Landscape influences on water quality composition and structure across multiple temporal and spatial scales in the hydrological unit of the Doce River in Minas Gerais

• Autor(a) principal: Santos, João Pedro dos
• Data de Publicação: 2023
• Outros Autores: Macedo, Diego Rodrigues
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: Revista Espinhaço
• Texto Completo: https://revistas.ufvjm.edu.br/revista-espinhaco/article/view/300
• topic: Qualidade da água; Métricas da paisagem; Extensões espaciais; Water quality; Landscape metrics; Spatial scales

The degradation of aquatic environments leads to water contamination, resulting in a deterioration of ecosystem quality. The Rio Doce basin has a history of environmental degradation, primarily due to industrial expansion and intensive agriculture. Additionally, in 2015, the Fundão tailings dam (Samarco, BHP Billiton, and Vale) ruptured, releasing ore tailings into the watercourses of the basin, abruptly altering its socio-environmental context and becoming one of the biggest environmental disasters in Brazilian history. The objective of this study was to analyze the relationship between landscape characteristics and water quality in the Minas Gerais portion of the Rio Doce basin across different spatial extents. Statistical models were employed to explore the relationship between water quality parameters and land use, land cover, and landscape metrics for 2008, 2013, and 2018. Three spatial extents were considered: sub-basin, corresponding to the entire upstream drainage area of the monitoring point; riparian, representing a 200-meter buffer zone on each side of the watercourse upstream of the monitoring point; and local, corresponding to a 200-meter buffer zone on each side of the watercourse within a 2-kilometer radius of the monitoring point. The parameters monitored included nitrate, total solids, and turbidity during both rainy and dry periods at 64 monitoring stations. The relationship between variables was assessed through multiple linear regression, allowing for constructing and selecting models with the best fits and identifying variables with the most significant impact on the models. The models with the best explanatory capacity were those of the sub-basin spatial extent (R²adj = 0.20-0.57; median: R²adj = 0.35) and riparian (R²adj = 0.23-0.48; median: R²adj = 0.39), while the local extent (R²adj = 0.07-0.47; median: R²adj = 0.23) performed less effectively. Human activities such as agriculture, urbanization, eucalyptus planting, and mining were associated with worsened water quality across all analyzed extents. Conversely, patches of natural vegetation were found to act as controllers of water quality. The study underscores the urgent need to halt deforestation and degradation in areas near watercourses and throughout the watershed.