Transport modeling of metal contaminants in a stream-aquifer system

Abstract

Pinal Creek, Arizona has been contaminated by discharge of acidic (pH = 5.8-6.3) and metal-rich ground water that was released from mining activities. In the stream, pH increases from approximately 6.0 to 7.8, while dissolved Mn(II) decreases from approximately 70.0 to 50.0 mg/L over 3 km downstream of the point of groundwater discharge. It was hypothesized that the spatial variation of in-stream pH is controlled by CO₂ gas-exchange and affects transport of dissolved Mn(II) through pH-dependent microbial oxidation in hyporheic zones. An existing transport model was extended to include carbonate equilibria, CO₂ degassing and pH-dependent Mn(II) removal processes and applied to predict the alkalinity-inorganic carbon (Cτ)-pH balance and transport of Mn(II) in natural stream based on field and laboratory experiments. The simulation results reproduced the overall trends of alkalinity, Cτ, and pH, and were in good agreement with dissolved Mn(II) in downstream concentrations. A multi-parametric sensitivity analysis (MPSA) was used to identify the relative sensitivity of predictions to physical and chemical parameters used in the extended transport model. MPSA results imply that CO₂ degassing and pH-dependent microbial oxidation are the most important factors controlling the spatial variation in pH and reactive uptake of dissolved Mn(II) in the stream system. Using stream tracer injections, streambed sediments and aquatic vegetation areas were identified as physical-storage zones within the stream where biogeochemical reactions were enhanced. The traditional one-storage transport model was extended to describe the hyporheic processes associated with two independent transient storage zones. The extended model was used to evaluate the applicability and accuracy of one-storage transport model to two-storage stream system. One-storage transport model has very strict limitation in its application for the two-storage stream system and is only valid if two retention times of transported solutes are very close (Tr₁/Tr2 1.0). The presented field and modeling approaches, which include model extension for the processes of CO₂ degassing and pH-dependent biogeochemical reactions, and generalized sensitivity analysis can improve our understanding of transport of metal contaminants in natural stream-aquifer system.