sampling designs, we must firstly cover the day-night interval, and secondarily increase daily- and hourly- sampling frequencies, decrease the number of sampling stations as well as sample during flood tide. Besides, it is better that we set a target species richness percentage in each single sampling time according to the sampling effort we could afford.
| Identifer | oai:union.ndltd.org:TW/101NTU05279027 |
| Date | January 2013 |
| Creators | Yu-Ming Shih, 石淯銘 |
| Contributors | Kwang-Tsao Shao, Chi-Lu Sun, 邵廣昭, 孫志陸 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 95 |
Page generated in 0.0016 seconds