Seasonal variations of rare earths and yttrium distribution in the lowland Havel River, Germany, by agricultural fertilization and effluents of sewage treatment plants

Applied Geochemistry 41 (2014) 62–72

Contents lists available at ScienceDirect

Applied Geochemistry journal homepage: www.elsevier.com/locate/apgeochem

Seasonal variations of rare earths and yttrium distribution in the lowland Havel River, Germany, by agricultural fertilization and effluents of sewage treatment plants Peter Möller a,⇑, Andrea Knappe b, Peter Dulski a a b

Helmholtz-Zentrum Potsdam, Deutsches GeoForschungsZentrum, Telegrafenberg, 14473 Potsdam, Germany Freie Universität Berlin, Dep. Earth Sciences, 12249 Berlin, Germany

a r t i c l e

i n f o

Article history: Received 29 April 2013 Accepted 19 November 2013 Available online 3 December 2013 Editorial handling by Philippe Negrel In memory of Asaf Pekdeger

a b s t r a c t REE and Y (REY) distribution in the lowland Havel River passing the Federal State of Berlin, Germany, depends on contributions of point sources such pharmaceutical and high-tech industries, acid water from the open pit lignite mining, and medical application of very stable organic Gd chelates. Another omnipresent dispersed source of REY are water-soluble Ca-phosphates containing micro-amounts of Eu(II)-bearing barite as components of common agricultural fertilizers. After distribution in the field during the cold season (October through March) these Ca-phosphates dissolve and secondary phosphates and calcite precipitate both being enriched in light REE. Heavy REE are preferably exported by runoff together with part of the micro-contaminant barite leading to high Yb/Nd ratios in the Havel water and REY distribution patterns with only small Eu deficits. During the warm season (April through September) light REE together with phosphate are leached from secondary soil minerals by runoff. The micro-component barite is retained in vegetation-covered soil. Thus, REY patterns of Havel water show significant Eu deficits. The high Gd anomalies result from medical applications of Gd-chelates which after urination pass the sewage treatment plants. The seasonal variations of total Gd in the Havel River are artifacts based on seasonal locally varied discharge of effluents from sewage treatment plants. The natural Gd concentration of 8 pmol/l in the northern Havel is enhanced to 3300 pmol/l, when the Havel River leaves Berlin territory. The elimination of phosphate from Lake Tegel water affects the fractionation of REE but not the concentration of total Gd. Although enhanced in total phosphorus (TP), the REE concentrations in the water from the Spree River and the Teltow Canal are less than in the Havel water before their confluence. Only Yb and Lu do not decrease. The contributors of the Havel River are high in total organic carbon (TOC) and dissolved organic carbon (DOC) compared to the Havel water before their indicating that REY are preferentially sorbed by settling organic matter. Applying PHREEQC and assuming that only 10% of TP is present as ortho-phosphate yields that only carbonate complexes are essential. Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction The naturally occurring suite of the rare earth elements La to Lu is ubiquitous in hydrological systems. Including Y this group of elements is henceforth presented as REY. To a substantial extent their distribution in the hydrosphere depends on (i) formation of soluble chemical complexes, (ii) sorption onto surfaces of crystalline and amorphous matter (Dia et al., 2000; Ingri et al., 2000; Johannesson et al., 2004; Tweed et al., 2006; Tang and Johannesson, 2006; Willis and Johannesson, 2011), (iii) coprecipitation with alteration products, (iv) scavenging by Fe–Al oxyhydroxides (Bau, 1999; Ingri et al., 2000; Ohta and Kawabe, 2001; Quinn et al., 2004, 2006), and (v) concentration in and later release from the biosphere (Stille ⇑ Corresponding author. Tel.: +49 03312881430. E-mail addresses: [email protected] (P. Möller), [email protected]. de (A. Knappe), [email protected] (P. Dulski). 0883-2927/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.apgeochem.2013.11.011

et al., 2006a, 2006b; Sun et al., 1997; Takahashi et al., 2005, 2007; Texier et al., 1999). Besides the inorganic ligands such as carbonates and phosphates, humates in vegetation-covered catchments are considered a particularly important category of ligands because of their strong affinity for metals and their ubiquity in soils, sediments, pore-, ground- and surface water (Dupre et al., 1999; Lead et al., 1998; Willis and Johannesson, 2011). Due to the interaction with bedrocks and sediments (Banks et al., 1999; Johannesson et al., 2000; Zhang et al., 2006) the REY abundance in river and lake water shows seasonal and temporal variations (Goldstein and Jacobson, 1988; Ingri et al., 2000; Leybourne and Johannesson, 2008; Sholkovitz, 1992, 1995). These variations are considered to reflect different types and rates of water/rock interactions. REE patterns of groundwaters from limestones resemble each other (Johannesson et al., 1997, 1999, 2000; Möller et al., 2003; Smedly, 1991). In groundwater from sandstones (Möller et al., 2003), basalts (Möller et al., 2003; Paces

P. Möller et al. / Applied Geochemistry 41 (2014) 62–72

et al., 2001) and granites (Möller et al., 1997) the REE patterns differ from those of the aquifer rocks because of incongruent dissolution of REE-bearing major, minor and tracer minerals (Möller, 2000). Anomalous Gd concentrations are recognized worldwide in rivers, coastal seawater, ground- and tap water because very stable organic Gd chelates of various chemical compositions but of similar chemical properties (Brücher, 2002) are applied in hospitals and by private radiologists as contrast enhancement agents in magnet resonance imaging (IMR) of the blood system since 20 years (Bau et al., 2006; Bau and Dulski, 1996; Elbaz-Poulichet et al., 2002; Hennebrüder et al., 2004; Knappe et al., 1999, 2005; Kulaksiz and Bau, 2007, 2011, 2013; Lawrence et al., 2006, 2009; Möller et al., 2000, 2002; Nozaki et al., 2000; Petelet-Giraud et al., 2009; Rabiet et al., 2009; Verplanck et al., 2005; Zhu et al., 2004). These Gd chelates are neither removed during sewage treatment nor by sorption onto particulate matter in surface water. They survive for months unchanged in the hydrosphere (Dulski et al., 2011). Recently anomalous concentrations of La and Sm were reported in the Rhine River, Germany, (Kulaksiz and Bau, 2011, 2013). The usage of REE in the high-tech products and their increase in garbage and industrial effluents will be responsible for long-term increase of REY in the hydrosphere. Another important dispersed source are agricultural fertilizers which often contain REY contents of up to 1400 lg/g depending on their source material (Otero et al., 2005). Being produced from marine phosphorites the widespread compound NPK (nitrogen– phosphorus–potassium) fertilizers are rich in REY. The runoff from fertilized areas has to be considered as an omnipresent dispersed source of REY in the countryside. Different from industrial and medical sources, agricultural fertilization is bound to defined seasons of the year. Thus, the pollution by fertilizer is expected to create a seasonal trend of REY in runoff and surface water. The aim of this exemplary study is to search for such seasonal variation of REY abundance in the lowland Havel River when passing the Federal State of Berlin, Germany (Fig. 1). In such a low-flux system the seasonal variations of REE abundance are expected to be higher than in fast running rivers because of enhanced environmental interactions.

