Utilize gadolinium as environmental tracer for surface water-groundwater interaction in Karst

https://doi.org/10.1016/j.jconhyd.2020.103710Get rights and content

Abstract

The rare earth element (REE) group is widely used for geochemical prospection and the hydrochemical differentiation of waters. Most of the currently applied methods use normalized REE patterns to determine enrichments or depletions of certain REE in comparison to standard materials, which are caused by specific environmental conditions. Contrast agents containing Gadolinium (Gd), which are used for magnetic resonance imaging (MRI), have been emitted into surface waters since the 1980s. Patients excrete these contrast agents shortly after ambulant medication in hospitals or at home. Sewage treatment is currently unable to hold back Gd from this anthropogenic source. Therefore, the Gd concentration in the receiving channel increases significantly and creates a Gd peak in the REE pattern. This anthropogenic peak propagates into adjacent groundwater bodies. In a karst aquifer showing a connection between a river/ponor (input) and three springs (output), such an anthropogenic Gd anomaly has been traced, the local Gd background quantified, and surface water groundwater interaction evaluated. In two sampling campaigns, water samples were taken every day at the input and output side during one week in February and three weeks in May. Sampling of springs and brooks in the vicinity of the karst aquifer proved excessive Gd from anthropogenic sources. The evaluation of concentration, mass flow and total mass of Gd show that Gd can be an environmental tracer to monitor surface water-groundwater interaction as well as the anthropogenic influence on water bodies. Further anthropogenic pollutants - diclofenac, carbamazepine, galaxolide, caffeine, and acesulfame-K - representing different classes of common organic substances were tested as co-indicators to Gd. However, out of these only acesulfame-K was detected and is related with the Gd-anomaly. Our results indicate that Gd is a more powerful indicator of surface water- groundwater interaction than most organic pollutants.

Introduction

Karst aquifers are a major freshwater source around the world and the geomorphological and hydrogeological characteristics of karst create springs with relatively large discharges and large catchments. One geomorphological characteristic of karst are ponors (swallow holes), where river water descends partly or completely underground. In many cases, this water discharges in springs, directly connected to the upstream ponor (Ford and Williams, 2007; Henry and Suk, 2018). As the stream water descends through the ponor, the natural attenuation such as removal of suspended solids and pathogenic microbes, biodegradation and adsorption is less compared e.g. to bank filtration, because the groundwater flow in karst conduits is too rapid for these processes to be effective. Therefore, karst aquifers are very vulnerable to anthropogenic influences. The surface water might contain anthropogenic contaminants and without treatment poses a threat for the use of the spring water e.g. for the public water supply (Dew and Hötzl, 1999; Ford and Williams, 2007; Henry and Suk, 2018; White et al., 2018).

Since the 1980s, gadolinium (Gd) complexes are used as contrast agents in magnetic resonance imaging (MRI). Due to their increasing application, the Gd concentration levels in aquatic environments increase as well (Lawrence, 2010; Kulaksız and Bau, 2013; Tepe et al., 2014; Merschel et al., 2015; Hatje et al., 2016).

MRI uses very stable Gd complexes, which pass the human body un-metabolized. Depending on the complex, patients excrete the contrast agents containing Gd 1.5 to 30 h after injection (Idée et al., 2006). Conventional sewage systems cannot remove them from wastewater (Lawrence et al., 2009; Cyris, 2013, pp. 82). Even hospitals with special wastewater treatment emit Gd (Kümmerer and Helmers, 2000; Kümmerer, 2001), but as practitioners apply MRI-diagnostics as well, hospitals are not the only source of anthropogenic Gd emissions. Gd has become an indicator of municipal sewage in general (Verplanck et al., 2005; Verplanck et al., 2010; Birka et al., 2013, Birka et al., 2016).

Naturally, the rare earth elements (REE) are widespread in the environment and show patterns reflecting the element concentrations in rock and soil (McLennan, 1982; Henderson, 1984; Taylor and McLennan, 1985, Taylor and McLennan, 1995) and/or the environmental conditions of the analyzed material (Guo et al., 2010; Laveuf et al., 2012; Petrosino et al., 2013). Because of this characteristic, the REE are a tool in mineral exploration, rock formation determination (Hallis et al., 2014; Gaschnig et al., 2016) and hydrothermal process modeling (Göb et al., 2013; Chudaev et al., 2017).

