Environmental Science and Pollution Research

, Volume 16, Issue 6, pp 630–640

Temporal concentration changes of DEET, TCEP, terbutryn, and nonylphenols in freshwater streams of Hesse, Germany: possible influence of mandatory regulations and voluntary environmental agreements

Authors

  • Kristin Quednow
    • Department of Environmental Analytical Chemistry, Institute of Atmospheric and Environmental SciencesJ. W. Goethe University Frankfurt am Main
    • Department of Environmental Analytical Chemistry, Institute of Atmospheric and Environmental SciencesJ. W. Goethe University Frankfurt am Main
AREA 2.2 • MONITORING OF CHEMICALS IN WATER • RESEARCH ARTICLE

DOI: 10.1007/s11356-009-0169-6

Cite this article as:
Quednow, K. & Püttmann, W. Environ Sci Pollut Res (2009) 16: 630. doi:10.1007/s11356-009-0169-6

Abstract

Background, aim, and scope

The present study focuses on the temporal concentration changes of four common organic pollutants in small freshwater streams of Hesse, Germany. The substances (tris(2-chloroethyl)phosphate (TCEP), the technical isomer mixture of 4-nonylphenol (NP), 2-(t-butylamino)-4-(ethylamino)-6-(methylthio)-s-triazine (terbutryn), and N,N-diethyl-m-toluamide (DEET)) are subject to differing regulations. Whereas the use of NP and the related nonylphenolethoxylates (NPEOs) are almost completely banned under EU directive 2003/53/EC, the herbicide terbutryn is only restricted for use as a herbicide in the majority of member states of the European Union (EU). In contrast, TCEP and DEET are not regulated by legislation, but have been replaced in some products through consumer pressure. The impact of regulation on the environmental concentrations of these pollutants is discussed.

Materials and methods

The substances were monitored in small freshwater streams in the Hessisches Ried region, Germany, during the period September 2003 to September 2006. The samples were extracted with solid phase extraction (SPE) and analyzed by coupled gas chromatography–mass spectrometry (GC–MS).

Results

All target compounds were detected frequently within the fresh water streams of the study area. Monitoring in the study area revealed a significant concentration decrease only for NP. For the other three compounds, no significant concentration decrease was observed. Terbutryn concentrations and loads showed a seasonal trend with higher levels in summer and autumn, but were also present in winter and spring. Concentrations of TCEP and DEET were in the range of prior investigations.

Discussion

The decrease of NP concentrations and loads during the sampling period indicates that the regulation of NP and NP ethoxylates has led to a significant improvement in reducing the occurrence of this compound in the aquatic environment. Furthermore, the ban on agricultural use of terbutryn at the end of 2003 had no discernable influence on terbutryn concentrations in the following years.

Conclusions

The benefits of national bans or self-regulations by manufacturers on several chemicals appear to be limited. In contrast, the European-wide ban (of NP) revealed to be effective in preventing the substance from entering the aquatic environment on a large scale and reduced the NP concentration to an acceptable level (i.e., below the PNEC).

Recommendations and perspectives

Further research is needed to investigate diffuse sources and point sources of terbutryn not related to agriculture. Further research is required to find an explanation for the ongoing high concentration of TCEP in river water despite of the supposed replacement of TCEP by TCPP already in the 1990s.

Keywords

DEETFlame retardantsMonitoringNonylphenolOrganic pollutantsTCEPTerbutrynVoluntary agreements