2. Hydrology of the study area The lowland rivers Havel and Spree drain the Pleistocene countryside of Berlin, Germany, under conditions of a continental climate with annual precipitation of 500–600 mm. The Havel River passes the western part of the study area from north to south, whereas the Spree River flows east to west through the City of Berlin (insert of Fig. 1). Both rivers follow glacial troughs filled by Late Pleistocene sediments. Their drainage basins are either covered by mostly pine forests or are used for cattle breeding and agriculture on poor gley and sand-loam soils. Part of the Spree water originates from open pit lignite mining. Based on different hydrological conditions the Havel River is subdivided into three sections (Fig. 1). The water chemistry of the Havel in Section 1 is dominantly controlled by the drainage basin north of Berlin with its wood-, grass- and farm land and to minor extend by the effluent of the stratified Lake Tegel. Lake Tegel receives contributions from the Nordgraben with its high organic and phosphorus loads from the sewage treatment plant STP-1, the Tegel-Fließ and some Havel water supplied by a pipeline. This water mix passes the phosphate elimination plant (PEP; Fig. 1) which was installed to reduce the eutrophication of the lake. By sorption onto FeOOH precipitates total phosphorus (TP) was reduced to levels of 8–20 lg/l (Schauser and Chorus, 2007). Due to

63

abstraction by bank filtration for production of drinking water Lake Tegel discharges less than 0.6 m3/s into the Havel River. In Section 2 downstream the lock Spandau (S-6; Fig. 1), the Havel water is mixed with a similar amount of Spree water which is strongly affected by effluents from STPs 2, 5, 6 and to a less degree from STP-1, water from the pharmaceutical industries from within the city of Berlin, and water from lignite mining before entering the study area. At the location S-20 the effluent from STP-2 is not included. In Section 3 the Teltow Canal collects the effluents from STPs 3, 4, 5 and 6, and water from the lignite open pits via the river Spree. STP-2 only discharges its effluent during March through September via a pipe line into the Teltow Canal. In October to March the effluents are discharged into the River Spree (insert of Fig. 1). 3. Sampling From May 2001 until December 2002 about monthly sampling of surface water was performed at 17 sampling locations (S) along the Havel River and its main contributors (Fig. 1). At all sampling points about 3 l of water were collected in polytetrafluorethylene (PTFE) bottles. Within 1–2 days these samples were filtered through encapsulated cellulose-acetate filter of 0.2 lm (SartobranÒ, Sartorius) into PTFE bottles using a peristaltic pump. The filtrate was acidified to pH 2 and spiked with 1 ml containing 100 ng/g of both Pr and Tm. Aliquots of 10 ml of the spiked water were used for direct measurements of the spikes Pr, Tm and of total Gd. All major and minor dissolved species, total phosphorus (TP), dissolved organic carbon (DOC) and total organic carbon (TOC) were analyzed by the Civil Service of Berlin in separate samples. Samples from Sections 1, 2 and 3 are characterized by numbers S-1 – S-10, S-20 – S-24 and S-30 – S-31, respectively. 4. Analytical procedure of REE The conditioned water was passed through a Sep-Pac C18Ò column (Water Corp., USA) preconditioned with a mixture of ethylhexyl-phosphates to collect rare earth elements (REE) and yttrium. The flow rate was 1 l per hour. Later, the columns were washed with 50 ml of sub-boiled 0.01 N HCl. REE were eluted with 40 ml of ultra-pure 6 M HCl. The eluates were evaporated to incipient dryness. Each of the residues was subsequently dissolved in 1 ml of 6 N ultra-pure HNO3, transferred to a volumetric flask, where 1 ml of 100 lg/g Re and Ru spike were added for internal shift corrections and filled up to 10 ml by 0.5 M sub-boiled HCl. ICP-MS was used to determine REY. The detailed procedure of measurements and the corrections for molecular ion interferences are given by Dulski (1994) and Bau and Dulski (1996). All samples were measured repeatedly by ICP-MS. Blanks were determined in 0.5 M HCl. The calibration solutions were prepared from stock solutions diluted by ultra-pure water to 0.5 N acidity. All measurements of the 350 water samples were corrected for the recovery of Pr and Tm spikes. The average efficiencies of Pr and Tm in the eluates were (94 ± 6)% and (90 ± 6)%, respectively. The Tm spikes were also high enough to be measured directly with recoveries of (95 ± 6)%, respectively. 5. Distribution of REE 5.1. REE patterns The PAAS-normalized REE patterns of water from all sampling points along the Havel River (of which only 5 are presented in Fig. 2) show decreasing seasonal variations and declining spread of abundance of light REE downstream. In Section 1 sample S-2

64

P. Möller et al. / Applied Geochemistry 41 (2014) 62–72

Fig. 1. Geographic sketch map of the Havel River passing the Federal State of Berlin, Germany, showing the locations of sampling (given by S-numbers) and the sewage treatment plants (STP) in the insert. Arrows indicate the flow directions. PEP = phosphate elimination plant. Countries surrounding Germany are given by their international codes: A Austria; B Belgium; CH Switzerland; CS Czech Republic; DK Denmark; F France; L Luxembourg; NL Netherlands; PL Poland.

and S-4 show enhanced spread of the light REE (LREE) compared with that of the heavy REE (HREE). Anomalous Eu and Gd vary more during the warmer than the cooler season (Fig. 2a and b). The Eu deficiencies are small to negligible during the cold season. The representatives of light and heavy REE, Nd and Yb, vary inversely (Fig. 3a). Thus the REE fractionation defined by the ratio of Yb/Nd is high and low during the cold and warm seasons, respectively, and is positively correlated with the runoff of the northern drainage area which is considered to be about half of the recorded flux of combined Havel and Spree water at the location S-21 (Fig. 3b). The REE patterns of S-8 – S-10 increase by 1–2 units from La to Lu with highly anomalous Gd (Fig. 4). After passing the phosphate elimination plant (PEP) the water mix of Tegel Fließ, Nordgraben and Havel water (S-9 and S-10) shows negligible seasonal variations of REE patterns and plot mostly between the patterns of the seasonally varied entries of S-8. Total Gd concentrations are

nearly unaffected by the PEP procedure. Immediately after passing PEP Yb/Nd ratios scatter and are up to ten times the ratio of the Havel water of Section 1 (S-9 in Fig. 3c). These high ratios decline within Lake Tegel and approach those of the Havel water as shown by water from S-10 and may be due to REY-bearing particles passing the 0.2 lm filter. In Section 2 the REE fractionation in sample S-21 during the cold and warm seasons is less than in Section 1 and is comparable with that in the Spree water (S-20; Fig. 2c and d). This type of REE fractionation is implanted on the Havel water S-21 in which the Spree water accounts for 40–50%. Both Havel (S-21) and Spree water (S-20) show less variations of Yb/Nd molar ratios (Fig. 3d) than the Havel water in Section 1 and are less related to the seasonal trend of runoff (Fig. 3b). In Section 3 the Yb/Nd ratio the water at S-31 is strongly enhanced by mixing with the heavily polluted water from the Teltow Canal (S-30) with its enhanced Yb/Nd ratios during the warm season (Fig. 3d).