These patterns are altered by the release of Gd-MRI contrast agents into surface waters via municipal sewage in densely populated areas as well as in rural areas because the patients excrete the Gd complex in the hospital or at home (Kümmerer and Helmers, 2000; Kümmerer, 2001). Seemingly, Gd emissions in the environment are related to point sources like discharge from wastewater treatment plants (WWTP). This possible source coincides with Gd-anomalies in the environment (Petrosino et al., 2013) and almost all WWTP in western countries are sources of Gd as the vast majority of people have access to MRI treatment (Cyris, 2013; Lindner, 2017). The emissions result in a widespread anthropogenic Gd anomaly of the REE pattern in surface water (Bau and Dulski, 1996; Möller et al., 2000; Kulaksız and Bau, 2011; Hatje et al., 2016). The Gd complexes are very stable in the environment leading to conservative behavior. In some cases, such anomalies were measured over long distances in rivers (Verplanck et al., 2005; Petelet-Giraud et al., 2009; Rabiet et al., 2009; Verplanck et al., 2010; Klaver et al., 2014) and the signal accumulates over time (Merschel et al., 2015; Hatje et al., 2016).

As the practitioners and the hospitals work mainly on weekdays, the anthropogenic Gd anomalies show a transient signal in the surface water as well as in adjacent groundwater influenced by bank filtration. In such coupled systems, it is possible to trace the surface water in the underground. Transit time is a key parameter for the aquifer system that can be calculated from water mixing and water flux data based on the Gd anomaly. Furthermore, transient signals in bank filtrate show how powerful Gd is as an environmental tracer for bank filtration and groundwater with surface water portions (Bichler et al., 2016; Brünjes et al., 2016).

In case of bank filtration for freshwater production, larger porous media aquifers and single wells, the anthropogenic Gd anomaly has been used as a tracer to support other tracer analysis and to calculate groundwater travel times (Johannesson et al., 1997; Massmann et al., 2007; Lawrence and Bariel, 2010; Schwesig and Bergmann, 2011).

Previous investigations of anthropogenic Gd anomalies use it as indicator for anthropogenic influence on surface water and groundwater in heavily polluted and densely populated areas where many point sources (WWTP) overlap each other and overprint the transient signals of weekdays or small emission changes. Dilution and adsorption effects might play a significant role as well but are not easy to evaluate. Although it has been shown, that not only WWTPs are unable to remove the anthropogenic Gd from the water cycle, neither can water treatment plants (WTP). The elevated Gd-concentration reaches even tap water for example in Berlin (Kulaksız and Bau, 2011; Tepe et al., 2014). There, artificial recharge from treated wastewater caused a Gd-anomaly in groundwater and raw water for drinking water supply (Massmann et al., 2004, Massmann et al., 2008; Tepe et al., 2014). Combining anthropogenic Gd as an environmental tracer with other anthropogenic organic pollutants proved to be a good approach in many cases where they show similar behavior. Especially acesulfame was used in several studies and supports results of Gd analyses at least over short infiltration distances (Engelhardt et al., 2014; Bichler et al., 2016); although its environmental stability is questionable (Kahl et al., 2018).

This study comprises the situation of a cuesta in Germany, where the local river descends partly into a karst aquifer by a set of ponors at the riverbank and contributes to the discharge of three interconnected freely draining springs.

The main objective of this research is to show the environmental tracer capabilities of Gd over long distances in a karst aquifer, which exhibits high flow velocities between surface water input and output. As the chosen aquifer, system was investigated already in the 1990s using dye tracers, conclusions from the Gd signal can be evaluated by the results of these earlier studies. Additionally, the study site offers a versatile application of Gd as an environmental tracer. Furthermore, overall and water type specific geogenic background values are calculated from the many data of this study applying a statistical method agreed by the environmental and geological surveys of the German Laender and the Federal Republic.

