Artificial groundwater treatment: biofilm activity and organic carbon removal performance.

The artificial recharge of sand aquifers with raw source waters is a means both explored and utilised by many water utilities to meet the future potable water demands for increasing urban populations. The microbial ecology within these systems is however, poorly understood, as is the role that microbial biofilms play in the quality of finished water. Knowledge of the ability of biofilm bacteria to metabolise natural organic matter (NOM) is limited, particularly in respect to the degradation of normally recalcitrant hydrophilic and hydrophobic humic acid fractions by sessile and planktonic microbial consortia within sand aquifer systems. To simulate the artificial recharge of sand aquifers that were proposed for the Greater Stockholm Area, four separate 4 m deep sand columns were fed raw lake water and examined over a 45-week study period. The simulated aquifer system (hydraulic retention time 9-16 h) demonstrated the removal of total organic carbon (TOC) (10+/-5%), direct total counts (DTC) of bacteria (74+/-11%), heterotrophic plate count (HPC) bacteria (87+/-5%) and assimilable organic carbon (AOC) (87+/-5%), thereby fulfilling an important barrier function, except for the removal of TOC. Hydrophilic humic acid fractions were more readily metabolised by microbiota (HPC and EUB338-positive cells) harvested from the raw source water (SSM-W), whilst hydrophobic humic acid fractions promoted higher activity by microbiota harvested from the sand matrix (SSM-S). The apparent low activity demonstrated by biofilm microbiota (approximately 40% and 25% of DTC were positive to EUB338 probing for sand matrix and slide biofilms, respectively) could be attributed to the highly recalcitrant nature of the organic loads, whilst at the same time explain the poor removal of TOC. Following nutrient activation (by the PAC assay) nonetheless, a 3-fold increase in the percentage of EUB-positive bacteria was observed on glass slides. Furthermore, the incubation of SSM-S with R2A increased probe-active cells from 57+/-8% to 75+/-7% of DTC and at the same time increased SSM-W from 38+/-8% to 50+/-10%. Whilst these results may imply a good potential for the biological treatment of water by shallow sand aquifers, further work should address the poor removal of TOC observed in this study.