The long-term development of water quality in managed aquifer recharge - Organic matter removal and related redox reactions

Managed aquifer recharge (MAR) has become a widely applied method for drinking water production. Still, the long-term impacts of MAR on groundwater are often questioned. The research objective was to investigate the long-term impact of MAR in terms of water quality development in an area where surface water naturally infiltrates the ground. At the studied site in Hollola, Finland, infiltration has been ongoing for thousands of years—a much longer period than can be reached at built bank infiltration facilities or by laboratory tests. Water quality development was characterized by measuring temperature, organic and inorganic carbon (TOC, DOC, DIC), dissolved oxygen (DO), Fe and Mn from lake water and groundwater samples, and by calculating the carbon budget. Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS) was used to determine the molecular composition of the organic matter. The water quality development was modeled with a 1D reaction transport model, and the parameter sensitivity of the model was tested by Generalized likelihood uncertainty estimation (GLUE). The natural system showed water quality development to be similar to built bank infiltration plants, with 80–90% removal of organic matter. A temperature-mediated seasonal pattern in concentrations of DO, Fe, and Mn and a decrease in concentration variations in the aquifer were observed. The carbon budget calculation showed that 25% of the dissolved carbon entering the aquifer was organic, mainly originating from the lake. Of the inorganic majority, 31% originated from the lake bank. Based on FT-ICR MS results, the removal of organic matter in the aquifer was selective: the relative abundance of oxygen-containing species decreased, while the relative abundance of sulfur- containing species increased. The TOC and DO concentrations in the model were most sensitive to the seepage velocity, the hydrodynamic dispersivity, and the reaction rate parameters controlling the oxygen-related degradation of TOC in the lake sediment and the aquifer. In addition to these parameters, Mn and Fe concentrations were sensitive to reaction rate factors for the dissolution of Mn in the aquifer, but the result was less clear. It was concluded that a MAR system can remove organic matter over the long term without losing its capacity, which is a promising result for the long-term sustainability of MAR. The results also highlighted the role of lake bottom sediments in bank infiltration. When modeling MAR, a model for bank processes should be included. DIC measurements and carbon budget calculation provide a tool for characterizing the lake bank processes in future studies. Identifying the key parameters in the groundwater model provided useful guidance for the calibration of MAR models and highlighted the importance of good conceptual understanding and flow velocity estimation in field studies.