The study site is located (Fig. 1 upper left: black rectangle in the map of Germany) in a Mesozoic sedimentary rock sequences of Germany. The land use of the area itself is dominated by agriculture. Forests occupy higher plateaus, which are separated from each other by narrower and wider valleys of the local rivers and creeks. As a result, the surface waters create different catchments on the surface by eroding the Triassic platform and forming cuestas, with separate groundwater bodies. The settlements locate mainly on the riverbanks in the broader valleys (Fig. 1). Just a few small villages with up to a few hundred inhabitants reside on the plateaus. Around 15,000 people live within the study area covering roughly 30 km2. An important geomorphologic feature is the river “Streu” (blue line in Fig. 1) flowing from the village “Nordheim” in the northwest to “Unsleben” in the south, traveling through “Mellrichstadt” and “Mittelstreu”, where the springs are located “Mittelstreuer Quellen”. The “Streu” embraces a central plateau, which is the primary recharge area for the local aquifer and the springs (Fig. 1).

At the study site, sedimentary rocks of lower Mesozoic (Triassic) age form the main geological units. The lowermost members are sandstones with claystones at the top of the “Buntsandstein” forming the regional base of the aquifer system of interest. Limestones of the Middle Triassic “Muschelkalk” are the next overlying unit. The lower and middle members of the “Muschelkalk” form a landscape with plateaus thinly covered with soil. In the valleys and close to the rivers, fluvial sediments cover the Triassic rocks. The overall dip of the sedimentary rocks is to the southwest. Up to 140 m thick limestones are the main aquifer of the study site. A few wells drilled on the plateaus indicate the karst water table at 40 to 60 m depth below the surface (Reder and Parchwitz, 2013). In the study area, the “Streu” has eroded into the karstified limestones and seeps partially into several ponors (swallow holes) marked as “input” in Fig. 1. Former dye tracer tests confirmed connecting karst conduits to the main springs of the study area, i.e. the “Mittelstreuer Quellen” marked as “output” in the southern section in Fig. 1. Earlier investigations including tracer tests confirm groundwater flow velocities between 125 and 3000 m/d (Hofmann et al., 1991; Reder and Parchwitz, 2013). The distance between ponors and springs is approximately 7500 m. The “Mittelstreuer Quellen” are three interconnected karst springs spreading over approximately 50 m along the hillside toe. All three are used for the local drinking water supply. One pipe collects the discharge of all three springs and directs it to the water treatment plant (WTP) approximately 700 m southeast of the springs (Fig. 1). The sampled water represents mixed spring water (Hofmann et al., 1991; Reder and Parchwitz, 2013).

Section snippets

Sampling

At three different sampling campaigns in 2017 and 2018, water samples were taken at the study. First, in summer 2017 six samples were taken at different locations: at a ponor of the “Streu” near the village of “Stockheim” (Fig. 1), at the WTP, before and after the treatment process, and at the three individual springs. The Bavarian Environmental Protection Agency (Bavarian EPA) mirrored this sample collection and analyzed the samples in its own laboratory in parallel. At the river's ponor we

Results and discussion

The results consist of three data types: first, the REE-concentrations at the two sampling sites (ponor and springs) in summer and winter, second, the REE-concentrations of the samples in the wider study area, which define the more detailed geogenic background concentrations of Gd, and third, the analysis of five organic pollutants, which have the same emission path (wastewater) in the study area as Gd. The second type includes the spring discharge measurements and the calculation of realistic

Conclusion

The study proves the comprehensive applicability of Gd as environmental tracer for the interaction of surface water and groundwater in karst. It can be a tool to investigate or monitor wastewater influences on water bodies in general. Hence, it is important to know the geogenic background of Gd in the water types and aquifer systems in question. The statistical calculation of the background using probability plots combines statistics and expert knowledge of the geological and environmental

Acknowledgements

Many thanks to Mr. Thomas Walter of the Environmental Protection Agency of the Saarland for providing the Excel based “Probnet” for background calculations and the time for introducing it to us. We acknowledge the support of the municipal water supplier (WZV Mellrichstädter Gruppe) and the local authorities (Verwaltungsgemeinschaft Mellrichstadt, Bürgermeister Streit) in maintaining the automated sampling equipment and the great help during the fieldwork as well as the constant interest in the

References (71)

  • S. Göb et al.

    Major and trace element compositions (including REE) of mineral, thermal, mine and surface waters in SW Germany and implications for water–rock interaction

    Appl. Geochem.

    (2013)
  • H. Guo et al.

    Geochemical controls on arsenic and rare earth elements approximately along a groundwater flow path in the shallow aquifer of the Hetao Basin, Inner Mongolia

    Chem. Geol.

