A framework model to integrate sources and pathways in the assessment of river water pollution.

Metal and nutrient pollution, soil erosion, and alterations in climate and hydrology are prevalent issues that impact the water quality of riverine systems. However, integrated approaches to assess and isolate causes and paths of river water pollution are scarce, especially in the case of watersheds impacted by multiple hazardous activities. Therefore, a framework model for investigating the multiple sources of river water pollution was developed. The chosen study area was the Paraopeba River basin located in the Minas Gerais, Brazil. Besides multiple agriculture, industrial, and urban pollution sources, this region was profoundly affected by the rupture of the B1 tailings dam (in January 2019) at the Córrego do Feijão mine, resulting in the release of metal-rich waste. Considering this situation, thirty-nine physicochemical and hydromorphological parameters were examined in the Paraopeba River basin, in the 2019-2023 period. The analysis involved various statistical techniques, including bivariate and multivariate methods such as correlation analysis, principal component analysis, and clustering. The Paraopeba River was mainly impacted by metal contamination resulting from the dam collapse, whereas nutrient contamination, mainly from urban and industrial discharges, predominantly affected its tributaries. Additionally, the elevated concentrations of aluminum, iron, nitrate, and sulfate in both main river and tributaries can be attributed to diffuse and point source pollution. In terms of hydromorphology and soil type, the interaction between woody vegetation and erosion-resistant soils, especially latosols, contributes to the stability of riverbanks in the main river. Meanwhile, in the tributaries, the presence of neosols and sparse vegetation in urbanized areas promoted riverbank erosion potentially amplifying pollution. While the study was conducted in a particular watershed, the findings are based on a methodology that can be applied universally. Hence, the insights on surface water quality from this research can be a valuable resource for researchers studying watersheds with diverse pollution sources.

Prognosis of metal concentrations in sediments and water of Paraopeba River following the collapse of B1 tailings dam in Brumadinho (Minas Gerais, Brazil).

In January 25, 2019, the B1 dam of Córrego do Feijão mine located in Brumadinho municipality (Minas Gerais, Brazil) collapsed and injected nearly 2.8 Mm3 of iron (Fe)- and manganese (Mn)-rich tailings in the Paraopeba River. This study assessed the contribution of tailings to the contamination of sediments and water by those metals. The dataset was built through daily to weekly samplings executed in the two years following the event, at 27 sites located along the Paraopeba plus 9 sites located at the confluence of main tributaries. The results evidenced a distinct contribution in the sectors "Anomalous" (8.6-63.3 km downstream from the dam) and "Natural" (115.8-341.6 km). The "Anomalous" sector presented large Fe/Al (12.2 ± 6.4) and Mn/Al (0.33 ± 0.19) ratios in sediments, thus being rich in tailings, while the "Natural" sector presented small ratios (2.4 ± 1.0; 0.06 ± 0.03) comparable to the natural sediments. A 500-700 m3/s stream flow discharge in the Paraopeba caused pronounced drops to the Fe/Al and Mn/Al ratios in the "Anomalous" sector, attributed to the mixture of contaminated sediments from the main water course with uncontaminated sediments injected by the tributaries during the event. Non-linear regressions showed Fe/Al and Mn/Al declines in the "Anomalous" sector, related with tailings mobilization downstream. The concentrations of Fe and Al in the sediments correlated positively with the corresponding concentrations in the Paraopeba water, conditioned by raising discharge rates and variations in the water pH. The contribution of tailings to the Fe correlation was demonstrated. No direct relation was established between the Mn concentrations in water and stream discharge, because manganese is associated with fine particles in the tailings that are mobilized to the water column even under low flows. The preliminary results of Seasonal Autoregressive Integrated Moving Average models predicted the return of Paraopeba to a pre-collapse condition in 7-11 years.