Temporal-spatial variability in the composition of catch by a stow net installed at Haizhou Bay and its influencing factors
TANG Yanli,CHENG Shasha,MA Shuyang,WANG Xinmeng
DOI: https://doi.org/10.3724/SP.J.1118.2017.16248
2017-01-01
Journal of Fishery Sciences of China
Abstract:The Administrative Department of Fisheries and Researchers has focused almost exclusively on the in-dustrialized fishing sector. Nevertheless, small-scale fisheries account for more than 90% of the world's capture fisheries and fish workers. Thus, complete understanding of the dynamics of small-scale fisheries and exploring the factors that drive fishery activity are urgently needed. Many multivariate methods have been applied to small-scale fishery data in order to distinguish the factors that influence catch composition. This study used mul-tivariate methods to evaluate the spatial variations of catch composition and analyzed the relationships between environmental/fishing factors and catch composition. The survey data from 2013 to 2015 were obtained from 3 stations in Haizhou Bay where stow nets are used. In all, 103 species were captured. We selected 23 species whose weight was more than 1% of all catch species, to analyze the factors that influenced the catch composition. The 23 catch species accounted for 85.79% of the total weight. Cluster analysis (CA) was used to assess the catch species composition, and the CA dendrogram showed two groups, which were significant at the 95% confidence interval (P<0.05). Approximately 15 catch species were included in the first cluster and 8 in the second cluster. According to the results of similarity percentage analysis, Amblychaeturichthys hexanema, Syngnathus acus, and Alpheus heterocarpus were the main contributors to the similarity between the samples of the first cluster, and their simi-larity contribution rates were 26.68%, 16.86%, and 11.01%, respectively. The second cluster was represented by Larimichthys polyactis, Portunus trituberculatus, and Odontamblyopus rubicundus, and their similarity contribu-tion rates were 41.74%, 18.50%, and 17.13%, respectively. Further, the two clusters were distinguished from each other by A. hexanema and L. polyactis, and their variance contributions were 11.87% and 10.28%, respectively. The non-metric multidimensional scaling (NMDS) discriminated stations, years, and months with respect to spe-cies catch composition. Both NMDS plots and one-way analysis of similarities indicated that catch species com-position changed significantly among different stations and different months (P<0.05), but not significantly among different years (P>0.05). The redundancy analysis (RDA) suggested that environmental and fishing factors had significant effects on catch species composition. These variables jointly explained 54.11% of the total variance of the selected species. Among the fishing factors, effective net was the most important factor that explained 67.49%of the total variance, and the daily catch explained 32.51% of the variance. Among the environmental factors, sea surface temperature was the most important factor that explained 27.74% of the total variance, followed by land distance (18.43%), chlorophyll-a content (16.22%), sea surface salinity (13.78%), and wind speed (13.68%); depth only explained 10.15% of the total variance. Effective net was the most important factors driving catch species composition in the first cluster. In contrast, catch composition of the second cluster was influenced by sea surface temperature and wind speed. The partial RDA that included only environmental variables explained 26.40% of the variance, showing that environmental factors explained a significant proportion of the variation in catch species composition. Approximately 8.18% of the variance was explained by the removal of environmental variables. For example, L. polyactis and P. trituberculatus in the second cluster showed a positive relationship with daily catch, and Engraulis japonicus showed a positive relationship with sea surface temperature, whereas A. hexanema had no significant association with any of the variables. Catch composition of stow nets installed at Haizhou Bay varied temporally and spatially and was influenced by the combined effects of a range of environmental and fishing fac-tors. Identifying factors influencing small-scale fisheries could be a critical step to implement marine fishery management measures that are more likely to succeed.