基于刺网的淀山湖鱼类群落结构时空变化及网目大小对其产生的影响

Our study aimed to analyze the spatial and monthly differences in fish community structure of Dianshan Lake. We explored the effect of monthly mean air temperature (MMAT) on fish species number. The effect of mesh size on species composition, size structure, and catch per unit effort (CPUE) were also investigated. From 2009 to 2010, using multi-mesh monolayer gillnets, monthly sampling of fishes was conducted at six stations. Using PRIMER 5.0, cluster analysis (CA), and non-metric multidimensional scaling (NMDS), spatial and monthly differences in the fish composition of Dianshan Lake and the effect of mesh size on species composition and size structure of fish assemblages were analyzed. Two-way analysis of variance was used to test for seasonal and spatial differences in CPUE. The best models for describing relationships between species number and MMAT, and between CPUE and mesh size were selected. A total of 28 fish species were collected. They belonged to six orders and eight families. The dominant species were Coilia nasus and Acheilognathus taenianalis. There were no significant spatial differences in fish species composition and CPUE, but these parameters differed significantly between seasons. The fish species composition in January was significantly different from that in other months. The number per unit effort (NPUE) in the winter was significantly lower than those in the summer and autumn. The biomass per unit effort (BPUE) in winter was significantly lower than those in the other three seasons. The species number caught per month varied as a quadratic polynomial function of MMAT. Species composition and size structure of catches were significantly influenced by mesh size. The catches collected by nets with a mesh size of 2, 4, 6, and 8 cm, 10 cm, and 12 cm clustered into three assemblages, respectively. They were separated by indicator species like A. taenianalis, C. nasus, Chanodichthys erythropterus, Cyprinus carpio, and Chanodichthys erythropterus. The size structure differed significantly between catches sampled with 2-10 cm mesh-size nets and those sampled with 12 cm mesh-size nets. They were divided by indicator body length class of 10-15 cm and 40-45 cm. With the increasing mesh size, cumulative species number tended to decrease and NPUE declined exponentially. However, BPUE showed no significant correlation with mesh size. In order to make a more reliable estimation of the fish community structure of the lakes, we recommend the addition of nets with mesh size less than 2 cm into the multi-mesh monolayer gillnet and combined utilization with other fishing gears.