1 Background, aim, and scope

In the last decade, studies have increasingly targeted the so-called emerging pollutants such as pharmaceuticals, personal care products, surfactants, flame retardants, and industrial and agricultural chemicals (i.e., Daughton 2004; Kolpin et al. 2002; Richardson and Ternes 2005). Many of these chemicals are characterized by high production volumes, high polarity, low biodegradability and consequent ubiquitous occurrence in the aquatic environment, and are the subject of public and scientific concern due to unknown or possible adverse effects on the environment and human health. As a result, the legislature of many countries presently endeavors to minimize risks by implementing new laws and regulations. In the European Union (EU), implementation of the Water Framework Directive no. 2000/60/EC of the European Community represents one approach for managing old and emerging pollutants. The directive defines a priority list of 33 substances or groups of substances of major concern for waters, and issues water quality objectives directed at achieving a ‘good ecological state’ for surface waters. Another legal measure is the European Union regulation 2006/1907 of 18 December 2006, which regulates the registration, evaluation, authorization, and restriction of chemicals (REACH). However, chemicals have been prohibited in the past with varying success for affected environments. In Germany, for example, the pesticide atrazine was banned in 1991 but was still detected in sewage treatment plant (STP) effluent in 1997 (Nitschke and Schussler 1998), and was found as a diffuse contaminant in groundwater until recently (Tappe et al. 2002). In France, the use of atrazine has been banned since September 2003, but a significant decrease in the concentration of this compound and its metabolite desethylatrazine has not been observed in the groundwater of an agricultural catchment in Breville, France (Baran et al. 2007). Voluntary agreements by manufacturers to abandon the use of some substances have played a major role in the past. For example, removal of phosphate from washing agents in the mid-1980s in Germany led to a noticeable improvement in surface water quality. However, the effectiveness of voluntary environmental programs and regulation is in question (Arimura et al. 2008; Blackman et al. 2007; Morgenstern and Pizer 2007). For example, Arimura et al. (2008) found two voluntary environmental actions (the ISO 14001 standard and public environmental reports) to be effective in reducing natural resource use, solid waste generation, and water wastage. Morgenstern and Pizer (2007) examined various voluntary environmental programs in the US, Japan, and the EU, and found the effects to be low (0–28%). Moreover, Ziegler and Rennings (2004) investigated the determinants of environmental innovation in companies and found that the European EMAS standard had no significant effect on environmental innovation.

In the present study, the occurrence of four common surface water pollutants was investigated in rivers of the Hessisches Ried catchment area near Frankfurt, Germany. These compounds have the characteristics of ‘emerging pollutants’, and include the insect repellent N,N-diethyl-m-toluamide (DEET), the chlorinated flame retardant tris(2-chloroethyl)phosphate (TCEP), the herbicide 2-(t-butylamino)-4-(ethylamino)-6-(methylthio)-s-triazine (terbutryn), and the technical mixture 4-nonylphenol isomers (NP), which are the degradation products of synthetic non-ionic surfactants (4-nonylphenol ethoxylates). The four compounds have different regulatory statuses. Under European Directive 2003/53/EC, the use of NP and the related ethoxylates have been largely prohibited in the EU since January 2005. According to this directive, NP and its ethoxylates may not be placed on the market or used as substance or constituent of preparations in concentrations equal or higher than 0.1% by mass. This restriction is valid for all uses resulting in discharges, emissions, or losses to the environment. Prior to this, NP use was subject to a risk assessment designed to yield predicted-no-effect concentrations (PNEC) of only 0.33 µg/l in the aquatic environment. NP can disrupt the endocrine system by interfering with the estrogen receptor (Jobling and Sumpter 1993). Apart from endocrine disruption, NP causes long-term toxic effects on aquatic organisms at concentrations of 3.3 µg/l (ECB 2002). From the 1980s to the early 2000s, NP was often detected in surface and wastewaters (reviewed by Thiele et al. 1997). Concentrations of NP in rivers in the 1980s occasionally reached values above 10 µg/l (Schaffner et al. 1987). Since the endocrine disruption and toxicity of NP was evident (Soto et al. 1991), the alkylphenolethoxylate-manufacturing industries in Germany decided in 1986 to voluntarily replace alkylphenolethoxylates (APEOs) in household, and later industrial detergents. A significant decrease of NP concentrations in sewage sludge (Jobst 1998) and surface waters (Leisewitz and Schwarz 1997) has been recorded since that time. However, according to Bolz et al. (2001) and Hegemann et al. (2002), the concentration of NP remained at high levels in STP effluents and in receiving waters in Germany in the late 1990s. NP concentrations up to 1.2 µg/l were still being reported in rivers in Germany in the early 2000s (Fries and Püttmann 2003b).

