Impactos espaciais e temporais do alagamento do reservatório de uma mega hidrelétrica no sudoeste da Amazônia brasileira sobre a diversidade taxonômica e funcional de morcegos Phyllostomidae

Abstract

The Brazilian Amazon has approximately 90% (~ 95 MW) of unexploited energy potential and government agencies have planned studies for the construction of 254 new hydroelectric dams in Amazon region in the near future. Although hydroelectric power stations have been promoted as cleaner sources of energy by most governmental agencies, river damming interrupts the continuity of the river and causes the flooding of adjacent riverside and terrestrial ecosystems. The flooding of the lowlands causes direct disturbance in the abundance and species composition, in particular those species directly associated to the flooded gradients. Most of the data currently available about the effects of the hydroelectric dams are like static pictures of certain post flooding stages resulting from short-term projects often conducted years after the reservoir is filled. For a better understanding of the environmental impacts of the hydroelectric dams on wild species, it is advisable to evaluate impacts before and after the reservoir floods. Bats are excellent indicators of the health of ecosystems, particularly the Phyllostomidae bats, which are highly diverse both in species richness and in ecology. Habitat loss from human interference has repeatedly been identified as the main threat to the survival of many bat species, but the effects of the establishment and operation of hydroelectric power stations on bat assemblages are still poorly understood. The general objective of this thesis was to evaluate the spatial and temporal effects of the flooding of the Santo Antônio Hydroelectric Plant reservoir, upper River Madeira, State of Rondônia, integrating bat data collected in areas that would be flooded and not flooded before of the dam construction with data collected in non-flooded areas after the river damming. The thesis is composed by three chapters that evaluate different aspects of the temporal and spatial effects of the dam filling on the taxonomic and functional composition of bats. In chapter 1, a total of 49 plots distributed along the areas going to be inundated by the dam and those remaining dry. As predictors for the species distribution, we tested the variables of vegetation structure and topography. A total of 2306 bats belonging to 58 species were captured. Species composition largely differed between the dry plots and the plots located in areas that will be flooded, and this was strongly associated with the variables of forest basal area and terrain elevation. Vegetation-related variables also had strong influence on the guild distribution. The flooding of lower elevations areas is expected to negatively affect the species number and abundance of frugivorous species. In contrast, it is likely that animalivores will be less vulnerable to dam-induced flooding, since they were abundant in the areas not expect to be inundated. In chapter 2, I sampled 12 plots that were going to be flooded (pre-flooded) after the establishing of the dam, and 34 plots that were to remain dry (pre-unflooded). Of the plots that remained dry, 25 of them were resampled after the river damming (post-dam). A total of 3096 bats belonging to 59 species were captured. Capture rates and number of species were similar among pre-flooded, pre-unflooded, and post-dam plots. On the other hand, the composition of species and frugivorous bats differed between the three plot types, with a clear discrimination of the várzea plots of the pre-flooded period. The comparison between pre- and post-dam dry plots showed that 60% of plots decreased in the abundance and 68% of the plots decreased in abundance of frugivorous bats. In chapter 3, the same plots as in chapter 2 were sampled. The flooding of the lowlands increased significantly the taxonomic and functional α-diversity and functional uniqueness of the bat assemblages. I found strong evidence for a shift in functional traits after river damming, with an increase in the frequency of animalivorous bats and a reduction in phytophagues. The results of functional β-diversity show that β total results from functional β replacement - substitution of traits weighted by the species abundance - and not by the loss or gain of functional traits. The results of this thesis showed that the effects of the Madeira River damming by the Santo Antônio Hydroelectric Power Plant were rapid, observed in the first two years after the flooding of the reservoir. Bat assemblages of the várzea forest were the most impacted by the formation of the reservoir and they were not represented among the remnant unflooded areas. The implementation of the reservoir did not reduce the number of bat species, but frugivorous bats responded negatively with a general abundance loss after flooding of the reservoir. As new hydroelectric dams are planned for the Amazon, I recommend the adoption of legislation aimed at hydroelectric dams that incorporate the creation of conservation units with special attention to the vegetation types permanently suppressed. This conservation management strategy must be financed by energy producing companies.