Environ. Sci. Technol. 2010, 44, 4877–4883
Micropollutant Loads in the Urban Water Cycle A N D R E A S M U S O L F F , * ,† SEBASTIAN LESCHIK,† FRIDO REINSTORF,‡ GERHARD STRAUCH,† AND MARIO SCHIRMER§ Department of Hydrogeology, UFZsHelmholtz Centre for Environmental Research, Permoserstr. 15, 04318 Leipzig, Germany, Department of Water Resources and Waste Management, University of Applied Sciences Magdeburg-Stendal, Breitscheidstr. 2, 39114 Magdeburg, Germany, and Department of Water Resources and Drinking Water, Eawag: Swiss Federal Institute of Aquatic Science and Technology, Ueberlandstr. 133, 8600 Duebendorf, Switzerland
Received December 17, 2009. Revised manuscript received April 20, 2010. Accepted April 20, 2010.
The assessment of micropollutants in the urban aquatic environment is a challenging task since both the water balance and the contaminant concentrations are characterized by a pronounced variability in time and space. In this study the water balance of a central European urban drainage catchment is quantified for a period of one year. On the basis of a concentration monitoring of several micropollutants, a contaminant mass balance for the study area’s wastewater, surface water, and groundwater is derived. The release of micropollutants from the catchment was mainly driven by the discharge of the wastewater treatment plant. However, combined sewer overflows (CSO) released significant loads of caffeine, bisphenol A, and technical 4-nonylphenol. Since an estimated fraction of 9.9-13.0% of the wastewater’s dry weather flow was lost as sewer leakages to the groundwater, considerable loads of bisphenol A and technical 4-nonylphenol were also released by the groundwater pathway. The different temporal dynamics of release loads by CSO as an intermittent source and groundwater as well as treated wastewater as continuous pathways may induce acute as well as chronic effects on the receiving aquatic ecosystem. This study points out the importance of the pollution pathway CSO and groundwater for the contamination assessments of urban water resources.
Introduction A variety of organic compounds from wastewater sources such as pharmaceuticals, additives of personal care products (collectively PPCP), household chemicals, and industrial chemicals are widely recognized as a potential threat to aquatic ecosystems and to human health. Because most of these contaminants are environmentally present in low concentration ranges of pg L-1 to ng L-1 often the term micropollutant is used (1). Major concerns regarding the risk micropollutants pose on aquatic ecosystems and on human health involve endocrine disrupting effects (2), chemosen* Corresponding author e-mail:
[email protected]. † UFZsHelmholtz Centre for Environmental Research. ‡ University of Applied Sciences Magdeburg-Stendal. § Eawag: Swiss Federal Institute of Aquatic Science and Technology. 10.1021/es903823a
2010 American Chemical Society
Published on Web 05/28/2010
sitizing effects (3), possible interactions of contaminant mixtures, and chronic effects by a long-term exposure (4). Since large amounts of wastewater are produced, transported, and treated in urban agglomerations, urban water resources are known hot spots of micropollutant contaminations (5, 6). Until now, most studies deal with the occurrence of micropollutants in the effluents of wastewater treatment plants (WWTP) and in the receiving waters. This source is considered most important because often micropollutants are only partially removed during the treatment process and thus continually released into the aquatic environment (7, 8). Fewer studies deal with the input of micropollutants by combined sewer overflow (CSO). In areas served by combined sewage systems, stormwater runoff can exceed the capacity of the sewer network and untreated wastewater can be released into surface waters. CSO is recently recognized as an important additional source for micropollutants in surface waters (9-11). Only a few studies deal with micropollutants in groundwater resources. Here, sewer leakages and the influence of contaminated surface water may deteriorate groundwater quality on a long-term (6, 12-16). To the best of our knowledge, so far no study shows the relevance of different wastewater pathways for the micropollutant contamination of groundwater and surface water in urban areas. Here, a comprehensive knowledge of the urban water flow as well as of micropollutant occurrence in the different water compartments is required. Surface sealing and exfiltration of groundwater to and losses from the sewage system strongly affect the water and contaminant balance (17). The resulting spatiotemporally variable distribution of water flow as well as the diffuse, line, and point character of contaminant sources lead to a complex pattern of micropollutant concentrations in urban waters (18, 19). We hypothesize that for complex urban areas a holistic approach including wastewater, groundwater, and surface water flow, and contamination, is capable of describing micropollutant loads in all the water compartments and, thus, helps to improve the knowledge of the relevance of the different environmental pathways. In the chosen study area the catchment of shallow groundwater is interacting with the matching drainage catchment of wastewater, the sewershed. Therefore, an integrative water and contaminant mass balance is possible, including the sewage system as a source of micropollutants to groundwater and surface water as recipients (5). In our opinion, the study area with unconsolidated aquifers and a shallow groundwater surface can be seen as a typical setting for central European cities: Most cities are founded in direct vicinity of a river with connected alluvial sediments. Brombach (20) reports 40% of Germany’s total wastewater flow is parasitic water mostly from groundwater exfiltration which also hints to intense interaction of the sewage systems with shallow groundwater. So far, research was focused on monitoring micropollutant concentrations in the different water compartments over the course of a year and evaluation on the basis of statistical methods (15, 19). This study focuses on the micropollutant loads in the urban water compartments for a sewershed within the city of Leipzig, Germany. We use the results of the monitoring program concerning micropollutant concentration probability in the treated and the untreated wastewater and in the groundwater. Numerical models of groundwater recharge and groundwater flow as well as the measured wastewater flow from the sewershed are combined with the micropollutant concentrations to estimate annual loads. Thus, we are able to (a) characterize the water flow in and VOL. 44, NO. 13, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