The herbicide terbutryn is regulated by the European Council Directive 91/414/EEC and the European Commission Regulation (EC) 2076/2002. Under these regulations, authorization of plant protection products containing terbutryn was withdrawn on 25 July 2003 (with a period of grace that expired on 31 December of 2003) for all members of the EU with the exception of the UK, Spain, Ireland, and Slovakia, where authorization is due to expire not later than 31 December 2007. In contrast, DEET and TCEP are not yet regulated by legislation, but have been voluntarily replaced by the main producers as a result of public concern. The risk of serious medical effects of DEET appears to be low (Bell et al. 2002) when applied with common sense, but in rare cases it has been toxic to humans (Davies et al. 1988). DEET is rapidly absorbed through the skin, metabolized and eliminated in the urine (Selim et al. 1995). It has also been suggested to be involved in the Gulf War Syndrome, causing toxic side effects due to synergistic interaction with other neurotoxic substances (Haley and Kurt 1997). In 1999, the Bayer group substituted the active agent in their products. For example, in the insect repellent Autan®, DEET was replaced by Bayrepel (1-piperidine carboxylic acid-2-(2-hydroxyethyl)-1-methylpropyl ester), which is as efficient as DEET and has no known adverse effects (Wahle et al. 1999). However, there are still products on the market containing DEET as the active agent. In Greece, DEET is used for indoor control of insects through vaporization of DEET-containing tablets in wall-mounted electric heating elements that slowly release the insecticide (Dall’Osto et al. 2007). In a recent review of the worldwide occurrence of DEET in the aquatic environment, concentrations ranging from 40 to 3,000 ng/l in various water bodies were reported (Costanzo et al. 2007). In Germany, DEET was found in surface waters at concentrations of up to 90 ng/l in River Lippe (Dsikowitzky 2002), 80 ng/l in the River Rhine (Knepper et al. 1996), and at 1.1 ng/l in maritime waters of the North Sea (Weigel et al. 2002).

TCEP is a flame retardant and plasticizer. The estimated consumption of TCEP in Germany in 1997 was 500–1,000 t (Leisewitz et al. 2001a), and on a global scale was about 4,000 t per annum in 1998 (ECB 2006). Until the mid-1990s, TCEP was used in a mixture of about 1:1 (Leisewitz et al. 2001b) with tris(1-chloro-2-propyl)phosphate (TCPP) as a flame retardant in polyurethane foams. As flame retardants evaporate from consumer products (Kemmlein et al. 2003), TCEP was detected in indoor air (Hartmann et al. 2004) and house dust (Marklund et al. 2003). TCEP has been shown to be carcinogenic and neurotoxic to animals (Umezu et al. 1998), and in the early 1990s the European Isocyanate Producers Association (ISOPA) recommended replacement of TCEP. The two German producers have ceased TCEP production, and it is replaced by TCPP in polyurethane foams (Fooken 1997). According to the European Flame Retardants Association (EFRA), TCEP is no longer produced in the EU-15 member states (LfU 2007), but processing/use of TCEP is not prohibited. Imports of TCEP to the EU in 2001 amounted to about 1,150 t (ECB 2006). TCEP is on the second EU priority list and currently a risk assessment is carried out. The risk assessment draft of 2 March 2006 proposed a PNEC for water of 65 µg/l (ECB 2006). As TCEP is commonly present in wastewater and is not eliminated in STPs (Höhne and Püttmann 2006), it enters the aquatic environment via this route (Fries and Püttmann 2003a; Prösch et al. 2002). Consequently, TCEP is often found in STP effluents and receiving rivers (Andresen et al. 2004; Fries and Püttmann 2003a).

