Padrões de distribuição e fatores condicionantes da riqueza e composição de opiliões na várzea do Rio Amazonas Brasil (arachnida: opiliones)

Abstract

One of the major interests of researchers working with biodiversity in the Amazon basin has been focused on the process that has influenced the diversification of species in this region. However, the process seems to be complex and influenced by different factors that may change among different taxa. The Amazon basin has undergone a series of large modifications for at least 90 millions of years, before present. Geologic, historic and ecological factors have played an important role on the recent biodiversity pattern. Many theories have been proposed and they are all controversial in many opinions. In this study the variation on the harvestmen species composition, richness and abundance through geographic gradient of longitude and e landscape structural variation have been investigated. Landscape variables were measured by the percentage of flooded and dried forests, flooded area, and percentage of fields, upland forest, and the index of land heterogeneity. The patterns of distribution of harvestmen species were evaluated along the geographic gradient, and the distributions probabilities of 7 species in relation of climatic variables, temperature and precipitation, and altitude and hydro stress, the Negro river acting as barrier for species distribution on the floodplains of Solimões-Amazonas were also formally tested. The study area was the Solimões-Amazonas floodplains. Samplings were carried out in a longitudinal transect of 3200 km, from Tabatinga, Amazonas state, and Gurupá, Pará state, in Brazil. Four from eight plots of 150 m2 were fixed for capturing specimens (30 x 5 m), in each plot harvestmen were captured by beating and visual nocturnal search. The harvestmen community was split up in two different biogeographic zones separated by the Negro, which is a breaking zone for species similarity. The structural factors of landscape has influence on the distributional patterns in two different zones, however this variation is largely related with the longitudinal gradient. Longitude was the strongest factor affecting this pattern by indirect and direct effect on landscape variation and on the community. Maximum entropy predictive models corroborated the harvestmen patterns of distribution and suggest the distribution in two different regions for the tested species.