4877
FIGURE 1. Study area with boundaries of the groundwater flow model, location of the sewershed, land use, and surface sealing. Upper right map shows location within Germany. WWPS is the wastewater pumping station; 0 indicates the outlet of combined sewer overflow facilities. between the different urban water compartments, (b) describe the micropollutant load balance, (c) specify the temporal distribution of micropollutant release to the environment, and (d) discuss the relevance of different micropollutant release pathways and the potential impact on the aquatic ecosystem.
Materials and Methods Study Area. The study area is located in the northwestern part of the central European city of Leipzig (Figure 1). We focus on a subcatchment of the sewage system, the sewershed, with a size of 5.37 km2. Here, 24.6% of the area is fully sealed by roads and buildings, and 31.8% is partly sealed. The sewershed is drained by a combined sewage system with a total length of 54.7 km. A fraction of 29% of the sewers is located beneath the average groundwater table. Wastewater is flowing to a station that pumps all wastewater to the municipal WWTP northeast of the study area, serving most parts of the city of Leipzig. Treated wastewater from the WWTP discharges into the river system of the Weisse Elster. A small stream, the Bauerngraben, located on the eastern and northern edge of the sewershed, receives no surface runoff except CSO water at six locations along the course. To ensure constant water flow within the stream Bauerngraben, river water from the Weisse Elster is bypassing through this stream. The study area is underlain by three connected quaternary fluvial sand and gravel aquifers and one discontinuously distributed tertiary sandy aquifer. Two discontinuously distributed boulder clay layers between the quaternary aquifers and one distinctive floodplain loam layer in the 4878
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 13, 2010
northern part of the area act as aquitards. Groundwater flows mainly from the south to the north forming a watershed in good agreement with the sewershed. Within the sewershed, groundwater is unconfined and oxic. Sample Collection and Additional Field Work. The monitoring of water quality was conducted from April 2007 to April 2008 and is comprehensively described in refs 15, 19. Briefly, within the sewershed seven groundwater observation wells were sampled every three months concerning micropollutant concentrations. Furthermore, two wells, the inflow after primary sedimentation, and the outflow of the municipal WWTP were sampled every month. Groundwater samples were taken by submerged pumps; wastewater samples were taken as grab samples. Additional fieldwork included automatic measurements of groundwater table at eight observation wells. Sample Analysis and Quality Assessment. A total number of 13 samples from the WWTP influent, 13 samples from the effluent, and 56 samples from the groundwater were considered in this study. Six micropollutants were chosen for analysis: bisphenol A (BPA), a precursor of polycarbonate production with estrogenic activity (21); technical 4-nonylphenol (NP), an isomer mixture from the degradation of nonylphenol polyethoxylates (used as surfactants) with estrogenic activity, with restricted use in the European Union since 2005 (22); caffeine (CAF), a stimulant from beverages and pharmaceuticals, an established marker substance for wastewater (23); galaxolide (HHCB) and tonalide (AHTN), polycyclic musk fragrances with a potential of bioaccumu-
FIGURE 2. Scheme of water flow between the urban water compartments. Gray arrows indicate the pathways relevant for the environmental release of micropollutants from the sewershed: WWPS, wastewater pumping station; WWTP, wastewater treatment plant; ET, evapotranspiration; P, precipitation; Qro, surface runoff to the sewage system (stormwater); Qww, strict wastewater flow; Qrch,rain, groundwater recharge from infiltrated rainwater; Qrch,ww, groundwater recharge from sewer leakages; Qcso, combined sewer overflow to the surface water; Qex, groundwater exfiltration flow to the sewage system; Qgw, groundwater discharge through the sewershed boundary; Qww,p, combined wastewater pumped to the WWTP; Qtww, treated wastewater from the WWTP. lation and chemosensitizing effects (3); carbamazepine (CBZ), antiepileptic drug with high persistence in the water cycle (12). The chemical analysis was performed using solid phase extractionatneutralconditionsandgaschromatography-mass spectrometry (see refs 24, 25). Analysis was performed without derivatization and used external standards for the quantification of CAF, HHCB, AHTN, and CBZ (every 5 analysis) and internal standards for NP (4-n-nonylphenol) and BPA (d16 bisphenol A). Concentrations below the limit of detection were set to 0 ng L-1. Micropollutant concentrations from samples with a problematic chemical analysis (e.g., low standard recovery, overlapping peaks in the chromatogram) were not considered. Water Flow Quantification. The annual water balance of the sewershed is based on the following equations P + Qww ) ETa + Qgw + Qcso + Qtww + ∆Sgw
(1)
Qww + Qro + Qex ) Qrch,ww + Qcso + Qtww
