Variable non-linear removal of viruses during transport through a saturated soil column.

Reduction of viral surrogates (bacteriophage MS2 and murine norovirus-1 [MNV-1]) and viruses naturally present in wastewater (enteroviruses, adenoviruses, Aichi viruses, reovirus, pepper mild mottle virus) was studied in a long-term experiment simulating soil-aquifer treatment of a non-disinfected secondary treated wastewater effluent blend using a 4.4 m deep saturated soil column (95% sand, 4% silt, 1% clay) with a hydraulic residence time of 15.4 days under predominantly anoxic redox conditions. Water samples were collected over a four-week period from the column inflow and outflow as well as from seven intermediate sampling ports at different depths. Removal of MS2 was 3.5 log10 over 4.4 m and removal of MNV-1 was 3 log10 over 0.3 m. Notably, MNV-1 was removed to below detection limit within 0.3 m of soil passage. In secondary treated wastewater effluent, MNV-1 RNA and MS2 RNA degraded at a first-order rate of 0.59 day-1 and 0.12 day-1, respectively. In 15.4 days, the time to pass the soil column, the RNA-degradation of MS2 would amount to 0.8 log10, and in one day that of MNV-1 0.3 log10 implying that attachment of MNV-1 and MS2 to the sandy soil took place. Among the indigenous viruses, genome copies reductions were observed for Aichi virus (4.9 log10) and for pepper mild mottle virus (4.4 log10). This study demonstrated that under saturated flow and predominantly anoxic redox conditions MS2 removal was non-linear and could be described well by a power-law relation. Pepper mild mottle virus was removed less than all of the other viruses studied, which substantiates field studies at managed aquifer recharge sites, suggesting it may be a conservative model/tracer for enteric virus transport through soil.

Introducing sequential managed aquifer recharge technology (SMART) - From laboratory to full-scale application.

Previous lab-scale studies demonstrated that stimulating the indigenous soil microbial community of groundwater recharge systems by manipulating the availability of biodegradable organic carbon (BDOC) and establishing sequential redox conditions in the subsurface resulted in enhanced removal of compounds with redox-dependent removal behavior such as trace organic chemicals. The aim of this study is to advance this concept from laboratory to full-scale application by introducing sequential managed aquifer recharge technology (SMART). To validate the concept of SMART, a full-scale managed aquifer recharge (MAR) facility in Colorado was studied for three years that featured the proposed sequential configuration: A short riverbank filtration passage followed by subsequent re-aeration and artificial recharge and recovery. Our findings demonstrate that sequential subsurface treatment zones characterized by carbon-rich (>3 mg/L BDOC) to carbon-depleted (≤1 mg/L BDOC) and predominant oxic redox conditions can be established at full-scale MAR facilities adopting the SMART concept. The sequential configuration resulted in substantially improved trace organic chemical removal (i.e. higher biodegradation rate coefficients) for moderately biodegradable compounds compared to conventional MAR systems with extended travel times in an anoxic aquifer. Furthermore, sorption batch experiments with clay materials dispersed in the subsurface implied that sorptive processes might also play a role in the attenuation and retardation of chlorinated flame retardants during MAR. Hence, understanding key factors controlling trace organic chemical removal performance during SMART allows for systems to be engineered for optimal efficiency, resulting in improved removal of constituents at shorter subsurface travel times and a potentially reduced physical footprint of MAR installations.

Start-up performance of a full-scale riverbank filtration site regarding removal of DOC, nutrients, and trace organic chemicals.

The performance of a full-scale riverbank filtration facility in Colorado was evaluated from initial start-up over a period of seven years including the impact of seasonal variations to determine whether sustainable attenuation of various chemical constituents could be achieved. Both, annual and seasonal average concentrations were determined for several wastewater-derived constituents including dissolved organic carbon (DOC), ultraviolet absorbance at 254 nm, nitrate, phosphate for the years 2006, 2009, 2010, 2012, and trace organic chemicals (TOrC) for years 2009, 2010, and 2012. ANOVA analyses and Student's t-tests were performed to evaluate the consistency of contaminant attenuation at the site. Findings revealed no significant statistical differences for any of the bulk parameters with the exception of phosphate suggesting a highly reliable attenuation of DOC and nitrate from start-up to full-scale performance. Phosphate attenuation, however, exhibited a steady decline, which was likely attributed to exhaustion of sorption sites in the subsurface porous media. The river's flow regime influenced both occurrence levels and attenuation of TOrC during riverbank filtration, i.e. less river discharge resulted in higher TOrC concentrations and lower proportion of river water in the recovered groundwater. Differences in removal performance between annual data sets for caffeine, trimethoprim, sulfamethoxazole, and carbamazepine were caused by variations in the source; concentrations in riverbank filtrate remained similar over several years. The seasonal assessment for TOrC revealed steady or improving removal between winter and summer seasons based on the statistical analysis with atenolol being the only exception likely due to an increased microbial activity at elevated temperatures.

Assessment of virus removal by managed aquifer recharge at three full-scale operations.

Managed aquifer recharge (MAR) systems such as riverbank filtration and soil-aquifer treatment all involve the use of natural subsurface systems to improve the quality of recharged water (i.e. surface water, stormwater, reclaimed water) before reuse. During MAR, water is either infiltrated via basins, subsurface injected or abstracted from wells adjacent to rivers. The goal of this study was to assess the removal of selected enteric viruses and a potential surrogate for virus removal at three full-scale MAR systems located in different regions of the United States (Arizona, Colorado, and California). Samples of source water (i.e., river water receiving treated wastewater and reclaimed water) before recharge and recovered groundwater at all three sites were tested for adenoviruses, enteroviruses, Aichi viruses and pepper mild mottle virus (PMMoV) by quantitative polymerase chain reaction (qPCR). Samples of groundwater positive for any virus were also tested for the presence of infectious virus by cell culture. PMMoV was the most commonly detected virus in the groundwater samples. Infectious enteric viruses (reovirus) were only detected in one groundwater sample with a subsurface residence time of 5 days. The results suggested that in groundwater with a residence time of greater than 14 days all of the viruses are removed below detection indicating a 1 to greater than 5 log removal depending upon the type of virus. Given its behavior, PMMoV may be suitable to serve as a conservative tracer of enteric virus removal in managed aquifer treatment systems.