2 Materials and methods

2.1 Study area and sampling

The study area is in the Hessisches Ried region, south of Frankfurt am Main, Germany. This is a densely populated and industrialized area which also supports agriculture. It is drained by four small rivers (the Schwarzbach, Modau, Winkelbach, and Weschnitz rivers), all of which are tributaries of the Rhine River and serve as receiving waters for treated wastewater. Figure 1 provides an overview of the study area and the location of sampling sites. These river systems discharge only small water volumes on an annual basis (Göbel 1996): 0.3 m3/s for the Winkelbach River, 1.1 m3/s for the Modau, 2.0 m3/s for the Schwarzbach, and 3.2 m3/s for the Weschnitz. The Schwarzbach and Weschnitz rivers receive the greatest volumes of treated wastewater, but due to the larger volume of the latter, the proportion of wastewater in the flow at its mouth is lowest (15%). In contrast, the Winkelbach, Modau, and Schwarzbach rivers contain a high proportion of wastewater (47%, 35%, and 50%, respectively) at their mouths. In extremely dry seasons, the discharge of the Schwarzbach River can comprise almost 100% wastewater.
https://static-content.springer.com/image/art%3A10.1007%2Fs11356-009-0169-6/MediaObjects/11356_2009_169_Fig1_HTML.gif
Fig. 1

Sampling area (Hessisches Ried) and location of the sampling points

A total of 330 water samples were collected during 13 sampling occasions from 26 sampling sites between September 2003 and September 2006. Parameters including water temperature, conductivity, and pH were recorded at each site during sampling. A volume of 2.5 l of surface water was collected at each of the 26 sampling sites, stored in a brown glass bottle and kept cool (4°C) during transport to the laboratory. Samples were processed the day after sampling.

2.2 Sample preparation and analysis

Samples were passed through pre-cleaned paper filters (Schleicher & Schuell, grade 597½) and then extracted by passage at low pressure (approximately 800 mbar) through solid phase extraction (SPE) cartridges (Bond Elut, 1 ml). Prior to use, the cartridges were rinsed with 1 ml methanol and then conditioned with 1 ml methanol/acetonitrile (50:50) and 1 ml distilled water. The SPE cartridges were eluted three times with 333 µl of a methanol/acetonitrile mixture (50:50). The solvent was removed under a nitrogen stream and the extract residue was dissolved in 100 µl acetonitrile. Squalane (4,000 ng) was added to each extract as an internal standard. The extracts were subjected to GC–MS analysis using a Fisons GC 800 gas chromatograph equipped with a BP-X5 capillary column (30 m length, 0.25 mm inner diameter, 0.25 μm film thickness), and with helium used as the carrier gas. The GC oven temperature varied from 80 to 300°C at a rate of 4°C per minute. Extracts (1 μl) were injected in splitless mode. The mass spectrometer (Fisons MD 800) used electron ionization (EI, 70 eV) and was operated in full-scan mode (50–600 m/z). The analytes were identified by comparison with reference standards of terbutryn, 4-nonylphenol and DEET (Ehrenstorfer), and TCEP (Sigma Aldrich).

For quantification of the technical mixture of 4-nonylphenol isomers, the method previously described by Fries and Püttmann (2003b) and Höhne and Püttmann (2008) was applied. TCEP was analyzed using the method described by Quednow and Püttmann (2008b) and terbutryn was analyzed according to Quednow and Püttmann (2007). DEET was quantified in the same extracts using m/z 119 as quantifier ion and m/z 91 and m/z 191 as qualifier ions. The peak area of the internal standard (squalane) has been measured in the total ion chromatogram while the signals of the analytes have been integrated in the mass chromatograms of the respective mass traces. The peak areas of the analytes in the mass traces were revised with correction factors and response factors as described by Fries and Püttmann (2003a, b). Both factors have been determined at each measuring campaign. Detection limits were determined in accordance with DIN 32645 (1994) with concentrations of 4 ng/l (terbutryn), 8 ng/l (DEET), 5 ng/l (TCEP), and 8 ng/l (NP). Recoveries were 80% (terbutryn), 81% (DEET), 110% (TCEP), and 77% (NP), respectively.

