Made available in DSpace on 2016-08-10T10:44:20Z (GMT). No. of bitstreams: 1 ROSENI DA SILVA SOUZA.pdf: 1686885 bytes, checksum: 0659d802253a7b64e281c9a3cd03a25e (MD5) Previous issue date: 2012-12-10 / This study aimed to assess the level of concentration of toxic metals in water (Cd, Cu, Mn, Pb and Zn) and fish (Al, Fe, Mn, Pb and Zn) from ten sub-basins of the Upper Tocantins River, Goiás State, Central Brazil, from July to August 2008. Each section of the watercourse chosen (50, 500 and 1000m for streams, creeks and rivers, respectively) were georeferenced and demarcated. In these section were sampled water and fish. The water was collected manually at the beginning and end of each section at 50 cm from the surface, then filtered, packaged in bottles and stored at 4 °C. Samples of whole fish or muscle pieces were defrosted, lyophilized, digested and stored in polyethylene bottles previously washed. The determination of the concentration of toxic metals from water samples was performed using the technique of optical emission spectrometry with inductively coupled plasma (ICPOES), and this fish was made using the technique of atomic absorption spectrometry method (flame-AAS). After determination of toxic metals in fish muscles, these were classified by trophic guild (omnivorous, invertivorous, detritivorous, piscivorous, insectivorous and herbivorous) through available literature. The Kruskal-Wallis analysis showed no significant differences between the concentrations of toxic metals in water between sub-basins. The comparison of the concentrations of toxic metals present in fish sampled indicated that there were significant differences between the trophic guilds considered for Al (p = 0.0057), Fe (p = 0.0000), Mn (p = 0.0000 ), Pb (p = 0.0001) and Zn (p = 0.0001). The detritivorous had higher concentration of Fe compared to omnivores, and higher concentrations of Al, Fe, Mn, Pb and Zn in relation to invertivores. The invertivores showed a higher concentration of Mn, Pb, Zn than omnivores. The insectivores have excelled in the concentration of Mn in relation to invertivores and Pb in relation to detritivores. In the case of piscivorous the highest concentration was of Fe and Mn in relation to Mn and detritivorous in relation to omnivores. The correlation analysis between the concentrations of Mn, Pb and Zn indicate that there is no correlation between the water and the fish, whereas positive correlations were observed between the concentrations of Pb and Mn (r = 0. 682834); Zn and Mn (r = 0.716368) and Zn and Pb (r = 0.770089) in sampled fish. It is concluded that the water did not influence directly the contamination of fish by toxic metals, but fish seem to have been contaminated by food, expressed in this study by the trophic guilds. / Este trabalho teve como objetivo avaliar o nível de concentração de metais tóxicos na água (Cd, Cu, Mn, Pb e Zn) e em peixes (Al, Fe, Mn, Pb e Zn) pertencentes a dez sub-bacias do alto rio Tocantins, Goiás, Brasil Central, no período de julho a agosto de 2008. Cada trecho do curso de água escolhido (50, 500 e 1000m para córregos, ribeirões e rios, respectivamente) foi georreferenciado e demarcado. Nesses trechos foram realizadas as coletas de água e peixes. A água foi coletada manualmente no início e no final de cada trecho a 50 cm da superfície, em seguida foi filtrada, acondicionada em frascos e armazenada a 4°C. As amostras de peixes inteiros ou pedaços de músculo foram descongelados e deixados em repouso até atingirem a temperatura ambiente, em seguida foram liofilizados e posteriormente digeridos e estocados em frascos de polietileno previamente lavados. A determinação da concentração dos metais tóxicos das amostras de água foi feita por meio da técnica de espectrometria de emissão óptica com plasma indutivamente acoplado (ICP-OES), e esta dos peixes foi feito por meio da técnica de espectrometria de absorção atômica modalidade chama (EAA-chama). Após a determinação dos metais tóxicos nos peixes, estes foram classificados por guilda trófica (onívoro, invertívoro, detritívoro, piscívoro, insetívoro e herbívoro) através da literatura. A análise de Kruskal-Wallis indicou que não houve diferenças significativas entre as concentrações de metais tóxicos na água entre as sub-bacias. Já na comparação das concentrações de metais tóxicos existentes nos peixes amostrados verificou-se que houve diferenças significativas entre as guildas tróficas consideradas para Al (p=0,0057); Fe (p= 0,0000); Mn (p=0,0000); Pb (p=0,0001) e Zn (p=0,0001). A guilda dos detritívoros tiveram maior concentração de Fe em relação aos onívoros; e maior concentração de Al, Fe, Mn, Pb e Zn em relação aos invertívoros. Os invertívoros apresentou uma maior concentração de Mn, Pb, Zn que os onívoros. Os insetívoros se destacaram na concentração de Mn em relação aos invertívoros e de Pb em relação aos detritívoros. No caso dos piscívoros a maior concentração foi de Fe e Mn em relação aos detritívoros e Mn em relação aos onívoros. A análise de correlação entre as concentrações dos metais tóxicos Mn, Pb e Zn indicam que não há correlação entre as concentrações dos três metais na água com os peixes, enquanto que são observadas correlações positivas significativas entre as concentrações de Pb e Mn (r=0,682834); Zn e Mn (r=0,716368) e Zn e Pb (r=0,770089) entre os peixes. Neste estudo conclui-se que a água não influenciou diretamente na contaminação dos peixes, os quais parecem terem sido contaminados via alimentação, expressa neste estudo pelas guildas tróficas.

Healthy coral reefs have become increasingly rare, and their continuous degradation has serious implications for loss of marine biodiversity. There is an urgent need to assess the strength of top-down versus bottom-up effects on reef communities, to better understand how food web alterations can change the structure and function of these vulnerable marine systems. In this study, I used fatty acid (FA) analysis to investigate the trophic and ecological relationships among potential key forage species of the critically endangered monk seal in the Hawaiian archipelago. A series of multivariate tests performed on groups of closely related and ecologically equivalent species of fishes and invertebrates using a restricted number of FAs revealed that FA differences among groups primarily reflected diet, but could also be related to habitat and ecology. The same groups were subsequently analysed using an alternate method in quantitative FA signature analysis (QFASA) simulations, which allowed for the effects of using various subsets of FAs to be evaluated. Overall, species groups were relatively well characterized using both methods. When present, overlap in FA composition principally occurred among groups with similar diet/ecology, and were more prominent at higher trophic levels. A last set of analyses which combined the multivariate and QFASA simulation methods revealed that despite taxonomical relatedness and similarities in trophic ecology, individual species of carnivorous fish could be reliably distinguished using FAs. Therefore, while increasing the number of FAs used in the analyses might be useful to refine the resolution of distinctions, using a restricted number of FAs can also result in reliable differentiation among species. My results suggested that despite tremendous diversity, finer scale variations in FA composition could be detected among groups, and among species which shared the same diet and trophic ecology. These findings have important implications for the study of food web interactions in the Hawaiian archipelago, as they provide the foundation for using the same species groups in diets estimations of monk seal, as well as other top predators in this ecosystem. Moreover, they provide a framework for using multiple approaches to link FA patterns to the foraging ecology of individual species.