Geochemical baselines (GBs) play a crucial role in discerning natural variability from anthropogenic impacts on elemental composition within the environment. However, their applicability in quantifying the contribution of pollution sources to heavy metal contamination in sediments remains understudied. This research aimed to assess the degree of contamination and local pollution source attribution by leveraging geochemical baselines derived from statistical techniques, specifically the relative cumulative frequency (RCF) and 2σ-iterative (2σ-I) methods. In the urban water systems of Ma’anshan City, the major iron ore centre in eastern China, we observed concentration ranges of Cr, Cu, Ni, Pb and Zn in 36 sediment samples ranging from 66.89 to 352.08 mg/kg, 22.01 to 133.37 mg/kg, 22.66 to 50.80 mg/kg, 14.66to 264.37 mg/kg and 73.30 to 2707.46 mg/kg, respectively. RCF and 2σ-I techniques yielded similar GBs with no significant differences (p > 0.05). The geo-accumulation index and contamination factor analysis showed a sediment heavy metal accumulation rank of Zn > Pb > Cr > Cu > Ni. The contribution percentage of pollution sources varied with land functional type of watershed. For industry-influenced sediments, the contribution of local sources to Cr, Cu, Pb and Zn was significant, with shares of 43%–88%. Overall, this study highlights the valuable insights provided by GBs for effective management of urban aquatic environments.