P. Möller et al. / Applied Geochemistry 41 (2014) 62–72

65

Fig. 3. Temporal changes of REE at various locations along the Havel River. (a) Temporal changes of Nd and Yb abundances in Section 1; (b) REE fractionation defined by the molar ratio of Yb/Nd in Havel water of Section 1 and the flux recorded at S-21; (c) Yb/Nd ratios in Lake Tegel water after passing the phosphate elimination plant; (d) Yb/Nd ratios of water from the rivers Spree (S-20) and Havel (S-21; S-31) and the Teltow Canal (S-30) downstream the confluence of the Havel and Spree Rivers. Parallel to the abscissa the warm seasons are indicated by bars. Note the different scales of the Yb/Nd-axis. Flux data were supplied on request by the Civil Service ‘‘Berliner Senatsverwaltung für Stadtentwicklung und Umwelt, Abteilung VIII, Referat VIII E’’.

Fig. 2. Spread of PAAS normalized REE patterns of Havel and Spree water during the warm (light grey) and cold (dark grey) annual seasons at various sampling locations along the Havel River from N to S (a, b, d and e) and the Spree River (c: S-20) near the confluence. PAAS after McLennan (1989).

5.2. Eu anomalies The Eu anomalies, Eu/Eu* (Eq. (1)), develop during the warm season but are less during the cold one in Section 1 (S-2 and S-4;

Fig. 5a). These anomalies are linearly correlated with the fractionation of REE presented by Yb/Nd ratio (Fig. 5b). The correlation of Eu anomalies and Yb/Nd ratios in Lake Tegel water (S-9) differs from those of the Havel water of Section 1 (S-2 and S-4) which may be due to the process of phosphate elimination. In Sections 2 and 3 the Eu anomalies in S-21 and S-31 are strongly influenced by water from the River Spree (S-20) and the Teltow Canal (S-30), respectively. The trends of Eu/Eu* in S-20 and S-21 is rather similar in Fig. 5a and b. Eu/Eu* in S-30 is nearly constant but associated with significant variations of Yb/Nd ratios.

P. Möller et al. / Applied Geochemistry 41 (2014) 62–72

Log REE, PAAS normalized

66

-4

Gd** depends on the discharge from STP-2 which is either to the River Spree or the Teltow Canal during the cold and warm season, respectively, which is is reflected in the inverse trends of Gd** in S-21 and S-31 (Fig. 6).

-5

Gd ¼ Gdtot  Gd

-6

Gd ¼ Gdsn  ðSmsn  Dysn Þ0:5

S-8; warm season S-9; cold season S-10; cold season

S-8; cold season S-10; warm season

00





ð2Þ ð3Þ

-7

5.4. Temporal variations of dissolved major and minor species

-8 La

Ce

Pr

Nd Pm Sm Eu

Gd

Tb

Dy

Ho

Er

Tm Yb

Lu

Fig. 4. Comparison of REE distribution pattern of the effluent of the phosphate elimination plant (S-9) and the water from Lake Tegel after mixing during the warm and cold seasons (S-10). S-8 represents the most important inflow into Lake Tegel.

The dissolved species such as Cl, Ca, bicarbonate and sulfate scatter but do not show seasonal variations (Fig. 7). In Section 1, the least sewage-affected sample S-4 shows lowest concentrations of Cl, Ca, and SO2 4 . The inflows to Lake Tegel (S-7 and S-8) are high est in Ca; HCO 3 ; ðClÞ, and sulfate. Ca and HCO3 concentrations in S-10 are similar to those in S-4, whereas Cl and sulfate resemble those in S-21. All the Havel samples after the confluence with the Spree River are significantly enriched in Cl (Na) and sulfate. Cl in S-30 reaches values similar to those in S-8, which are the samples most polluted by effluents from STPs. Contrasting the dissolved inorganic species, pH, DOC and TOC in Havel and Spree water are high and low during the warm and cold seasons, respectively (Fig. 8). The trends of pH values and water temperature are similar but not well synchronized. Excepting S-10, TP is low and high during the cold and warm season, respectively (Fig. 9). In S-1 through S-6 the seasonal variations of TP are lowest. The STP-effluent-loaded Nordgraben (S-8) and Teltow Canal (S-30) rarely exceed 0.3 mg/l TP. Thus the high values in S-7 and S-24 indicate a high source of phosphate. After the elimination of phosphate the water from Lake Tegel (S-10) shows minima in June but enhanced concentrations during September–February which strongly deviates from the behavior of TP in all the other water samples (Fig. 9b). The saturation indices (SI) of calcite and barite throughout the year are approached by using the common domain program PHREEQC (Parkhurst and Appelo, 2010). As shown for Havel water (S-4) barite saturation follows closely the minima and maxima of the measured flux of the Havel River at the position of S-21 and follows inversely the trend of water temperature (Fig. 10). The calcite

Fig. 5. Temporal changes of Eu anomalies water from the rivers Havel and Spree, Lake Tegel and Teltow Canal. (a) Correlation of Eu/Eu* ratios with the recorded flux at S-21. It is supposed that the trend of the recorded flux is similar to the runoff of the individual catchments involved. (b) Correlation of Eu anomalies and REE fractionation given by molar Yb/Nd ratios. Parallel to the abscissa the warm seasons are indicated by bars. Flux data were supplied on request by the Civil Service ‘‘Berliner Senatsverwaltung für Stadtentwicklung und Umwelt, Abteilung VIII, Referat VIII E’’. 0:33

Eu=Eu ¼ Eusn =ðSm0:67 sn  Tbsn Þ

ð1Þ

Index sn = PAAS normalized (McLennan, 1989). 5.3. Gd anomalies The anthropogenic amount of Gd chelates Gd** is estimated by Eq. (2), in which the natural contribution Gd* is estimated according to Eq. (3). In Section 1 of the Havel River the estimated anthropogenic contributions are about 8 pmol/l and only sporadically up to 50 pmol/l (Fig. 6). The Gd** level in Spree water is about 100 pmol/l before STP-2 discharges its effluents during the cold season. Down-stream the confluent of the rivers Spree and Havel the anthropogenic Gd** range from 100 to 3000 pmol/l. In S-31

Fig. 6. Temporal changes of anthropogenic Gd input, Gd** in pmol/l at various location along the Havel River and its contributors. Note the logarithmic scale of the Y-axis. Parallel to the abscissa the warm seasons are indicated by bars which correspond to the periods of discharging effluents from STP-2 into the Teltow Canal.