3 Results and discussion

3.1 NP (technical isomer mixture)

For the NP technical isomer mixture, a maximum concentration of 0.77 µg/l was detected in the Schwarzbach River in September 2003, when overall highest concentrations (compared to the other sampling periods) were detected. After September 2003, the mean NP concentrations decreased from 0.4 to 0.02 µg/l (in September 2006), and the detection frequency of NP in the water samples also decreased. Whereas in September 2003, NP was present in all samples at concentrations above the detection limit of 0.008 µg/l; in September 2006, NP could be detected in only 60% of the samples. During two sampling campaigns, in April 2005 and June 2005, NP was not detected in any water sample analyzed. In total, NP was detected in 56% of the water samples. In addition to decreased concentrations, mean NP loads decreased from 256 µg/s in September 2003 to 9 µg/s in September 2006 (Fig. 2). The decreasing loads indicate that the observed NP decrease was due to lower inputs, and not just a function of dilution effects due to higher discharge volumes. The decrease in NP can be attributed to the implementation of European Directive 2003/53/EC, which has prohibited the use of NP and related nonylphenolethoxylates (NPEOs) in industrial and commercial cleaning agents since January 2005. Similar outcomes have also been observed in wastewaters and rivers in Switzerland (Voutsa et al. 2006), and from STP effluents in Germany (Höhne and Püttmann 2006). Moreover, in a previous study (Quednow and Püttmann, 2008a), the occurrence of octylphenol (OP) has been investigated in the rivers of the Hessisches Ried. In contrast to NP, the concentration of OP showed no decrease during the same sampling period. OP has not been banned until now. The main use of octylphenol (80% of the total quantity) is in the production of para-tert-octylphenol (PTOP)-based resins, which are used as tackifiers in tire manufacture.
https://static-content.springer.com/image/art%3A10.1007%2Fs11356-009-0169-6/MediaObjects/11356_2009_169_Fig2_HTML.gif
Fig. 2

Nonylphenol concentration and load from September 2003 to September 2006 in the rivers of the Hessisches Ried

At the end of the sampling campaign in September 2006, NP was still present in the study rivers, albeit only in small concentrations (average approximately 20 ng/l). Although the EU directive bans the use of NP and NPEOs in textiles within the EU, there is no ban on imported textiles containing these substances. In a recent study for the Swedish Society for Nature Conservation (SSFNC), the NPEO content of 20 towels produced by different companies was analyzed, and NPEOs were found to be present in all samples at concentrations up to 10.6 g/kg (Hök 2007). On the basis of these results, the SSFNC estimated the amount of NP imported in textiles from countries outside the EU and entering STPs was 9 t/a in Stockholm alone. Furthermore, a representative from SASOL Germany, a producer of APEOs, estimated the quantity of NPEOs in textiles imported to the EU to be 500–1,000 t/a (C.D. Hager, personal communication 2006). Therefore, imported textiles still might represent an important source of NP entering the aquatic environment in Europe. Apart from imported products, plant protection substances can also be a source of NP entering aquatic environments, since NP is still used as an emulsifying agent in some pesticides (Guenther et al. 2002). The use of these products was not prohibited with implementation of European Directive 2003/53/EC. In addition, diffuse sources including re-solution from river sediments has to be considered, since NP tends to adsorb to suspended matter and is often found in sediments (Bolz et al. 2001; Heemken et al. 2001; Isobe et al. 2001). Although data on NP in sediments of the study rivers are not available, it is likely that sediments represent an NP pool as a result of high NP fluxes in the past.

