Studies On Fish Food Web Structures And Energy Flow Of Daning River In The Three Gorges Reservoir Areas

Quantifying food web structures and dynamics is a fundamental ecological research for understanding ecosystem structure and functioning, with implications for future management and rehabilitation of rivers. Daning River is one important tributary of the Yangtze River in the middle region of the Three Gorges Reservoir (TGR). After impoundment of the TGR, the river had been slowed down and deepened in the downstream. A new river/reservoir ecosystem had been formed in the Daning River. The main objectives of the present study were to elucidate the contributions of carbon sources and trophic relationships among fish species within fish food web, and then further construct the food web structure and energy flow pathways in the Daning River based on survey data from 2011 to 2014. Stable isotopes analysis and gut content analysis as well as Ecopath model were utilized to depict the food web characteristics and energy flow in Daning River ecosystems. The main results were summarized as follows:(1) Natural abundances of stable carbon (δ13C) and nitrogen (δ15N) isotopes of aquatic consumers and potential organic carbon sources were measured to quantify patterns of carbon flow through aquatic food webs using the mixing IsoSource model in the Daning River. The results revealed that primary producers (phytoplankton, benthic algae, terrestrial C3, C4 plants) ranged from -33.41‰ to -10.44‰ in δ13C values and from -0.39 to 11.99‰ in 815N values. These four basic producers showed distinct δ13C signatures, which allowed for a powerful discrimination of carbon sources. Carbon derived from autochthonous algae and C3 riparian plants were the dominant energy sources supporting consumers across the river. However, relative contributions differed among reaches, with phytoplankton the most important basal carbon sources contributing to consumer biomass in the downstream and C3 riparian plants greater in the upstream. This distinction was likely associated with hydrogeomorphic features changes along the river. As newly-impounded zone, the downstream had a higher productivity and biomass of pelagic primary producers as a result of greater water clarity, residence times and depth, and plants on large area of hydro-fluctuation belt as potential source materials consumed by aquatic organisms. However, the terrestrial C3 plants were abundant in riparian regions and were not affected by impoundment of the reservoir in the upper reaches and, in consequence consumers were relying more on these nutrients.(2) Stable nitrogen isotope ratios of the main fish species were measured for the lotic and impounded reaches in Daning River respectively to identify each trophic level. Mean trophic position and six community metrics in the isotopic niche of several conspecifics from lentic waters were compared during different hydrological periods. The results showed that the δ15N values of species in the lentic downstream ranged from 4.54%‰ to 17.51‰, whilst the trophic levels of these organisms were 1.51～3.88 with the average of 2.49. In the upstream,δ15N values of species varied from 2.25‰ to 10.81‰. However, the trophic levels were in the range of 1.49～4.01 with the average of 2.87. The higher average trophic levels of fish species from the upstream than downstream suggested that fish tended to consume preferable invertebrate in the lotic areas due to their large diversity and abundant biomass. The mean trophic levels of conspecifics during the flood period in the impounded waters were significantly declined with the comparison of the non-flood period, inferring that hydrological disturbance could influence the availability of intermediate predators on the one hand, and on the other increased food competition during the low-water period could have forced the fish downstream forage on sufficient prey food which were at lower trophic levels. The trophic structure of fish communities assembled by the same species at each period exhibited little variations on account of great food abundance in the Daning River.(3) Trophic relationships among the fishes were also confirmed by use of analysis of gut contents to further reveal trophic patterns. Trophic positions of these fish were also calculated and compared using dietary analysis and stable nitrogen isotope ratios. The result showed that four feeding guilds including planktivores, herbivores, omnivores and piscivores were classified by cluster analysis according to percent weight of diet items for 20 fish species. There was a moderate difference in ecological specialization among these species, and Levin’s standardized niche breadths (Ba) ranged from 0.003 to 0.371, with a mean of 0.230. Although a wide range of food resources were identified, the lower values of Ba indicates that the number of important of food items for each fish was small. The Pianka’s dietary overlap index among most fishes were also very low with only 7.89% of combinations higher than 0.6, which manifested that there were no intense competitions and food was unlimited in the river. The diversity of food resources and habitat could be the main explanation of lower competitions. It appeared that the trophic levels estimation based on diet analysis were significant higher than that calculated from stable isotope values for most fishes. Discrepancies between dietary and isotopic estimates of trophic positions were likely due to different trophic baseline of each method and biases associated with individual size.(4) Hydroacoustic surveys were conducted in October and November,2011 (at the high-water level period) and March,2014 (day and night acoustic surveys) to study the temporal and spatial distribution of fish in the impounded section of Daning River using a split-beam Biosonics DT-X echo-sounder (208 KHz). Fish densities of the entire water column were estimated. The results suggested that the target strength (TS) of the fish varied significantly between two months with (-62.97±5.61) dB in October and (-63.94±5.49) dB in November. The average water depth of fish target in November was significant deeper than that in October and fish tended to move to bottom layers in November. No significant difference was observed in mean fish density in both month, which averaged 5.70 ind./1000m3 in October and 5.43 ind./1000m3 in November, but fish density particularly in valley area of the impounded reaches varied significantly between two months. For different types of survey subarea, a higher fish density occurred in open unconstrained water zone where the average was 7.52 ind./1000m3 than in constrained zone where the average was 2.70 ind./1000m3. There were no significant differences in fish density between day and night surveys. However, the vertical position of fish target was detected lower during daytime, suggesting that there could be diel vertical migration of the fish. We discussed that food distribution rather than water temperature was the most important ecological factors which affected the distribution of fish in different type water zone, and the behavior of fish in diel migration depended on trade-off between feeding opportunity and predation risk as well.(5) In order to reveal trophic interactions and energy flow pathways of the newly-impounded area of Daning River, an Ecopath model based on the 2011-2012 survey data of fishery resources and eco-environment was constructed to describe the downstream ecosystem. The effective trophic levels of 20 functional groups varied between 1.0 for detritus and primary producer group and 3.3 for large catfish group, and all of the groups were aggregated in a food chain of six discrete trophic levels with the trophic flows primarily occurring through Ⅰ to Ⅳ trophic levels. The transfer efficiencies were low varying between 1.7% and 22.1%, with the average 5.7%. The proportion of total energy flow originating from detritus was 45.87% while the other 54.13% came from primary producers. Total system throughput of the ecosystem was found to be 35239.94 t/(km2·a), and total net primary production(TPP), total respiration(TR), total biomass(B) was 14544.35 t/(km2·a),1088.78 t/(km2·a),130.35 t /(km2·a), respectively. Overall ecosystem property indices showed that this reservoir had a relative higher value of TPP/TR and TPP/B, together with low connectance index (CI), system omnivory index (SOI), Finn’s cycled index and Finn’s mean path length, which indicating that Daning reservoir ecosystem was an immature ecosystem and still at its development stage. These results would provide guidline for fisheries policy development and ecological restoration of the Daning River.