67

P. Möller et al. / Applied Geochemistry 41 (2014) 62–72 S-4 S-10 S-30

S-7 S-20

S-4 S-10 S-30

S-8 S-21

Ca

140

Concentration mg/l

Concentration mg/l

S-8 S-21

140

160

120 100 C

80 60

a

40 0

Cl

120 100 80 C

60 40

b

20 3

6

9

12

15

18

21

24

0

3

6

9

12

15

18

21

24

15

18

21

24

200

350 300

Concentration mg/l

Concentration mg/l

S-7 S-20

250

HCO 3200 150

c

100 0

160

120

SO 4280

d

40 3

6

9

12

15

18

21

24

Months after 1.1.2001

0

3

6

9

12

Months after 1.1.2001

2 Fig. 7. Temporal variations of (a) Ca, (b) Cl, (c) HCO 3 and (d) SO4 at various locations along the Havel River and its contributors. All data are supplied on request by the Civil Service ‘‘Berliner Senatsverwaltung für Stadtentwicklung und Umwelt, Abteilung VIII, Referat VIII E’’.

saturation indices reach their maxima with increasing temperature and have their minima at highest runoff. Gypsum is unsaturated all the time. 5.5. Downstream variations of dissolved species Using the averages of the years 2001 and 2002 significant spatial variations along the Havel River occur due to the confluence with the Spree River at Havel-km 10.3 and the Teltow Canal at Havel-km 28.9 (Fig. 11a) All major dissolved species increase because the salinity of the Spree River and the Teltow Canal are enhanced by effluents of several STPs (cf. insert of Fig. 1) and by water from the open pits of lignite mines as evident from the enhanced sulfate concentration (Fig. 7d). The downstream variations of REY abundance (Fig. 11b) reveal that LREE (represented by Nd) and HREE (Dy, Er) decrease from north to south, whereas Yb and Lu slightly increase. The small contribution of water from Lake Tegel slightly increases REY abundance in the Havel River but significantly increase total Gd (Fig. 11c). Total Gd increases in Section 1 due to mixing with water from Lake Tegel with its high Gd anomaly (Fig. 4) and in Section 2 after the confluence with the Spree River which is due to the many effluents of STPs discharging into this river (Fig. 1). In Section 2 total Gd shows high variations caused by the seasonally varied discharge of effluents from STP-2 into the Spree River from October through March and from April through September into the Teltow Canal. Due to incomplete mixing of Teltow Canal and Havel water at this sampling point seasonally different concentrations in S-31 at Havel-km 28.9 (cf. Fig. 6) are found. 6. Discussion 6.1. Impact of environmental parameters on REE abundance Important seasonal changes are temperature and precipitation affecting both runoff and vegetation. The seasonal variation of

water temperature affects the growth of algae, phytoplankton and zooplankton and is positively correlated with pH (Fig. 8) and TP (Fig. 9). This has some impact on REE distribution by competing with both inorganic and organic complexation (Johannesson et al., 2004; Stille et al., 2006a, 2006b; Takahashi et al., 2005, 2007). The annual trends of runoffs from the different catchments are considered to resemble the trend of the flux at sampling point S-21 after the confluence of the rivers Havel and Spree. Runoff and water temperature are anti-correlated. The runoff in Section 1 is positively correlated with REY fractionation (Fig. 3b), Eu anomalies (Fig. 5a) and saturation of barite SIbarite but less with SIcalcite (Fig. 10). In Sections 2 and 3 these correlations diminish because of mixing with the highly polluted water from the Spree River and the Teltow Canal. 6.2. REE mobilization in soil The correlation of REE fractionation with runoff in Section 1 suggests that leaching REY from soil may play an important role. Indeed, the largest pools of REE in cultivated soils are water-soluble Ca(H2PO4)2 and CaHPO4 mineral components of artificial fertilizers and their secondary soil minerals. The source of REY in the commercial compound fertilizers of the type NPK (nitrogen-phosphorus-potassium) is the high natural REY abundance in phosphorites from which most of the water-soluble phosphates are produced by phosphoric acid treatment. Otero et al. (2005) reported trace elements including some REY in NPK fertilizers containing ortho- and polyphosphates in the range of 2–7% P2O5 (Fig. 12). In general, the REY abundance of these phosphates is about one order of magnitude lower than in their raw material but with largely similar REE patterns (Fig. 12). The P/RREY molar ratios in such fertilizers and the annual average in surface water are about 2300 and 10,000, respectively, showing that REE are less leached from soil than phosphates. Tyler (2004) and Zhang et al. (2006) therefore argued that such artificial fertilizers are ‘‘spoiling

68

P. Möller et al. / Applied Geochemistry 41 (2014) 62–72

Fig. 8. Temporal distribution of pH, TOC and DOC. (a) Temporal variations of pH and water temperature at various locations along the Havel River. (b) Temporal variation of dissolved organic carbon (DOC) and total organic carbon (TOC) in Havel water of Section 1 (S-5) and Section 2 (S-24). All data are supplied on request by the Civil Service ‘‘Berliner Senatsverwaltung für Stadtentwicklung und Umwelt, Abteilung VIII, Referat VIII E’’.

soils by REE’’. Most of the REY contents are retained by sorption onto solid matter or by incorporation in secondary precipitates forming along with dissolution of Ca phosphates. Although agricultural fertilization in the study area takes place either in late autumn or early spring, TP increases in surface water only during the warm season May to October (Fig. 8a) or as shown for the chemical composition of S-4 in Table 1. LREE abundance is significantly less during the cold season and HREE and Y behave reversely. This feature is lost after mixing with water from both the River Spree and the Teltow Canal in which all REY are enhanced during the warm season (S-24; Table 1). PHREEQC (Parkhurst and Appelo, 2010) estimates reveal that secondary LREE phosphates may precipitate (Table 2) when Ca-phosphates dissolve in soils during the cold season. LREE are also coprecipitated with calcite (Zongh and Mucci, 1995). Thus the secondary soil minerals are enriched in LREE leaving the pore water enhanced in HREE+Y and therefore their export is favored by high runoff from less vegetation-covered soil (Figs. 3a and b). Besides organic complexes (Dupre et al., 1999; Johannesson et al., 2004) carbonate complexes are the most important ones, whereas phosphate complexes are minor as shown by PHREEQC estimates (Table 3). The percentage  of MCOþ 3 ; MðCO3 Þ2 and MPO4° species differ for light and heavy REE (M = REE)) during the cold and warm season which may have some importance with respect to sorption of these species. Competing with reversible sorption by colloids and labile hydromorphical substances in soils (Dupre et al., 1999; Johannesson et al., 2004; Leybourne and Johannesson, 2008) the mobility of REE

Fig. 9. Temporal distribution of total phosphorus (TP) at various locations along the Havel River and its contributors (a) and the Tegel water (S-10) after passing the phosphate elimination plant (b). Note the different temporal trends of TP in inputs to Lake Tegel (S-7 and S-8). Parallel to the abscissa the warm seasons are indicated by bars. All other data are supplied on request by the Civil Service ‘‘Berliner Senatsverwaltung für Stadtentwicklung und Umwelt, Abteilung VIII, Referat VIII E’’.