(2)
where P is the precipitation; Qww is the strict wastewater flow; ETa is the actual evapotranspiration; Qgw is the groundwater discharge through the sewershed boundary; Qcso is the combined sewer overflow to the surface water; Qtww is the treated wastewater flow from the WWTP; ∆Sgw is the storage change of groundwater within the monitored year; Qro is the surface runoff to the sewage system (stormwater); Qex is the groundwater exfiltration flow to the sewage system; Qrch,ww is the groundwater recharge from sewer leakages. An overview of the considered water flow between the urban water compartments is given in Figure 2. The quantification of the sewershed’s water balance is based on a combination of different model and estimation approaches for the different water compartments. The methodology and data basis for the water flow quantification is given as an overview in Table SI1 and explained in detail in the Supporting Information. Load Estimation. Annual contaminant loads were estimated by multiplying the measured micropollutant concentration with the estimated annual water flow. Loads of micropollutant i out of the sewershed by untreated wastewater to the WWTP (Mi,uww), by treated wastewater (Mi,tww) and CSO (Mi,cso) to the receiving surface waters, and by
groundwater discharge out of the sewershed boundary (Mi,gw) were derived. To account for the statistical distribution of concentrations, we used the median as well as the 5th, 25th, 75th, and 95th percentiles for the groundwater samples and the median, 25th, and 75th percentiles for the wastewater samples. Details on the quantification are given in the Supporting Information. Concentration measurements are published in ref 19. Groundwater samples taken outside the sewershed are not considered here. Therefore, Table SI2 in the Supporting Information gives an overview of concentration percentiles of micropollutants used in this study. Since it was not feasible to sample wastewater in the pumping station of the sewershed, untreated wastewater samples were taken in the municipal WWTP. Thus, it was assumed that untreated wastewater from the sewershed had the same properties as the taken wastewater samples. The estimation of Mi,tww to the surface water does not take seasonality of micropollutant concentrations in the treated wastewater into account. A seasonality of contaminant loads was limited to NP (19) but did not have influence on the annual load from the WWTP since the wastewater discharge was uniform over the year. For groundwater samples, no spatial trend of concentrations was found (19) while temporal trends were limited to an infiltration of contaminated water from the stream Bauerngraben to the groundwater outside of the sewershed. Thus, groundwater samples from inside the sewershed were assumed to be representative for all the groundwater flowing through the boundaries of the sewershed.
Results and Discussion Urban Water Balance. An overview of the major water flow to and from the sewershed is given in Table 1. Precipitation (P) was the major source of water in the sewershed. P compares well with the mean precipitation (597 mm a-1, standard deviation 86 mm a-1, 1994-2008) in the city of Leipzig. Additionally, water was imported to the sewershed by water mains and released as strict wastewater flow into the sewage system (Qww, corresponds to 31% of the precipitation amount). For the part of the sewage system beneath the groundwater surface, groundwater exfiltration into the sewer section (Qex) made up 16% of the combined wastewater pumped out of the sewershed. The amount of wastewater losses to the groundwater (Qrch,ww) based on the measured micropollutant concentrations is a matter for discussion below. Although the sewershed is largely sealed, only 11% of the precipitation is discharging as stormwater runoff (Qro) to the sewage system. Thus, a considerable amount of the sealed surface was not contributing to the sewage system or allowed water to infiltrate through cracks and joints into the subsurface. Nevertheless, the sealed surface reduced the evapotranspiration. This resulted in a high groundwater recharge rate (Qrch,rain, 37% of the precipitation). The measured groundwater surface as well as the modeled groundwater recharge rates indicate a very fast response of the water table to rainfall with maxima in a time period of 18-28 h. Thus, groundwater was dominantly recharged following rainfall events; a slow recharge component was not observed. The modeled annual discharge of groundwater out of the sewershed boundary (Qgw) was lower than the groundwater recharge. Consequently, there was a change of groundwater storage ∆Sgw in the aquifers through the monitored year. In the stream Bauerngraben, CSO from the sewershed (Qcso) corresponds to 16% of the annual undisturbed surface water flow. Overall, 33 events with a mean duration of 4 h were registered. The release of untreated wastewater by CSO corresponds to a mean of 2.5% (min 1.9%; max 3.0%) from the annual untreated wastewater discharge (dry weather flow) VOL. 44, NO. 13, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
4879
TABLE 1. Major Water Flow of the Sewersheda mm a-1
flow type
description
P ET Qww,p Qww,ex Qww
precipitation potential evapotranspiration pumped combined wastewater pumped combined wastewater without stormwater strict wastewater flow, without stormwater and exfiltration flow from groundwater groundwater exfiltration flow to the sewage system storm water runoff to the sewage system combined sewer overflow, max > mean > min wastewater in combined sewer overflow, max > mean > min groundwater recharge sewer leakages groundwater discharge through the sewershed boundary storage change of groundwater
Qex Qro Qcso Qww,cso Qrch Qrch,ww Qgw ∆Sgw a
663 364 307 252 247
9721 9574 4497 3699 3618
46 73 27 > 26 > 24 9>7>5 242 28-37 118 124
670 1064 396 > 376 > 348 126 > 106 > 78 3549 417-549 1724 1825
The estimation of sewer leakages is discussed below.
