Quantitative risk assessment of metals in natural environments includes the development of site-specific conceptual models for transport and reaction of contaminants. A major research ambition is the development of more fundamental descriptions of reaction processes that utilise general reaction parameters and rely less on empirical, site-specific assessment of model parameters. For metals contaminants, the mobility, bioavailability and toxicity depends crucially on chemical speciation, particularly the distribution of a particular element between aqueous and solid phases. Of crucial importance is the redox transformation of Fe and Mn which provide significant adsorption capacity as higher valent oxide and hydroxide minerals which also act as sources of contamination upon their reductive dissolution. In contrast, sulphate reduction linked to microbial respiration creates a sink for a number of dissolved metal species by forming sparingly soluble metal sulphide mineral phases. Recent approaches to characterising the redox status of sub-surface environments includes use of electrochemical measurements, hydrochemical and mineral chemical analysis of solutes and mineral phases, monitoring the levels of biogenic gases such as H2, H2S and CH4, and microbiological techniques to characterise the function and activity of specific groups of organisms. Each of these approaches has drawbacks and advantages. More importantly, combining information from a variety of methods, with general reaction process models, has the potential to help determine model parameter values for important in situ reactions such as Fe, Mn and S mineral solubility and microbial respiration coupled to reductive dissolution of Fe and Mn minerals and formation of sulphide mineral phases.