    (2010)
  • L.J. Hallis et al.

    Trace-element modelling of mare basalt parental melts - implications for a heterogeneous lunar mantle

    Geochim. Cosmochim. Acta

    (2014)
  • C. Hissler et al.

    Rare earth elements as hydrological tracers of anthropogenic and critical zone contributions: a case study at the Alzette River basin scale

    Proc. Earth Planet. Sci.

    (2014)
  • K.H. Johannesson et al.

    Rare earth elements as geochemical tracers of regional groundwater mixing

    Geochim. Cosmochim. Acta

    (1997)
  • G. Klaver et al.

    Anthropogenic rare earth element in rivers: gadolinium and lanthanum. Partitioning between the dissolved and particulate phases in the Rhine River and spatial propagation through the Rhine-Meuse Delta (the Netherlands)

    Appl. Geochem.

    (2014)
  • S. Kulaksız et al.

    Anthropogenic gadolinium as a microcontaminant in tap water used as drinking water in urban areas and megacities

    Appl. Geochem.

    (2011)
  • S. Kulaksız et al.

    Anthropogenic dissolved and colloid/nanoparticle-bound samarium, lanthanum and gadolinium in the Rhine River and the impending destruction of the natural rare earth element distribution in rivers

    Earth Planet. Sci. Lett.

    (2013)
  • K. Kümmerer

    Drugs in the environment: emission of drugs, diagnostic aids and disinfectants into wastewater by hospitals in relation to other sources – a review

    Chemosphere

    (2001)
  • C. Laveuf et al.

    The impact of redox conditions on the rare earth element signature of redoximorphic features in a soil sequence developed from limestone

    Geoderma

    (2012)
  • M.G. Lawrence

    Detection of anthropogenic gadolinium in the Brisbane River plume in Moreton Bay, Queensland, Australia

    Mar. Pollut. Bull.

    (2010)
  • M.G. Lawrence et al.

    Tracing treated wastewater in an inland catchment using anthropogenic gadolinium

    Chemosphere

    (2010)
  • M.G. Lawrence et al.

    Detection of anthropogenic gadolinium in treated wastewater in south East Queensland, Australia

    Water Res.

    (2009)
  • H. Liu et al.

    Rare earth elements as indicators of groundwater mixing in the North China plain: a case study in the area of Hengshui City, China

    Proc. Earth Planet. Sci.

    (2017)
  • S.M. McLennan

    On the geochemical evolution of sedimentary rocks

    Chem. Geol.

    (1982)
  • G. Merschel et al.

    Tracing and tracking wastewater-derived substances in freshwater lakes and reservoirs - anthropogenic gadolinium and geogenic REEs in Lake Paranoá, Brasilia

    Compt. Rendus Geosci.

    (2015)
  • P. Möller et al.

    Anthropogenic gadolinium as a conservative tracer in hydrology

    J. Geochem. Explor.

    (2000)
  • E. Petelet-Giraud et al.

    Natural versus anthropogenic sources in the surface- and groundwater dissolved load of the Dommel river (Meuse basin)

    J. Hydrol.

    (2009)
  • P. Petrosino et al.

    REE contents in solid sample media and stream water from different geological contexts

    J. Geochem. Explor.

    (2013)
  • M. Rabiet et al.

    Positive gadolinium anomalies in wastewater treatment plant effluents and aquatic environment in the Herault watershed (South France)

    Chemosphere

    (2009)
  • A.J. Sinclair

    A fundamental approach to threshold estimation in exploration geochemistry: probability plots revisited

    J. Geochem. Explor.

    (1991)
  • N. Tepe et al.

    High-technology metals as emerging contaminants: strong increase of anthropogenic gadolinium levels in tap water of Berlin, Germany, from 2009 to 2012

    Appl. Geochem.

    (2014)
  • A. Bichler et al.

    Quantification of river water infiltration in shallow aquifers using acesulfame and anthropogenic gadolinium

    Hydrol. Process.

    (2016)
  • M. Cyris

    Behavior of Gadolinium-based Diagnostics in Water Treatment

    (2013)
  • D. Dew et al.

    Karst Hydrogeology and Human Activities: Impacts, Consequences, and Implications

    (1999)
  • Cited by (0)

    View full text