FIGURE 3. Distribution of median loads concerning the environmental release of micropollutants from the sewershed by (dark shading) treated wastewater discharge into the Weisse Elster river; (light shading) combined sewer overflow to the stream Bauerngraben; (white) groundwater discharge through the sewershed boundaries. to the surface water. The measured number of events is below the median overflow frequency in German sewage systems (26) (45-55 events per year). Micropollutant Balance. The annual cumulative micropollutant loads are based on the cumulative water flow and the probability distribution of measured concentrations. Derived micropollutant loads are given in Figure SI3 in the Supporting Information. Micropollutant release from the sewershed was evaluated by comparing the loads from treated wastewater (Mtww), from CSO (Mcso), and by groundwater discharge out of the sewershed boundary (Mgw). For the total released micropollutant mass, HHCB, BPA, and NP dominated over the other micropollutants (see mass released per area, Table SI3 in the Supporting Information). Concerning the relevance of different pathways, significant differences are apparent (Figure 3). For the release of CBZ, AHTN, HHCB, NP, and BPA, treated wastewater was the most important pathway. Nevertheless, Mcso is of significance. Median CAF output by CSO made up 88.8% of the total mass released to the environment. Also for BPA (10.3%) and NP (8.2%), CSO contributed significant loads to the environment. For BPA (26.0%) and NP (9.5%), a considerable amount of the total release from the sewershed can be assigned to the loads in groundwater discharge (Mgw). Several authors describe the interplay of dilution by stormwater and by the receiving surface water and the removal efficiency in WWTPs to be driving agents of 4880
m3 d-1
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 13, 2010
micropollutant enrichment in surface waters following CSO (10, 27). Loads of micropollutants with pronounced removal in the WWTP (here: CAF; BPA; NP; AHTN) were characterized by a higher fraction of CSO input than loads of more persistent contaminants (here: HHCB, CBZ). The importance of CSOrelated release of micropollutants to surface waters is described in several studies. Findings from Buerge et al. (9) (in summer, 90% of CAF in a Swiss lake came from CSO inputs) and from Phillips et al. (11) (in spring and summer, 90% of CAF in a North American lake came from CSO inputs) agree with the results presented here. In accordance with our findings of BPA release by CSO, Jonker et al. (28) improved a river mass flow balance for BPA by adding untreated wastewater sources during wet weather. The pattern of high NP and BPA loads and low CAF, HHCB, AHTN, and CBZ loads in groundwater is a result of the source concentrations and persistence of micropollutants on the pathway to and within the groundwater. Subsurface degradation and persistence of micropollutants in the unsaturated zone and within groundwater are discussed in several studies, but results are not always transferable since they depend on local conditions. NP and BPA were found to be degradable in oxic groundwater but persistent in under anoxic conditions (29, 30). Despite its degradability, BPA was detected in 20% of groundwater samples from U.S. drinking water production wells (31). NP can be formed from its parent compounds as well as degraded in the unsaturated zone and within groundwater (29, 32). HHCB, AHTN, and CAF concentration patterns in the study area’s groundwater hint to removal processes in the subsurface (19) which is in agreement with the findings of several other authors (32, 33). CBZ is a very persistent micropollutant (34) and was detected in drinking water production wells in the United States in the same frequency as BPA (31). On the basis of our measurements and the literature’s findings, we assume the high loads of BPA and NP in groundwater to be a result of only little or no removal while NP may additionally be formed in the subsurface. Removal is assumed to be effective for HHCB, AHTN, and CAF leading to low groundwater loads. CBZ is assumed to be persistent, and low groundwater loads are a result of the low concentrations in the untreated wastewater source. Sewer Leakages. Results of this study point at a considerable fraction of micropollutant released from the sewershed by groundwater discharge. In previous studies (15, 35) sewer leakages were found to be the most likely reason for micropollutant occurrence in the sewershed’s groundwater. The knowledge of the total groundwater recharge rate and the micropollutant concentrations in untreated wastewater as the source and in groundwater as the recipient allows for