Fig. 10. Temporal variation of saturation indices of barite and calcite together with water temperatures at S-4 and the runoff assumed to be proportional to the flux of water recorded at S-21. Parallel to the abscissa the warm seasons indicated by orange beams. Excepting Ba the data are supplied by the Civil Service ‘‘Berliner Senatsverwaltung für Stadtentwicklung und Umwelt, Abteilung VIII, Referat VIII E’’.

depends on either positive or negative charge of the involved REE complexes (Table 3). Significant amounts of the dissolved

69

P. Möller et al. / Applied Geochemistry 41 (2014) 62–72

Table 1 Chemical composition of surface water samples S-4 (Section 1) and S-24 (Section 2) in July (warm season) and December (cold season) 2001. Note the different changes ot REY abundance at S-4 and S-24. Water sample S-4

Water sample S-24

Temperature °C Sampling date

19.5 7/3/2001

2.7 12/6/2001

14.1 7/2/2001

3.8 12/5/2001

pH Na (mg/l) K (mg/l) Ca (mg/l) Mg (Mg/l) NHþ 4 (mg/l) Cl (mg/l)

7.8 17 4.69 73 8.47 0.13 31 81

7.6 16.5 4.5 65.9 6.6 0.180 32 75

8.6 34.7 8.86 81.6 10.7 0.05 53 142

7.7 35.7 8.02 80.5 9.41 0.3 59 131

177 0.028

177 0.052

140 0.01

171 0.12

0.092 0.115 1.8 116 153 88.9 16.9 1.24 19.1 2.00 10.2 1.99 6.08 6.25 1.10 135

0.087 <0.1 5.9 55.3 71.2 52.1 11.5 1.75 18.1 2.00 12.3 2.89 9.59 9.92 1.78 173

0.082

0.15

0.023 17.4 21.0 14.9 3.21 0.53 440 0.62 4.17 1.02 3.62 7.26 1.58 59.9

7.07 25.0 30.8 23.0 5.90 0.98 1041 1.02 6.37 1.52 5.80 14.08 1.93 86.5

SO2 4 (mg/l) HCO 3 (mg/l) PO3 4 (mg/l) Total phosphorus mg/l B (lg/l) SiO2 (mg/l) La (pmol/l) Ce (pmol/l) Nd (pmol/l) Sm (pmol/l) Eu (pmol/l) Gd (pmol/l) Tb (pmol/l) Dy (pmol/l) Ho (pmol/l) Er (pmol/l) Yb (pmol/l) Lu (pmol/l) Y (pmol/l)

Fig. 11. Spatial changes of average concentrations along the Havel in the study area of dissolved major species (a), REE (b), and total Gd (c). Mixing with Spree and Lake Tegel water is indicated by black and green vertical dashed line.

fraction passing the 0.2 lm filter are bound to colloids. Generally, this fraction decreases from La to Lu Ingri et al., 2000; Kulaksiz and Bau, 2013) During the warm season, characterized by low runoff and enhanced vegetation, the secondary LREE-bearing soil minerals dissolve, favored by the enhanced partial pressure of CO2 in soils. Under such conditions pore water is enhanced in LREE and consequently also in the runoff and in surface water (cf. S-4 in Table 1). TP in surface water is also high during this season (Fig. 9). Thus, mainly the formation and time-shifted dissolution of secondary REE bearing soil minerals control the fractionation of REY as shown by Yb/Nd ratio in Havel water of Section 1 varying between 0.2–0.3 and 0.05–0.1 during the cold and warm seasons, respectively (Fig. 3b). Maximum Yb/Nd ratios in NPK fertilizers are about 0.36 but can be much lower depending on the raw material used in the production of fertilizers (Otero et al., 2005). Thus the high Yb/Nd ratios of surface water during the cold season seemingly approaches those of the fertilizer mix applied in the countryside. LREE are enriched in plant tissues of deciduous woods and their degradation produces LREE enrichment in soils after degradation in spite of REE complexation by abundant humic acids (Dupre et al., 1999; Johannesson et al., 2004). Thus, LREE are depleted in large river systems an feature that disappears at higher latitudes and may be related to the disappearance of vegetation (Stille et al., 2006a, 2006b). This enrichment of LREE by vegetation may be regionally counterbalanced by enlarged leaching of LREE from fertilized fields during the warm season. The higher the runoff the more the suspended load and colloids are exported and thereby LREE increase in streams (Ingri et al., 2000). Thus, the organic matter yields similar effects as fertilizers.

6.3. The impact of barite Fig. 12. REE distribution pattern of phosphorite from Bu-Graa, Spanish Sahara (unpublished) and a NPK 12/12/17 compound fertiliser (12% N; 12% P2O5; 17% K2O) after Otero et al. (2005). Note the anomalous Eu in the REE pattern of the fertilizer.

Analyses of agricultural fertilizers (Otero et al., 2005) reveal that particularly the very common compound NPK fertilizers contain Ba

70

P. Möller et al. / Applied Geochemistry 41 (2014) 62–72

Table 2 Saturation indices of various REE mineral phases calcite, monohydrocalcite, and hydroxylapatite in surface waters S-4 and S-24 (Table 1) resulting from PHREEQC using LLNL database (Parkhurst and Appelo, 2010). Supersaturation is indicated by bold figures. Saturation indices of S-4

Saturation indices of S-24

Cold season

Warm season

Nd SI SI SI SI

MPO4:10H2O MOHCO3 M2(CO3)3 (M(OH)3

Yb

0.7 23.09 7.09 8.14

Monohydrocalcite Hydroxylapaptite Calcite Partial pressure CO2

Nd 0.61 2.71 6.52 7.75

1.2 11.44 6.24

0.95 2.52 0.16 2.53

Cold season Yb 1.78 11.75 6.11

0.47 0.28 0.36 2.61

Nd 0.67 3.38 7.79 6.31

Warm season Yb 0.75 11.17 5.93

Nd 0.97 2.06 8.13 4.71

Yb 2.61 11.95 4.85

0.16 0.72 0.98 3.56

0.79 0.67 0 2.64

Table 3 Speciation of REE in S-4 and S-24 Havel waters (Table 1) in July (warm season) and December (cold season) 2001 estimated by PHREEQC using LLNL database (Parkhurst and Appelo, 2010). M represents members of the rare earth suite. Only inorganic complexes are considered. Missing data are below 0.01%. Free phosphate ions are assumed as 10% of total phosphorus in agreement with few determinations of orthophosphate in Havel water. Dissolved species

REE speciation in sample S-4 in% 2.7 °C

MCOþ 3 MðCO3 Þ 2 MPO4° MðPO4 Þ3 2 M3+ þ MSO4

REE speciation in sample S-24 in% 19.5 °C

3.8 °C

14.1 °C

Nd

Yb

Nd

Yb

Nd

Yb

Nd

73

56

63

56

68

49

27

14

24 1.9

40 4 0.1

35 0.9

42 1.8

27 4

43 8 0.6

72 0.2

86 0.2

0.5 0.3

0.1

0.05

0.5 0.5

0.1

0.03

and SO2 4 up to 430 ppm and 12%, respectively. Out of the several hundred lg/g of Ba in NPK fertilizer only 1 lg/g is water soluble (Otero et al., 2005) either as free ion or sorbed onto particulate matter indicating the presence of barite inherited from the source phosphorites. Barite is known to enrich Eu(II) (0.139VIII nm) which fits into Ba(II) sites (0.152VIII nm after Shannon, 1976), whereas all trivalent REE are much too small to be coprecipitated with barite (Möller, 2000). The REE patterns of the fertilizers show lower REE abundance and enhanced deficits of Eu than in the source phosphorite (Fig. 12) indicating loss of Eu(II)-bearing barite during the processing of phosphorites. Under conditions of high precipitation and less vegetation during the cold season the fertilizer granulates decompose in the topsoil. Part of the barite particles are washed into the streamlets discharging ultimately into the Havel River where they either dissolve or are present as nanoparticles passing the 0.2 lm filter. This leads to small Eu anomalies only. During the warm season characterized by less precipitation and high vegetation surface transport of barite seems to be negligible. Furthermore, barite may precipitate in the reducing horizon of gleysol that contains high amounts of organic matter resulting in low redox potentials and thereby reduce Eu(III) to Eu(II) which coprecipitates with barite. All these processes result in REY abundance with high Eu deficits in Havel water. In Section 1 the Eu/Eu** ratios are therefore positively correlated with the drainage of the catchment of the Havel River north of Berlin (Fig. 5a) but they are negatively correlated with TP in surface water (Fig. 9). In Section 2 and even more in Section 3 the seasonal variations of Eu anomalies decline.

