Avaliação de condições ecológicas em riachos no cerrado: bases para sua conservação

Abstract

Freshwater ecosystems are among the most threatened by anthropogenic activities on a global scale. The high demand for water has resulted in an ongoing process of degradation leading to the loss of biodiversity in these ecosystem s. In the Brazilian savanna the intense expansion of agribusiness results in deforestation, habitat loss, introduction of exotic species, soil erosion, water pollution, among others. The general objective of this thesis was to evaluate the ecological conditions of savanna streams through the development of effective tools to efficiently conserve and manage these resources. Through a probabilistic sampling design, streams belonging to the hydrological units of four hydroelectric projects were selected (Nova Ponte, Três Marias, Volta Grande and São Simão) Simão). Macroinvertebrate sampling and physical habitat characterization followed the st andardized protocol of the US Environmental Protection Agency (US EPA) adapted for savanna streams. The objective of the first chapter was to evaluate the efficiency of macroinvertebrate taxonomic richness sampling effort obtained through samplings at local and basin scales. In addition, it was determined which environmental factors are associated with a greater sampling effort required to characterize the taxonomic richness on a local sca le. It was verified that the sample effort varies with the scale of study and with the evaluated taxonomic resolution. Metrics describing habitat heterogeneity and structure, anthropogenic disturbances in the riparian zone and water quality explained a sig nificant amount of the effort required to characterize the macroinvertebrate richness. In the second chapter the objective was to develop a multimetric index (MMI) capable of reflecting anthropogenic impacts using benthic macroinvertebrate assemblage metri cs. Steps of MMI development included: 1) selection of least and most disturbed sites ; 2) classification and selection of biological metrics; 3) selection of the best MMI model; 4) assignment of condition classes; 5) evaluation of the relationships betwee n MMI and anthropogenic stressors; 6) assessment of ecological conditions at local and regional scales. The final MMI was able to distinguish between least and most disturbed sites, responding to anthropogenic impacts related to land use, water quality, a nd physical habitat structure. In the third chapter, the association between anthropogenic stressors and the biological condition represented by MMI was evaluated at local (in each hydrological unit) and regional scales. For that, the relative risk (RR) an d relative stressor extent (RE) approach was used, which evaluates the importance and magnitude of a given stressor for the biological condition, respectively. It was verified that in regional scale turbidity, of fine sediment and % of agriculture are th e main stressors associated with poor biological conditions. In local scale % of pasture and total nitrogen are also associated with poor biological conditions but occurring in small extensions. In general, evaluating RR and RE together, it was found that reducing the im pact of excess sediment on the streambed can greatly improve the biological condition. The products of this thesis aim to provide subsidies for managers and decision makers to implement monitoring programs and environmental policies focused on the conservation of aquatic ecosystems.