the quantification of sewer leakages. We applied the approach of Lerner (17) for conservative contaminants with a distinct origin and steady-state conditions. This approach does not distinguish between wastewater losses due to direct interaction of groundwater with the sewage system or vertical wastewater seepage and passage through the unsaturated zone to the groundwater. Ci,gw )
1 ((Qrch,wwCi,uww) + (Qrch,rainCi,rain)) Qrch
(3)
Here, Ci,gw is the average concentration of contaminant i in groundwater; Qrch is the annual amount of total groundwater recharge, Qrch,ww is the amount of recharge from untreated wastewater (sewer leakages), Qrch,rain is the amount of recharge from precipitation, Ci,uww is the average concentration in untreated wastewater, and Ci,rain is the average contaminant concentration in the groundwater recharge from precipitation. Since Ci,rain is assumed to be zero, the equation simplifies and can be transposed to Qrch,ww. We assume the amount of total recharge Qrch to be equal to the modeled recharge Qrch,rain within the urbanized part of the study area. Here, the numerical groundwater recharge model was calibrated to the total recharge on the basis of the water level fluctuation method (36), which includes the water quantity from sewer leakages. Sewer leakages were quantified for each micropollutant (Table SI4 in the Supporting Information). Due to the different levels of persistence of the studied micropollutants, the derived leakage rates vary over 2 orders of magnitude. On the basis of the discussion above we assume most reliable leakage rates from the concentrations of CBZ and BPA. The used median CBZ concentration in untreated wastewater on the basis of grab samples may be limitedly representative. As discussed in Ort et al. (37), a representative sampling of CBZ in sewage systems is best done by flow proportional samples since the high temporal concentration variability of CBZ is a result of only a low daily number of input pulses. The differences in untreated and treated wastewater concentrations presented here also hint to the limited validity of grab samples and possibly a more error-prone chemical analysis in the complex matrix of untreated wastewater (see also ref 19). We therefore rely on two micropollutants with comparable persistence in the subsurface. To show the possible uncertainty of leakage rates based on the untreated wastewater samples of CBZ we included the leakage estimation on the basis of the treated wastewater concentrations in Table SI4 in the Supporting Information. The estimated leakage rate of 28-37 mm a-1 (0.09-0.12 -1 L s km-1 referring to the total length of sewer sections; 9.9-13.0% from the dry weather flow of wastewater, Qww,ex + Qrch,ww) corresponds to 12-16% of the total groundwater recharge rate. Results are on the same order of magnitude as those found in a study by Leschik et al. (16) on a sewer section within the sewershed by applying integral pumping tests and direct measurements in a blocked sewer section (0.3-0.7 L s-1 km-1). In comprehensive reviews, Rutsch et al. (38) state leakage rates of 1-10% of the wastewater’s dry weather flow, while Chisala et al. (39) found rates of 0.01-0.10 L s-1 km-1 to be most likely. The leakage rates estimated here represent the upper limit of these reviews’ findings. Since 12-16% of the total groundwater recharge rate can be assigned to sewer leakages, the temporal dynamics of leakages probably follows the temporal dynamics of raininduced groundwater recharge. The calibration of the groundwater flow model on the basis of the measured hydraulic heads was only feasible taking distinct groundwater recharge events into account. Therefore, we assume sewer leakages to occur simultaneously with natural groundwater recharge during and shortly after rainfall and to be not evenly distributed over time. This agrees with findings that the
FIGURE 4. Cumulative release of caffeine (black lines) and bisphenol A (gray lines) from the sewershed by treated wastewater, combined sewer overflow (CSO), and groundwater discharge.