6.4. Effect of phosphate elimination on REE distribution The temporal variations of phosphate concentrations in S-10 from Lake Tegel (Fig. 8b) deviates from the general trend of TP in Havel water (Fig. 9a). Due to diagenetic processes in the Lake’s

Yb

sediments during September to November phosphate is released. This process is only recognized at low TP of the lake water in S-10 (Fig. 9b). The diagenetic release of phosphate from sediments is a long-time effect of enhanced usage of fertilizers in agriculture in the past. The temperature-controlled mineralization process in lake sediments is associated with redox-controlled desorption of phosphate from Fe(III)-oxyhydroxides precipitates when Fe(III) is reduced to Fe(II) (Schauser, 2011). This type of release is also noticeable as small humps on the declining flanks of the TP seasonal main peaks in September–November (Fig. 9). The REE patterns of the Tegel water S-9 and S-10 show negligible seasonal variations (Fig. 4). Although TP is reduced about 40 times, water from S-10 indicates that REE abundance is not controlled neither by organic nor inorganic phosphates. Applying PHREEQC to the water in Table 1 and assuming only 10% of TP being present as ortho-phosphate may reveal that only the carbonate complexes are essential (Table 3).

6.5. Variations of Gd anomalies Gd anomalies in surface water originate from the effluents of STPs. Thus the distribution of Gd anomalies is related to their effluents. The Gd anomalies in Lake Tegel originate only from STP-1 (Fig. 1) which discharges its effluents through the Nordgraben into Lake Tegel. Similar Gd anomalies are also present in the Tegel Fließ because of its connection with the Nordgraben. During the cold and warm season the effluent of STP-2 are discharged into the Spree River and via a pipe line to the Teltow Canal, respectively, to keep the Havel River in the Berlin area less polluted during the warm season for recreational reasons. This explains the seasonal changes of Gd anomalies in Section 2 of the Havel River (S-21) and the Teltow Canal (S-30) (Fig. 11c). During the warm period total Gd is enhanced but constant in Section 2 of the Havel River. It increases drastically after mixing with water from the

P. Möller et al. / Applied Geochemistry 41 (2014) 62–72

Teltow Canal at S-31. During the cold season total Gd in Section 2 of the Havel River is high and slightly decreases downstream to S31 mainly due to dilution by precipitation and probably by transmetallation (Möller and Dulski, 2011). At this level the contribution by the Teltow Canal does not lead to variations of total Gd. The different concentrations of total Gd at Havel-km 28.9 may be due to incomplete mixing of waters from the Teltow Canal and the Havel River at the sampling point S-31. Transmetallation of Gd chelates may occur by species such as Fe(III), Cu(II) and the HREE and Y (Eq. (4)) (Möller and Dulski, 2010). This process is very slow because of the low concentrations of the interacting partners. Although Fe(III) is considered being able to replace Gd in its chelates, the effect during phosphate elimination is negligible (Fig. 4). However, Cu concentrations in Spree water (5–20 lg/l) and Havel water (5–5.8 lg/l) are reasonable to cause some transmetallation. Using the first rate constant kCu of 4  106 (nmol/L)1 h1 (Dulski et al., 2011) and assuming that only 10% of total dissolved Cu is present as free ion, Gd** decreases by only about 3% due to transmetallation (Eq. (5)) within one month which corresponds to the time needed by Havel water to pass the study area (Eq. (5)). 3þ

Gd-chelate þ M3þ () M-chelate þ Gd

ð4Þ

½Cu-chelate ¼ ½Gd-chelate  ½Cu2þ   kCu  24  30

ð5Þ

7. Conclusions Leaching of REY from fertilized soils depends on seasonal changes of precipitation and vegetation, water temperature and runoff. Subsequent to agricultural fertilization in autumn and early spring the Ca-phosphate components of artificial fertilizers dissolve and LREE coprecipitate with calcite and possibly form secondary phosphates. HREE are preferably exported from soil into rivers and lakes as organic and carbonate complexes or sorbed by particulate matter. Although the pore water is saturated with respect to Eu(II)-bearing barite from fertilizers, this micro-contaminant is partially exported by runoff due to poor vegetation. Thus, rivers and lakes inherit REY patterns with only small Eu deficits. During the warm season with less precipitation but high vegetation cover the secondary soil mineral dissolve and REY are exported together with phosphate into rivers and lakes. Barite, however, is retained from being moved by runoff and thus the river water shows enhanced light REE distribution patterns with high Eu deficits. The medical application of Gd chelates is responsible for the increases of total Gd from 8 up to 3300 pmol/l in the Havel River when passing the study area. The very stable Gd chelates pass the sewage treatment plants which discharge their effluents into the river system of Berlin. The seasonal variation of total Gd in the Havel River downwards the confluence with the Spree River is an artifact because the effluents of STP-2 discharged either into the Spree River or the Teltow Canal. The elimination of phosphate from surface water by precipitation with Fe(III) has only minor effects on the fractionation of REE. Diagenetic processes of the sediments associated with redox-controlled desorption of phosphate and reduction of Fe(III) to Fe(II) shows little effect on REY patterns. This indicates that phosphate has little influence on complexation of REY. Assuming that only 10% of TP is orthophosphate carbonate complexes dominate according to PHREEQC simulation. REY distribution is controlled neither by organic nor inorganic phosphate complexation. The Gd chelates are neither significantly reduced by the phosphate elimination process using sorption onto FeOOH precipitates nor by transmetallation within one month .

71

The REY abundance in Havel water before the confluence with the river Spree is much higher than thereafter (excepting Yb, Lu and total Gd). This is due to low REY abundance in the Spree water resulting from much higher loads of dissolved inorganic species, DOC and TOC. Before the confluence DOC is greater in the Havel water than thereafter but TOC behaves reverse. Thus sorption of REY onto settling organic matter may play a dominant role. Acknowledgement The authors deeply acknowledge the assistance of the Civil Service ‘‘Berliner Senatsverwaltung für Stadtentwicklung und Umwelt, Abteilung VIII, Referat VIII E’’ who made their analytical data files accessible and supplied a boat for water sampling campaigns and the Brandenburg State Office for Environment for supplying information on annual precipitation. We greatly acknowledge the assistance of S. Meier (FUB) and B. Richert, C. Wiesenberg and B. Zander (GFZ) for performing the analyses.