3.2 TCEP

Mean TCEP concentrations in the study rivers varied significantly, from 328 ng/l in the Schwarzbach River to 242 ng/l in the Modau River, 108 ng/l in Weschnitz River, and 71 ng/l in the Winkelbach River (Table 1). Generally, TCEP concentrations were higher at sampling locations that were influenced by high wastewater input, including all sampling locations of the Schwarzbach–Landgraben River, with the exception of Sw1, and all sampling locations near the mouths of the Weschnitz, Modau, and Winkelbach rivers. Furthermore, a positive correlation (R = 0.5) between TCEP concentration and associated population equivalent was observed. The highest TCEP concentrations (up to 2 µg/l) were found in September 2005 in the Schwarzbach River which, together with the Weschnitz River, had the highest transported TCEP loads (201 and 180 µg/s, respectively). Thus, the input of TCEP into both rivers is roughly the same, with the lower concentrations in the Weschnitz River resulting from dilution due to its greater discharge volume. The lowest TCEP concentration and load (average 71 ng/l and 20 µg/s, respectively) were in the Winkelbach River, which can be attributed to the low wastewater input and generally low discharge volumes (see Table 1) in this river. TCEP loads in the Modau River were intermediate (75 µg/s), because only small STP effluent volumes are discharged into the river’s lower reaches, and the river has a lower discharge volume as related to the Weschnitz River. Consequently, TCEP concentrations were higher in the Modau River than in the Weschnitz River.
Table 1

Proportion of samples with concentrations of the analytes above detection limit concentration [ng/l] and load of the analytes in the rivers of the Hessisches Ried

Analyte

River

% > DL

Concentration [ng/l]

Load [µg/s]

Min

Max

Mean

Mean

DEET

Schwarzbach

87

<DL

1,292

245

143

Modau

72

<DL

936

82

30

Winkelbach

46

<DL

564

70

20

Weschnitz

69

<DL

380

64

119

Total

70

<DL

1,292

124

83

TCEP

Schwarzbach

99

<DL

2,019

328

201

Modau

95

<DL

1,190

242

75

Winkelbach

71

<DL

561

71

20

Weschnitz

97

<DL

464

108

180

Total

92

<DL

2,019

203

127

Terbutryn

Schwarzbach

78

<DL

246

63

42

Modau

90

<DL

3,067

583

184

Winkelbach

40

<DL

554

28

7

Weschnitz

75

<DL

5,600

536

976

Total

73

<DL

5,600

306

292

NP

Schwarzbach

53

<DL

770

44

29

Modau

60

<DL

723

62

19

Winkelbach

55

<DL

618

48

13

Weschnitz

56

<DL

717

40

62

Total

56

<DL

770

49

31

Number of analyzed samples n = 330

There was a temporal variation in TCEP loads, which was significantly higher in summer (173 µg/s) and autumn (146 µg/s) (Fig. 3). This was probably due to lower rainwater input and therefore comparatively higher proportions of wastewater being present during these periods. No overall trend of decreasing TCEP concentrations was observed during the study sampling period (Fig. 4). The detection frequency was about 92% with a detection limit of 5 ng/l, suggesting that TCEP still present as a flame retardant in consumer products is entering surface waters via STP effluents. This confirms other studies reporting frequent detection of TCEP in various water analyses. Andresen et al. (2004) reported STP effluent and river water concentrations of 120 and 50 ng/l, respectively, in the Ruhr/Rhine area, and Dsikowitzky et al. (2004b) found TCEP concentrations up to 0.2 µg/l in the Lippe River. Prösch et al. (2002) reported decreasing concentrations of TCEP in the 1990s, when most conversion from TCEP to TCPP occurred. However, this measure was clearly not sufficient, with TCEP continuing to enter the aquatic environment via STPs as a consequence of ongoing importation of TCEP-containing products from non-EU states.
https://static-content.springer.com/image/art%3A10.1007%2Fs11356-009-0169-6/MediaObjects/11356_2009_169_Fig3_HTML.gif
Fig. 3

Seasonal variation of TCEP, DEET, and terbutryn load (mean) in the rivers of the Hessisches Ried

https://static-content.springer.com/image/art%3A10.1007%2Fs11356-009-0169-6/MediaObjects/11356_2009_169_Fig4_HTML.gif
Fig. 4