temporal dynamics of wastewater composition and of wastewater hydraulic heads may enforce the intermittent character of sewer leakages (40). Temporal Dynamics of Release to the Aquatic Environment. For the risk assessment of micropollutant release from urban areas, a comprehensive knowledge of concentrations and load dynamics in receiving waters is crucial. Temporal dynamics of concentrations in the surface water and in the groundwater were discussed in ref 19. Micropollutant load dynamics of the release by groundwater, treated wastewater, and CSO are shown in Figure 4. Micropollutant loads from treated wastewater inputs and groundwater discharge were evenly distributed over the course of the year. On the contrary, micropollutant release via CSO was distinctly distributed over 33 events, operating in only 1.5% of the monitored time of one year. Up to 17% of the annual load was released in single events. For the events, a mean fraction of 28% of untreated wastewater in the CSO water was estimated. Adding the undisturbed flow of the receiving stream Bauerngraben, during CSO a mean fraction of 18% of untreated wastewater in the streamwater events was estimated. As a consequence, in the stream Bauerngraben higher micropollutant concentrations than in the effluents of the WWTP occurred. On the larger, catchment, scale, an increase of micropollutant concentrations due to CSO strongly depends on the relation of river water flow to the wastewater inputs. For the Weisse Elster catchment, recipient of the Bauerngraben water, elevated micropollutant loads but not concentrations were observed after rainfall (19). Therefore, the results obtained in the Bauerngraben are best transferable to smaller urban streams. Consequences. The applied holistic approach of incorporating wastewater, surface water, and groundwater flow in a sewershed to reveal the related micropollutant loads enables us to draw conclusions on the relevance of pathways of micropollutant releases. Urban areas are emitters of micropollutants. Without doubt, the continuous release of treated wastewater from the sewershed is the most relevant source of micropollutants for the receiving surface waters. This results in a long-term chronic exposure of the aquatic ecosystem and a potentially sustained contamination of surface water resources. However, depending on the removal efficiency of the WWTP, CSO can contribute a significant portion of micropollutant loads to the surface waters. Here, the intermittent character with high micropollutant concentration may pose the risk of acute effects on the aquatic ecosystem in small streams. An estimated fraction of 9.9-13.0% of the dry weather flow VOL. 44, NO. 13, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
4881
wastewater in the investigated sewer sections was directly lost to the groundwater. Consequently, also continuous groundwater discharge contributed to a considerable part of the micropollutant release from the sewershed. As a result of this study, the pollution pathways of CSO as well as of groundwater should be adequately considered in assessments of micropollutant contamination of urban water resources. The micropollutant inputs to surface waters, with natural attenuation by natural processes such as volatilization, photolysis, or biodegradation, may level part of the contamination (19, 41). Micropollutant loads by urban groundwater discharge were found to be relevant and can exceed CSO loads from the same area. Here, the large volume of contaminated groundwater, the slow flow velocities, and the possible persistence of micropollutants should be kept in mind. There is strong evidence that micropollutants persist in groundwater for decades (14, 29). As a consequence, a long-term and probably sustained contamination of groundwater with micropollutants beneath urban areas can be expected.
Acknowledgments We express our thanks to the Kommunale Wasserwerke Leipzig GmbH for the provision of data on the sewage system and the willingness for allowing sampling in their facilities. The authors gratefully acknowledge all colleagues at the Helmoltz Centre for Environmental Research - UFZ from the research cluster “Risk of micropollutants in water and soil in the urban environment”. Ronald Krieg, Mathias Falke and Christiane Neumann are thanked for the help in the field.
Note Added after ASAP Publication Due to a production error, this paper published ASAP May 28, 2010 with errors in Table 1; the corrected version published ASAP June 3, 2010.
Supporting Information Available Additional details, figures, and tables. This material is available free of charge via the Internet at http://pubs.acs.org.
