Prawn larvae of Macrobrachium amazonicum (Decapoda, Palaemonidae) in contaminated oligohaline creeks on the Amazon estuary

The authors thank the National Council for Scientific and Technologic Development (CNPq) for their financial support (Project N. 481157/2010-7); the Institutional Program for Graduates? Scholarship of Federal University of Par? (PIBIC/UFPA) for the scholarship granted to the first author; members of the Research Group on Amazonian Crustacean Ecology (GPECA); and all colleagues who helped during field campaigns, with special thanks to Bianca Bentes (IECOS-UFPA), Simone de F?tima Pinheiro Pereira (UFPA), the Bureau of Environment (DEMA) staff for the safe working conditions provided in the field, and the boatman, Mr. Parau, for his availability. Universidade Federal do Par?. N?cleo de Ecologia Aqu?tica e Pesca da Amaz?nia. Grupo de Pesquisa em Ecologia de Crust?ceos da Amaz?nia. Bel?m, PA, Brasil. Universidade Federal do Par?. N?cleo de Ecologia Aqu?tica e Pesca da Amaz?nia. Grupo de Pesquisa em Ecologia de Crust?ceos da Amaz?nia. Bel?m, PA, Brasil. Universidade Federal do Par?. Instituto de Ci?ncias Biol?gicas. Bel?m, PA, Brasil. Minist?rio da Sa?de. Secretaria de Vigil?ncia em Sa?de. Instituto Evandro Chagas. Ananindeua, PA, Brasil. Universidade Federal do Par?. N?cleo de Ecologia Aqu?tica e Pesca da Amaz?nia. Grupo de Pesquisa em Ecologia de Crust?ceos da Amaz?nia. Bel?m, PA, Brasil. The objective of this study was to assess larval stage densities of Macrobrachium amazonicum in two highly contaminated amazonian creeks (Tucunduba and Mata Fome) and in one creek less impacted by human activities (Combu). The hypothesis tested is that larvae are more abundant in protected areas, independently of their development stage. Zooplankton was collected once every 3 months by conducting horizontal hauls at the sub-surface of the water column using a 300 ?m mesh size plankton net during flood and ebb tides. Temperature, pH, conductivity, salinity, dissolved oxygen, water hardness, biochemical oxygen demand, chemical oxygen demand, Cl, N-NO, NO, NH3, NH4, PO4, SO4, Fe, Mg, Li, K and Na values or concentrations were estimated for each creek in order to determine whether these factors influence larval stage densities. Zoea stages I, V, VIII, X and XI were recorded: zoea stage I was present all along Combu creek and downstream of Tucunduba and Mata Fome creeks; the other stages were encountered only in the uncontaminated creek. Larval stage zoea XI was considered ?occasional?. Densities were statistically different between months. It was concluded that spawning ground of M. amazonicum was close to the three creeks because the first larval stage was found indiscriminately at all downstream sampling stations. The more advanced stages (zoea V or higher) were found mid-downstream but only in the most preserved creek (Combu); no other larval stage beside zoea I was encountered in the creeks with greater anthropic influence. Further, not only is the adult population affected by water contamination, but also the occurrence of larval stages is associated with more protected waters. The high concentration of coliforms was associated with nitrates, and the low larval density of the Amazonian prawn was a clear response to this contamination of the water.