The characterisation of river basins and the development of environmental quality diagnostic tools to subsidise the sustainable management of water resources require interdisciplinary approaches. The aim of this study was to integrate ecological and geographical metrics to diagnose water quality in the Velles river basin, tributary of the San Francisco river (Minas Gerais, Brazil). The characteristics of the composition and structure of benthic bioindicator communities collected in 19 different tributaries in the Vello river basin were used. The geographical information has been acquired for each of the sub-catchment areas of the sampling site. Urban and rural population density, sanitary infrastructure and land use and coverage have been synthesised through a major component analysis (PCA). The associations between ecological and geographical metrics have been analysed by linear regression using the three most representative axes of PCA as predictors, summarising: (I) urban influence (44.37 %), (ii) rural influence (16.88 %) and (iii) influence of natural coverage (14.63 %). The models generated for biological metrics taxo-rich, diversity of Shannon Wienner, BMWP, ASPT and% Oligochaeta were significant (R2 ≥ 44-90, p < 0,002) and all were dependent on the first axis of the PCA, showing that biological metrics are sensitive to urban pressures. We conclude that urbanisation is the main source of influence for the degradation of the water quality of the river of the Velles, with the potential to compromise the San Francisco river basin. The characterisation of Hydrological basins and the development of environmental evaluation tools to subsidise the sustainable management of water resources request interdisciplinary approaches. The aim of this study was to integrate ecologic and geological metrics to evaluate water quality in the Velles river basin, a tributary of the San Francisco river (Minas Gerais, Brazil). For that was used data on the position of Benthic macroinvertebrate assemblies of 19 different taxes in the Velhriver basin. Geographic information was identified for each of the sampling site differentiation sub-basins, making urban and rural population density, sanitary infrastructure and land use and cover. These metrics were synthesised through a Principal Component Analysis (PCA) and the associations between ecologic and geographical metrics were adapted using linear regressions, with the three most representative axes of PCA as predictors Synthesising: (I) urban influence (44.37 %), (ii) rural influence (16.88 %), and (iii) influence of natural cohesion (14.63 %). The models produced for the biological metrics Richness, diversity of Shannon Wienner, BMWP, ASPT and% of Oligochaeta showed a significant correlation (R2 ≥ 44-90, p < 0.002) and all were dependent on the first axis of the PCA, highlighting that the biological metrics are perceived to urban assumptions. We conclude that Urbanisation is the main cause of degraded water quality in the Velles River Hydrological basin, with potential to commit it for the entire São Francisco River basin.