Literature Cited (1) Schwarzenbach, R. P.; Escher, B. I.; Fenner, K.; Hofstetter, T. B.; Johnson, C. A.; von Gunten, U.; Wehrli, B. The challenge of micropollutants in aquatic systems. Science 2006, 313 (5790), 1072–1077. (2) Jobling, S.; Nolan, M.; Tyler, C. R.; Brighty, G.; Sumpter, J. P. Widespread sexual disruption in wild fish. Environ. Sci. Technol. 1998, 32 (17), 2498–2506. (3) Luckenbach, T.; Epel, D. Nitromusk and polycyclic musk compounds as long-term inhibitors of cellular xenobiotic defense systems mediated by multidrug transporters. Environ. Health Perspect. 2005, 113 (1), 17–24. (4) Fent, K.; Weston, A. A.; Caminada, D. Ecotoxicology of human pharmaceuticals. Aquat. Toxicol. 2006, 76 (2), 122–159. (5) Schirmer, M.; Strauch, G.; Schirmer, K.; Reinstorf, F. Urban hydrogeology - Challenges in research and practice. Grundwasser 2007, 12 (3), 178–188. (6) Reinstorf, F.; Strauch, G.; Schirmer, K.; Gla¨ser, H. R.; Mo¨der, M.; Wennrich, R.; Osenbru ¨ ck, K.; Schirmer, M. Mass fluxes and spatial trends of xenobiotics in the waters of the city of Halle, Germany. Environ. Pollut. 2008, 152 (2), 452–460. (7) Barber, L. B.; Murphy, S. F.; Verplanck, P. L.; Sandstrom, M. W.; Taylor, H. E.; Furlong, E. T. Chemical loading into surface water along a hydrological, biogeochemical, and land use gradient: A holistic watershed approach. Environ. Sci. Technol. 2006, 40 (2), 475–486. (8) Ternes, T. A. Occurrence of drugs in German sewage treatment plants and rivers. Water Res. 1998, 32 (11), 3245–3260. (9) Buerge, I. J.; Poiger, T.; Mu ¨ ller, M. D.; Buser, H. R. Combined sewer overflows to surface waters detected by the anthropogenic marker caffeine. Environ. Sci. Technol. 2006, 40 (13), 4096– 4102. 4882
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 13, 2010
(10) Benotti, M. J.; Brownawell, B. J. Distributions of pharmaceuticals in an urban estuary during both dry- and wet-weather conditions. Environ. Sci. Technol. 2007, 41 (16), 5795–5802. (11) Phillips, P.; Chalmers, A. Wastewater effluent, combined sewer overflows, and other sources of organic compounds to Lake Champlain. J. Am. Water Resour. Assoc. 2009, 45 (1), 45–57. (12) Fenz, R.; Blaschke, A. P.; Clara, M.; Kroiss, H.; Mascher, D.; Zessner, M. Quantification of sewer exfiltration using the antiepileptic drug carbamazepine as marker species for wastewater. Water Sci. Technol. 2005, 52 (9), 209–217. (13) Greskowiak, J.; Prommer, H.; Massmann, G.; Nu ¨ tzmann, G. Modeling seasonal redox dynamics and the corresponding fate of the pharmaceutical residue phenazone during artificial recharge of groundwater. Environ. Sci. Technol. 2006, 40 (21), 6615–6621. (14) Massmann, G.; Su ¨ ltenfuss, J.; Du ¨ nnbier, U.; Knappe, A.; Taute, T.; Pekdeger, A. Investigation of groundwater a residence times during bank filtration in Berlin: multi-tracer approach. Hydrol. Processes 2008, 22 (6), 788–801. (15) Musolff, A.; Leschik, S.; Reinstorf, F.; Strauch, G.; Schirmer, M. Assessing emerging contaminants - case study of the city of Leipzig, Germany. In Groundwater Quality: Securing Groundwater Quality in Urban and Industrial Environments; Trefry, M. G., Ed.; IAHS Publication 324, IAHS Press, Wallingford, UK, pp 178-185. (16) Leschik, S.; Musolff, A.; Martienssen, M.; Krieg, R.; Bayer-Raich, M.; Reinstorf, F.; Strauch, G.; Schirmer, M. Investigation of sewer exfiltration using integral pumping tests and wastewater indicators. J. Contam. Hydrol. 2009, 110 (3-4), 118–129. (17) Lerner, D. N. Identifying and quantifying urban recharge: a review. Hydrogeol. J. 2002, 10 (1), 143–152. (18) Musolff, A. Micropollutants - challenges in hydrogeology. Hydrogeol. J. 2009, 17 (4), 763–766. (19) Musolff, A.; Leschik, S.; Mo¨der, M.; Strauch, G.; Reinstorf, F.; Schirmer, M. Temporal and spatial patterns of micropollutants in urban receiving waters. Environ. Pollut. 2009, 157 (11), 3069–3077. (20) Brombach, H. Sewerage and storm water treatment facilities in Germany as mirrored by statistics. Korresp. Abwasser 2010, 57 (1), 28–36. (21) Jin, S.; Yang, F.; Liao, T.; Hui, Y.; Xu, Y. Seasonal variations of estrogenic compounds and their estrogenicities in influent and effluent from a municipal sewage treatment plant in China. Environ. Toxicol. Chem. 2008, 27 (1), 146–153. (22) Ho¨hne, C.; Pu ¨ ttmann, W. Occurrence and temporal variations of the xenoestrogens bisphenol A, 4-tert-octylphenol, and tech. 4-nonylphenol in two German wastewater treatment plants. Environ. Sci. Pollut. Res. 2008, 15 (5), 405–416. (23) Buerge, I. J.; Poiger, T.; Mu ¨ ller, M. D.; Buser, H. R. Caffeine, an anthropogenic marker for wastewater contamination of surface waters. Environ. Sci. Technol. 