Article pubs.acs.org/est
Degradation of Polar Organic Micropollutants during Riverbank Filtration: Complementary Results from Spatiotemporal Sampling and Push−Pull Tests Sebastian Huntscha,†,‡ Diana M. Rodriguez Velosa,† Martin H. Schroth,‡ and Juliane Hollender*,†,‡ †
Eawag, Swiss Federal Institute of Aquatic Science and Technology, 8600 Dübendorf, Switzerland Institute of Biogeochemistry and Pollutant Dynamics (IBP), ETH Zurich, 8092 Zurich, Switzerland
‡
S Supporting Information *
ABSTRACT: The fate of polar organic micropollutants (logDOW (pH 7) between −4.2 and +3.5) during riverbank filtration (RBF) at the river Thur was studied using both spatiotemporally resolved sampling and single-well push−pull tests (PPT), followed by LCMS/MS analysis. The Thur is a dynamic prealpine river with an alluvial sandy-gravel aquifer, which is characterized by short groundwater travel times (a few days) from surface water infiltration to groundwater extraction. The spatiotemporal sampling allowed tracing concentration dynamics in the river and the groundwater and revealed persistence for the drug carbamazepine, while the herbicide MCPA (2-methyl-4-chlorophenoxyacetic acid) and the drug 4-acetamidoantipyrine were very quickly degraded under the prevalent aerobic conditions. The corrosion inhibitor 1H-benzotriazole was degraded slightly, particularly in a transect influenced by river restoration measures. For the first time in situ first-order degradation rate constants for three pesticides and two pharmaceuticals were determined by PPTs, which confirmed the results of the spatiotemporal sampling. Atenolol was transformed almost completely to atenolol acid. Rate constants of 0.1−1.3 h−1 for MCPA, 2,4-D, mecoprop, atenolol, and diclofenac, corresponding to half-lives of 0.6−6.3 h, demonstrated the great potential of RBF systems to degrade organic micropollutants and simultaneously the applicability of PPTs for micropollutants in such dynamic systems.
■
INTRODUCTION
RBF is dilution with groundwater that is not influenced by surface water. Several studies on the fate of organic micropollutants in RBF and related systems such as lake bank filtration and managed aquifer recharge (MAR) systems were conducted in the past decade, including pharmaceuticals in many cases6−15 and pesticides in some cases.9,16,17 However, the focus of these investigations was mostly on systems with long groundwater travel times (up to several years) from surface water infiltration to groundwater extraction. Removal efficiencies (%) for micropollutants in the underground were obtained by comparing average concentrations of long-term monitoring data of the respective river, lake, or infiltration pond and the corresponding groundwater. This way, a redox potential dependence of the degradation potential in the subsurface was found for many micropollutants, with aerobic conditions being favorable for the degradation of some compounds (e.g., 4-acetamidoantipyrine and clindamycin8,13) and anaerobic
Riverbank filtration (RBF) is a method to achieve high quality drinking water without expensive technical treatment,1 which is widely used in Europe and to a lesser extent in the USA,2 but has worldwide potential.3 During the underground passage of infiltrating river water, particulate, dissolved, and bacterial contaminants are removed efficiently by natural attenuation processes. Thereby, the hyporheic zone, i.e. the interface between surface water and groundwater, plays an important role as it is a region of intensified biogeochemical activity.1,4 Many RBF sites are located in intensively used areas where the rivers receive considerable amounts of treated wastewater from wastewater treatment plants (WWTP) as well as runoff from agriculture. Hence, organic micropollutants such as pesticides, pharmaceuticals, and industrial chemicals may impair the drinking water quality of RBF systems if they are not removed in the aquifer. Generally, RBF removes organic micropollutants through biodegradation and sorption processes.5 However, sorption plays a minor role for polar organic micropollutants and often results merely in compound retardation rather than permanent removal in RBF aquifers. Another process that leads to concentration decreases during © 2013 American Chemical Society
Received: Revised: Accepted: Published: 11512
April 25, 2013 September 2, 2013 September 13, 2013 September 13, 2013 dx.doi.org/10.1021/es401802z | Environ. Sci. Technol. 2013, 47, 11512−11521
Environmental Science & Technology
Article
to reveal the spatial and temporal differences in the biodegradation potential of an alluvial sandy-gravel aquifer. Degradation kinetics were studied in more detail in PPTs for well-degradable compounds.
conditions for other compounds (e.g., carbamazepine and sulfamethoxazole8,14). Several compounds appeared to be persistent under all redox conditions (e.g., AMDOPH (1acetyl-1-methyl-2-dimethyloxamoyl-2-phenylhydrazide) 13) or degraded without showing a clear redox-dependency (e.g., diclofenac and sotalol 14). Consequently, it was concluded that the occurrence of oxic and anoxic conditions at a bank filtration site together with a high residence time and travel distance was beneficial for the degradation of organic micropollutants in aquifers.18 The methods applied in those studies are not suitable to investigate the fate of micropollutants in RBF systems with short groundwater residence times and a highly dynamic input, such as those found on the Swiss plateau in the Upper Rhine catchment. Here, a considerable amount of drinking water is produced from alluvial aquifers that are fed by prealpine rivers without a lake or reservoir and that are, thus, subject to large variations in discharge. In Switzerland, the legislation requires a minimum subsurface residence time of only 10 days for drinking water abstraction, compared with 50 days in the European Union. As a result, short-term fluctuations resulting from floods or chemical spills may have a big impact on the water quality. Such dynamic RBF systems require a much higher sampling frequency or selective methods to quantify degradation of micropollutants during the underground passage. River restoration projects that are increasingly conducted at Swiss rivers as modern flood protection measures and to achieve a “good ecological status” of surface waters as required by the European Water Framework Directive19 even increase the heterogeneity of RBF systems. Their typical measures include the removal of bank fixation, the elimination of overbanks and the formation of gravel bars, islands, and meanders similar to those found in a natural state,20 which influence the exchange between river and groundwater and thus potentially the fate of micropollutants. A method that enables the determination of in situ microbial reaction rates in groundwater is the single-well push−pull test (PPT).21,22 This reactive tracer method consists of the injection of a test solution with one or more reactants together with a nonreactive tracer into the aquifer and the evaluation of breakthrough curves during a subsequent extraction phase. Over the last two decades, PPTs have been used to study contaminant degradation by aerobic respiration in petroleumcontaminated aquifers,23,24 sorption and transport of surfactants and perfluorinated compounds,25,26 and degradation of chlorinated hydrocarbons and PAHs27−29 as well as denitrification rates23,30 in different aquifers. The goal of this study was to increase the understanding of the degradation behavior of polar organic micropollutants from agricultural and urban sources in a highly dynamic RBF system by applying two different methods. On the one hand, highly spatiotemporally resolved sampling aimed to trace pesticide concentration peaks and concentration dynamics of pharmaceuticals and corrosion inhibitors in the river and during the subsequent RBF at two adjacent groundwater transects with young infiltrate (