Concentrations (mean values) of TCEP, DEET, and terbutryn from September 2003 to September 2006 in the rivers of the Hessisches Ried

TCEP concentrations had previously (1992–2000) been determined in rivers from the same catchment area examined in our study (Fooken 2000), and a comparison of both data sets is shown in Fig. 5. The results show a decrease in the concentration of TCEP in river water around the year 2000, but a further decrease since that time is not evident.
https://static-content.springer.com/image/art%3A10.1007%2Fs11356-009-0169-6/MediaObjects/11356_2009_169_Fig5_HTML.gif
Fig. 5

TCEP concentrations of this study in comparison to older monitoring dates

3.3 Terbutryn

The occurrence of terbutryn in the study area is shown in Fig. 4. The terbutryn data have been discussed in detail previously (Quednow and Püttmann 2007). The highest terbutryn concentration (5.6 µg/l) was found in June 2004 in the Weschnitz River (We3). This concentration is very high considering that terbutryn is toxic to algae (EbC50—72 h) at concentrations of 2 µg/l (Okamura et al. 2000), and has adverse effects on primary production of periphytic algae at concentrations above 0.6 µg/l (Brust et al. 2001). Furthermore, Richter and Nagel (2007) recently identified water as the primary route of terbutryn uptake in two benthic organisms (Gammarus fossarum and Asellus aquaticus). They determined bioaccumulation factors of 13 and 30, respectively, in these organisms, indicating a moderate potential for bioaccumulation. Terbutryn concentrations showed a clear spatial trend, with high concentrations in the Weschnitz and Modau rivers (0.54 and 0.58 µg/l, respectively), and low concentrations in the Winkelbach and Schwarzbach rivers (0.06 and 0.03 µg/l, respectively). The Weschnitz River had a higher load (399 µg/s) than the Modau River (124 µg/s), indicating a greater input of terbutryn. Furthermore, terbutryn loads were higher in summer and autumn compared to winter and spring (see Fig. 3), which might argue for a agricultural source of terbutryn in both rivers. However, additional sources for terbutryn in surface waters such as run-off from roof paints and house paints have to be considered (Menge 2005). The very high concentrations of terbutryn (>1 µg/l) in the Weschnitz and Modau rivers occur both downstream from sewage treatment plants. Recently, a pigment factory producing house paints containing terbutryn as biocide has been reported to cause the high terbutryn concentrations in the Modau river downstream from Ober-Ramstadt (Fritz 2008). In the case of the Weschnitz river, a possible point source for the high concentrations of terbutryn downstream from Weinheim has not been identified until now. But the high terbutryn concentrations exceeding 1 µg/l in both rivers have to be related rather to the discharge of contaminated sewage water from point sources than from diffuse sources such as agriculture or run-off from roofs and painted houses. In contrast, the comparably low terbutryn concentrations observed in Winkelbach and Schwarzbach rivers (<0.06 µg/l) are instead caused by diffuse sources.

Although the application of terbutryn as a herbicide in Germany has been strictly prohibited since 31 December 2003, a trend of decreasing terbutryn concentration in the study rivers was not detected (see Fig. 4). In contrast to the expected decrease after the ban, terbutryn concentrations were highest in the 2004 growing season. A possible explanation could be the deliberate exhaustion of remaining herbicide stocks by farmers. This explanation, however, would not explain the fact that terbutryn levels in the water samples remained at the same levels in 2005 and 2006.

The ban of terbutryn in Germany by the end of 2003 was restricted to the application in agriculture, but the further use of terbutryn as biocide, i.e., in roof paints and outside wall paints, was not forbidden. That might be the reason why terbutryn concentrations remained detectable with concentrations up to 0.1 µg/l in all study rivers, possibly due to the influence of run-off from painted outside walls and roofs. Is those rivers (Modau and Weschnitz), obtaining sewage waters from point sources such as a paint factory (in case of Modau), terbutryn concentrations exceeded even 1 µg/l. The results show that the ban of terbutryn application in agriculture by the end of 2003 had no significant effect on the degree of terbutryn pollution in the rivers under investigation. Apparently, the use of terbutryn simply shifted from the application as herbicide in agriculture to several other applications such as biocide in various paint products.