2003, 37 (4), 691–700. (24) Braun, P.; Mo¨der, M.; Schrader, S.; Popp, P.; Kuschk, R.; Engewald, W. Trace analysis of technical nonylphenol, bisphenol A and 17 alpha-ethinylestradiol in wastewater using solid-phase microextraction and gas chromatography-mass spectrometry. J. Chromatogr. A 2003, 988 (1), 41–51. (25) Einsle, T.; Paschke, H.; Bruns, K.; Schrader, S.; Popp, P.; Mo¨der, M. Membrane-assisted liquid-liquid extraction coupled with gas chromatography-mass spectrometry for determination of selected polycyclic musk compounds and drugs in water samples. J. Chromatogr. A 2006, 1124 (1-2), 196–204. (26) Brombach, H.; Wo¨hrle, C. Measured overflow activity of combined sewer overflows (CSO) tanks. Korresp. Abwasser 1997, 44 (1), 44–66. (27) Fono, L.; Sedlak, D. L. Use of the chiral pharmaceutical propranolol to identify sewage discharges into surface waters. Environ. Sci. Technol. 2005, 39 (23), 9244–9252. (28) Jonkers, N.; Kohler, H.-P. E.; Dammsha¨user, A.; Giger, W. Mass flows of endocrine disruptors in the Glatt River during varying weather conditions. Environ. Pollut. 2009, 157 (3), 714–723. (29) Barber, L. B.; Keefe, S. H.; Leblanc, D. R.; Bradley, P. M.; Chapelle, F. H.; Meyer, M. T.; Loftin, K. A.; Kolpin, D. W.; Rubio, F. Fate of sulfamethoxazole, 4-nonylphenol, and 17 beta-estradiol in groundwater contaminated by wastewater treatment plant effluent. Environ. Sci. Technol. 2009, 43 (13), 4843–4850. (30) Ying, G. G.; Toze, S.; Hanna, J.; Yu, X. Y.; Dillon, P.; Kookan, R. S. Decay of endocrine-disrupting chemicals in aerobic and anoxic groundwater. Water Res. 2008, 42 (4-5), 1133–1141. (31) Focazio, M. J.; Kolpin, D. W.; Barnes, K. K.; Furlong, E. T.; Meyer, M. T.; Zaugg, S. D.; Barber, L. B.; Thurman, M. E. A national reconnaissance for pharmaceuticals and other organic wastewater contaminants in the United States - (II) Untreated drinking water sources. Sci. Total Environ. 2008, 402 (2-3), 201–216.
(32) Conn, K. E.; Siegrist, R. L.; Barber, L. B.; Meyer, M. T. Fate of trace organic compounds during vadose zone soil treatment in an onsite wastewater system. Environ. Toxicol. Chem. 2010, 29 (2), 285–293. (33) Swartz, C. H.; Reddy, S.; Benotti, M. J.; Yin, H. F.; Barber, L. B.; Brownawell, B. J.; Rudel, R. A. Steroid estrogens, nonylphenol ethoxylate metabolites, and other wastewater contaminants in groundwater affected by a residential septic system on Cape Cod, MA. Environ. Sci. Technol. 2006, 40 (16), 4894–4902. (34) Clara, M.; Strenn, B.; Kreuzinger, N. Carbamazepine as a possible anthropogenic marker in the aquatic environment: investigations on the behaviour of Carbamazepine in wastewater treatment and during groundwater infiltration. Water Res. 2004, 38 (4), 947–954. (35) Musolff, A.; Leschik, S.; Reinstorf, F.; Strauch, G.; Mo¨der, M.; Schirmer, M. Xenobiotics in groundwater and surface water of the city of Leipzig. Grundwasser 2007, 12 (3), 217–231.
(36) Crosbie, R. S.; Binning, P.; Kalma, J. D. A time series approach to inferring groundwater recharge using the water table fluctuation method. Water Resour. Res. 2005, 41 (1); DOI 10.1029/2004WR003077. (37) Ort, C.; Gujer, W. Sampling for representative micropollutant loads in sewer systems. Water Sci. Technol. 2006, 54 (6-7), 169–176. (38) Rutsch, M.; Rieckermann, J.; Krebs, P. Quantification of sewer leakage: a review. Water Sci. Technol. 2006, 54 (6-7), 135–144. (39) Chisala, B. N.; Lerner, D. N. Distribution of sewer exfiltration to urban groundwater. Proc. Inst. Civ. Eng. Water Manage. 2008, 161 (6), 333–341. (40) Karpf, C.; Traenckner, J.; Krebs, P. Hydraulic modelling of sewage exfiltration. Water Sci. Technol. 2009, 59 (8), 1559–1565. (41) Gurr, C. J.; Reinhard, M. Harnessing natural attenuation of pharmaceuticals and hormones in rivers. Environ. Sci. Technol. 2006, 40 (9), 2872–2876.
ES903823A
VOL. 44, NO. 13, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
4883
