Determinantes ambientais da ictiofauna de riachos e conhecimento ictiológico tradicional dos ribeirinhos da região Norte de Goiás

Abstract

Understanding how environmental impacts influence biodiversity in different spatial scales is fundamental, as it allows the identification of the most appropriate scales for the management of natural resources. In addition, the record and the formalization of popular knowledge on this topic can contribute to the expansion of knowledge on the socio-economic aspects related to the issue of environmental impacts and their influence on water resources and aquatic biota. The objective of this study is to evaluate the influence of environmental variables at the local and landscape scale on the abundance of stream fish communities. Additionally, we intend to also investigate the traditional ichthyological knowledge (TIK) and perception of people living near streams regarding the environmental impacts incidents in these environments. The study was carried out in streams of sub-basin of the river Santa Teresa, the Upper Tocantins river system in the northern region of Goiás. Fish individuals were sampled in 29 streams and the ethnoichthyological survey were performed with 15 interviews. The abundance of fish species were correlated with local and landscape environmental variables. The ethnoichthyological data were collected through interviews by using semi-structured questionnaire and the information obtained were analyzed from a qualitative and quantitative perspective. Were sampled 8,461 individuals belonging to 47 species, 15 families and five orders. Both global models resulted in significant relationships (local variables, R2 = 0.243, p = 0.015; landscape variables, R2 = 0.073; p = 0.027). Three local variables (percentage of consolidated substrate, proportion of litter and turbidity) were selected for the reduced model and explained 15.54%. Among the landscape variables, the area of micro-basin was retained in the reduced model, explaining 7.25%. Together environmental predictors accounted for 22% of the total variation in the abundance of the fish community. Local predictors accounted exclusively for 14.75% (p = 0.001) of the variation in species composition, while the landscape predictors accounted for 6.50% (p = 0.001). The proportion of the variance explained by both sets of predictors was 0.8%. Our results indicate greater contribution of local variables on the variation in the abundance of fish assemblages of streams. This highlights the importance of conservation of the features of these environments, especially the structural complexity of the habitat to maintain the diversity of stream fish. The people living near streams has perception related to the environment in the region and the environmental impacts that have caused the loss of regional environmental integrity. They highlighted the reduction and/or disappearance of some species. The ecological information about the fish refer mainly to habitat and trophic aspects of the species. The most commonly used criterion for etnoclassification was the morphology of ethnospecies. People from riverine communities have ample TIK, their perceptions of the importance of environmental conservation for the maintenance of species and to their quality of life reveals an environmental conscience positioning.