COS 1-10
Rapid responses of fish size spectrum to environmental changes in a recently formed Neotropical reservoir

Monday, August 10, 2015: 4:40 PM
301, Baltimore Convention Center
Mateus Ferrareze, Zoologia e Botânica, UNESP, São José do Rio Preto, Brazil
Henrique C. Giacomini Sr., Ecology & Evolutionary Biology, University of Toronto, Toronto, ON, Canada
Donald A. Jackson, Ecology & Evolutionary Biology, University of Toronto, Toronto, ON, Canada
Brian J. Shuter, Ecology & Evolutionary Biology, University of Toronto, Toronto, ON, Canada
Background/Question/Methods

Large dams are known to have a major impact on fish communities of rivers and can drastically change the biological diversity. While the effects of large dams on the environment merit highlighting, most developments around the world are for small size undertakings. Size spectrum approaches have increasingly been applied as indicators of the aquatic ecosystem structure, and the use of size distribution models may give us a better understanding about how these developments affect the aquatic communities. The aim of this study is to assess the responses of the fish size spectrum to a longitudinal gradient in a small size reservoir in Brazil during the first and second year after dam closing. To reach this goal, eight samplings were carried out quarterly. Samples were obtained in three different regions of the reservoir (lotic, intermediate, and lentic). Large fish were collected with gill nets, whereas small fish were sampled with trawl nets. Concomitantly to the fish sampling, data of 16 limnological variables were taken. To test the hypotheses and whether there is any association between the abiotic components with the patterns obtained, a size spectrum gradient estimated by maximum likelihood estimator (EMV) of the Type 1 Pareto distribution was used.

Results/Conclusions

The size spectrum slope, given by the maximum likelihood estimator of the Pareto type I distribution, vary in a systematic manner across zones and years. For the first sampling year, there is a trend of decreasing slope from the lotic zone to downriver, implying a longitudinal gradient in the relative contribution of smaller fish. The second year is characterized by shallower size spectra in all zones, although the difference is magnified at more impacted zones, implying a reduced relative contribution of smaller fish there. Longitudinally, the size spectrum slope did not change in the lotic zone, where the riverine condition is better preserved. The lentic zone, representing the more altered habitat, had a much steeper size distribution in the first year, implying in relatively higher dominance by small fish, but then showed the greatest change in slope in the second year towards the values typical for the lotic zone. The transition zone showed intermediate trends in slope. These results can be explained by concomitant changes in phytoplankton production, water transparency and occurrence of predator fish observed between the two years, suggesting that the fish size distribution can readily track changes in environmental conditions following major perturbations.