References Banks, D., Hall, G., Reimann, C., Siewers, U., 1999. Distribution of rare earth elements in crystalline bedrock groundwaters: Oslo and Bergen regions, Norway. Appl. Geochem. 14, 27–39. Bau, M., 1999. Scavenging of dissolved yttrium and rare earths by precipitating iron oxyhydroxide: Experimental evidence of Ce oxidation, Y-Ho fractionation, and lanthanide tetrad effect. Geochim. Cosmochim. Acta 63, 67–77. Bau, M., Dulski, P., 1996. Anthropogenic origin of positive gadolinium anomalies in river waters. Earth Planet. Sci. Lett. 143, 245–255. Bau, M., Knappe, A., Dulski, P., 2006. Anthropogenic gadolinium as a micropollutant in river waters in Pennsylvania and in Erie Lake, northeastern United States. Chemie der Erde – Geochem. 66, 143–152. Brücher, E., 2002. Kinetic stabilities of Gadolinium(III) chelates used as MRI contrast agents. Top. Curr. Chem. 221, 103–122. Dia, A., Gruau, G., Olivie-Lauquet, G., Riau, C., Molenat, J., Curmi, P., 2000. The distribution of rare earth elements in groundwater: assessing the role of sourcerock composition, redox changes and colloidal particles. Geochim. Cosmochim. Acta 64, 4131–4151. Dulski, P., 1994. Interferences of oxide, hydroxide and chloride analyte species in the determination of rare earth elements in geological samples by inductively coupled plasma-mass spectrometry. Fresenius J. Anal. Chem. 350, 194–203. Dulski, P., Möller, P., Pekdeger, A., 2011. Comparison of gadopentetic acid (GdDTPA) and bromide in a dual-tracer field experiment. Hydrogeol. J. 19, 823–834. Dupre, B., Viers, J., Dandurand, J.L., Polve, M., Benezeth, P., Vervier, P., Braun, J.J., 1999. Major and trace elements associated with colloids in organic-rich river waters; ultrafiltration of natural and spiked solutions. Chem. Geol. 160, 63–80. Elbaz-Poulichet, F., Seidel, J., Othoniel, C., 2002. Occurrence of an anthropogenic gadolinium anomaly in river and coastal waters of southern France. Water Res. 36, 1102–1105. Goldstein, S.J., Jacobson, S.B., 1988. Rare earth elements in river waters. Earth Planet. Sci. Lett. 89, 35–47. Hennebrüder, K., Wennrich, R., Mattusch, J., Stärk, H.-J., Engewald, W., 2004. Determination of gadolinium in river water by SPE preconcentration and ICPMS. Talanta 63, 309–316. Ingri, J., Widerlund, A., Land, M., Gustafson, Ö., Andersson, P., Öhlander, B., 2000. Temporal variations in the fractionation of rare earth elements in a boreal river; the role of colloid particles. Chem. Geol. 166, 23–45. Johannesson, K.H., Farnham, I.M., Guo, C., Stetzenbach, K.J., 1999. Rare earth element fractionation and concentration variations along a groundwater flow path within shallow, basin-fill aquifer, southern Nevada, USA. Geochim. Cosmochim. Acta 63, 2697–2708. Johannesson, K.H., Stetzenbach, K.J., Hodge, V.F., Kreamer, D.K., Zhou, X., 1997. Delineation of ground-water flow systems in the southern Great Basin using aqueous rare earth element distributions. Ground Water 35, 807–819. Johannesson, K.H., Tang, J., Daniels, J.M., Bounds, W.J., Burdige, D.J., 2004. Rare earth element concentrations and speciation in organic-rich blackwaters of the Great Dismal Swamp, Virginia, USA. Chem. Geol. 209, 271–294. Johannesson, K.H., Zhou, X., Guo, C., Stetzenbach, K.L., Hodge, V.F., 2000. Origin of rare earth element signatures in groundwaters of circumneutral pH from southern Nevada and eastern California, USA. Chem. Geol. 164, 239–257. Knappe, A., Möller, P., Dulski, P., Pekdeger, A., 2005. Positive gadolinium anomaly in surface water and ground water of urban area Berlin, Germany. Chemie der Erde–Geochem. 65, 167–189. Knappe, A., Sommer-von Jarmersted, C., Pekdeger, A., Bau, M., Dulski, P., 1999. Gadolinium in aquatic systems as indicator for sewage water contamination. In: Ármannsson, H. (Ed.), Proc. 5th Intern. Symp. Geochemistry of the Earth’s Surface. Balkema, Rotterdam, pp. 187–190.

72

P. Möller et al. / Applied Geochemistry 41 (2014) 62–72

Kulaksiz, S., Bau, M., 2007. Contrasting behaviour of the gadolinium and natural rare earth elements in estuaries and the gadolinium input into the North Sea. Earth Planet. Sci. Lett. 260, 361–371. Kulaksiz, S., Bau, M., 2011. Rare earth elements in the Rhine River, Germany: First case of anthropogenic lanthanum as a dissolved microcontaminant in the hydrosphere. Environ. Int. 5, 973–979. Kulaksiz, S., Bau, M., 2013. 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. 362, 43–50. Lawrence, M.G., Jupiter, S.D., Kamber, B.S., 2006. Aquatic geochemistry of the rare earth elements and yttrium in the Pioneer River catchment, Australia. Mar. Freshwater Res. 57, 725–736. Lawrence, M.G., Ort, C., Keller, J., 2009. Detection of anthropogenic gadolinium in treated wastewater in South East Queensland, Australia. Water Res. 43, 3534– 3540. Lead, J.P., Hamilton-Taylor, J., Peters, A., Reiner, S., Tipping, E., 1998. Eu binding by fulvic acids. Anal. Chim. Acta 369, 171–180. Leybourne, M.I., Johannesson, K.H., 2008. Rare earth elements (REE) and yttrium in stream waters, stream sediments, and Fe-Mn oxyhydroxides: fractionation, speciation, and controls over REE+Y patterns in the surface environment. Geochim. Cosmochim. Acta 72, 5962–5983. McLennan, S.M., 1989. Rare earth elements in sedimentary rocks. Influence of provenance and sedimentary processes. In: Lipin, B.R., McKay, G.A. (Eds.), Geochemistry and Mineralogy of Rare Earth Elements. Rev. Mineral., vol. 21, pp. 169–200. Möller, P., 2000. Rare earth elements and yttrium as geochemical indicators of the source of minerals and thermal waters. In: Stober, I., Bucher, K. (Eds.), Hydrology of Crystalline Rocks. Kluwer Academic Press, pp. 227–246. Möller, P., Dulski, P., 2010. Transmetallation of Gd-DTPA by Cu, Y and lanthanides and its impact on the hydrosphere. Appl. Geochem. 25, 48–59. Möller, P., Rosenthal, E., Dulski, P., Geyer, S., Guttman, Y., 2003. Rare earths and yttrium hydrostratigraphy along the Lake Kinneret-Dead Sea-Arava transform fault, Israel and adjoining territories. Appl. Geochem. 18, 1613–1628. Möller, P., Dulski, P., Bau, M., Knappe, A., Pekdeger, A., Sommer-von Jarmersted, C., 2000. Anthropogenic gadolinium as a conservative tracer in hydrology. J. Geochem. Explor. 69 (70), 409–414. Möller, P., Paces, T., Dulski, P., Morteani, G., 2002. Anthropogenic Gd in surface water, drainage systems, and the water supply of the city of Prague, Czech Republic. Environ. Sci. Technol. 26, 2387–2394. Möller, P., Stober, I., Dulski, P., 1997. Seltenerdelement-, Yttrium-Gehalte und Bleiisotope in Thermal- und Mineralwässern des Schwarzwaldes. Grundwasser 3, 118–132. Nozaki, Y., Lerche, D., Alibo, D.S., Tsutsumi, M., 2000. Dissolved indium and rare earth elements in three Japanese rivers and Tokyo Bay: evidence for anthropogenic Gd and In. Geochim. Cosmochim. Acta 64, 3975–3982. Ohta, A., Kawabe, I., 2001. REE(III) adsorption onto Mn dioxide (d-MnO2) and Fe oxyhydroxide: Ce(III) oxidation by d-MnO2. Geochim. Cosmochim. Acta 65, 695–703. Otero, N., Vitoria, L., Soler, A., Canals, A., 2005. Fertiliser characterisation: Major, trace and rare earth elements. Appl. Geochem. 20, 1473–1488. Paces, T., Möller, P., Fuganti, A., Morteani, G., Pecek, J., 2001. Sparkling mineral water at the western rim of the Doupovske hory Mountains (Czech Republic): genesis by water-rock interaction and deep-seated CO2. Bull. Czech Geol. Surv. 76, 189– 202. Parkhurst, D.L., Appelo, C.A.J., 2010. User’s Guide to PHREEQC (Version 2.17.5)-A Computer Program for Speciation, Batch-reaction, One-dimensional Transport and Inverse Geochemical Calculations. . Petelet-Giraud, E., Klaver, G., Negrel, P., 2009. Natural versus anthropogenic sources in the surface- and groundwater dissolved load of the Dommel River (Meuse Basin): Constraints by boron and strontium isotopes and gadolinium anomaly. J. Hydrol. 369, 336–349. Quinn, K.A., Byrne, R.H., Schijf, J., 2004. Comparative scavenging of yttrium and the rare earth elements in seawater: competitive influences of solution and surface chemistry. Aquatic Chem. 100, 59–80.

Quinn, K.A., Byrne, R.H., Schijf, J., 2006. Sorption of yttrium and rare earth elements by amorphous ferric hydroxide: influence of pH and ionic strength. Mar. Chem. 99, 128–150. Rabiet, M., Brissaud, F., Seidel, J.L., Pistre, S., Elbaz-Poulichet, F., 2009. Positive gadolinium anomalies in waste water treatment plant effluents and aquatic environment in the Herault watershed (South France). Chemosphere 75, 1057– 1064. Schauser, I., 2011. Causes and thresholds of P release from sediments. In: Chorus, I., Schauser, I. (Eds.), Oligotrophication of Lake Tegel and Schlachtensee, Berlin. . Schauser, I., Chorus, I., 2007. Assessment of the success of internal and external lake restauration measures in two Berlin lakes. Lake Reserv. Manage. 23, 366–376. Shannon, R.D., 1976. Revized effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides. Acta Crystallogr. A32, 751–767. Sholkovitz, E.P., 1992. Chemical evolution of rare earth elements: Fractionation between colloidal and solution phases of filtered river water. Earth Planet. Sci. Lett. 114, 77–84. Sholkovitz, E.P., 1995. The aquatic chemistry of rare earth elements in rivers and estuaries. Aquatic Geochem. 1, 1–34. Smedly, P.L., 1991. The geochemistry of rare earth elements in groundwater from the Carnmenellis area, Southwest England. Geochim. Cosmochim. Acta 55, 2767–2779. Stille, P., Steinmann, M., Pierret, M.C., Gaithier-Lafaye, F., Aubert, D., Probst, A., Viville, D., Chabaux, F., 2006a. The impact of vegetation on fractionation of rare earth elements (REE) during water rock interaction. J. Geochem. Explor. 88, 341–344. Stille, P., Steinmann, M., Pierret, M.C., Gaithier-Lafaye, F., Chabaux, F., Viville, D., Pourcelot, L., Matera, V., Aouad, G., Aubert, D., 2006b. The impact of vegetation on REE fractionation in stream waters of a small forested catchment (the Strengbach case). Geochim. Cosmochim. Acta 70, 3217–3230. Sun, H., Wang, X., Wang, L., 1997. Bioconcentration of rare earth elements lanthanum, gadolinium and Yttrium in algae (Chlorella Vulgarize Beijerinck): influence of chemical species. Chemosphere 34, 1753–1760. Takahashi, Y., Chatellier, X., Hattori, K.H., Kato, K., Fortin, D., 2005. Adsorption of rare earth elements onto bacterial cell walls and its implication for REE sorption onto natural microbial mats. Chem. Geol. 219, 53–67. Takahashi, Y., Hirata, T., Shimizu, H., Ozaki, T., Fortin, D., 2007. A rare earth element signature of bacteria in natural waters? Chem. Geol. 244, 569–583. Tang, J., Johannesson, K.H., 2006. Controls of the geochemistry of rare earth elements along a groundwater flow path in the Carrizo sand aquifer, Texas, USA. Geochem. Trans. 5, 49–60. Texier, A.C., Andres, Y., Le Cloirec, P., 1999. Selective biosorption of lanthanide (La, Eu, Yb) ions by Pseudomonas aeruginosa. Environ. Sci. Technol. 33, 489–495. Tweed, S.O., Weaver, T.R., Cartwright, I., Schaefer, B., 2006. Behaviour of rare earth elements in groundwater during flow and mixing in fractured rock aquifers: an example from the Dandenong Ranges, southeast Australia. Chem. Geol. 234, 291–307. Tyler, G., 2004. Rare earth elements in soil and plant systems – a review. Plant Soil 267, 191–206. Verplanck, P.L., Taylor, H., Nordstrom, D.K., Barber, L., 2005. Aqueous stability of gadolinium in surface waters receiving sewage treatment plant effluent, Boulder Creek Colorado. Environ. Sc. Technol. 39, 6923–6929. Willis, S.S., Johannesson, K.H., 2011. Controls on the geochemistry of rare earth elements in sediments and groundwaters of the Aquia aquifer, Maryland, USA. Chem. Geol. 285, 32–49. Zhang, Y., Deng, L.B., Ye, W.H., 2006. Does the long-term application of calcium superphosphate lead to an increase of soil rare earth element contents? J. Environ. 18, 130–134 (China). Zhu, Y., Hoshino, M., Yamada, H., Itoh, A., Haraguchi, H., 2004. Gadolinium anomaly in the distribution of rare earth elements observed for coastal seawater and river waters around Nagoya city. Bul. Chem. Soc. Jpn. 77, 1835–1842. Zongh, S., Mucci, A., 1995. Partitioning of rare earth elements (REEs) between calcite and seawater solutions at 25 and 1 atm, and high dissolved REE concentrations. Geochim. Cosmochim. Acta. 59, 443–453.