3.4 DEET

The DEET concentrations in river samples are listed in Table 1. Significantly higher mean concentrations (245 ng/l) were found in the Schwarzbach River compared to the Weschnitz, Winkelbach and Modau rivers (64, 70, and 82 ng/l, respectively). The highest overall DEET concentration (1.3 µg/l) was found in the Schwarzbach River in June 2004. High DEET loads were detected in the Schwarzbach and Weschnitz rivers. As for TCEP, the higher discharge of the Weschnitz River results in lower concentrations than, but comparable loads to, the Schwarzbach River. The correlation of DEET concentrations with the population equivalent was also positive (R = 0.5), and the correlation with TCEP was strong (R = 0.62, two-tailed, 0.05 probability level). This is not surprising since both DEET and TCEP are common wastewater ingredients (Dsikowitzky et al. 2004a). In parallel with insect populations, mean DEET concentrations and loads (see Fig. 3) were significantly higher in summer (0.15 µg/l) than in autumn, winter, and spring (0.06, 0.04, and 0.05 µg/l, respectively). In the 1990s, Knepper et al. (1996) found concentration peaks of DEET (0.03–0.08 µg/l) in the Rhine River during summer and autumn, paralleling seasonal application as an insect repellent. Similar to the results of the present study, Knepper et al. (1996) detected DEET in the rivers outside the insect season, and later investigations revealed elimination of DEET in STPs only at influent concentrations above 0.3 µg/l, with an elimination rate of about 40% (Knepper and Maes 2003). Higher influent concentrations only occurred in summer. Furthermore, the occurrence of DEET in sewage water in winter and spring (when DEET application as an insect repellent in Germany is virtually non-existent) may be the result of washing of clothes following holidays in warmer regions (Knepper 2004a). This explanation seems plausible since DEET is one of the most efficient insect repellents (Fradin and Day 2002) and is often recommended for travel in the tropics. According to Knepper (2004b), concentrations of DEET in sewage and surface water decreased after 1999 when DEET was replaced by Bayrepel in many commercial insect repellents. However, during our sampling period (September 2003 to September 2006), no trend of further decrease was observed (see Fig. 4). DEET concentrations varied remarkably over the sampling period and remained at a median level of about 0.06 µg/l, which is in the range of concentrations reported for the late 1990s.

4 Conclusions

All target compounds were consistently detected in rivers in the study area. The lowest concentration, load, and detection frequency over the sampling period were observed for NP, which is subject to the strongest regulation through an EU-wide ban. With respect to NP concentrations in the aquatic environment in the last two decades, we conclude that self-regulation by manufacturers has led to an improvement in the pollution status of this compound in Germany, but environmental concentrations have not been reduced to an acceptable level (i.e., below the PNEC). In contrast, the 2005 European-wide ban seemed to be effective in preventing the substance from entering the aquatic environment on a large scale.

Concentrations of TCEP, DEET, and terbutryn showed no decrease during the sampling period. This is particularly surprising for terbutryn, as it indicates that the 2003 ban on the use of terbutryn as an herbicide in agriculture has had no detectable effect on the concentrations of this compound in the investigated river systems. TCEP and DEET were both withdrawn before 2000 under voluntary agreements, and concentrations were stable at the 2000 level. Therefore, it appears that neither the ban in only one of several fields of applications (terbutryn) nor voluntary withdrawal (TCEP and DEET) at the national level are sufficient to prevent the substances from entering the aquatic environment. As postulated by Morgenstern and Pizer (2007), it seems that, although voluntary programs can offer environmental gains, the benefits of such programs are limited.

5 Recommendations and perspectives

The results from the study have shown that, in the case of terbutryn, the prohibition of its use as herbicide in agriculture since 31 December 2003 was not followed by the expected decrease of river pollution in the Hessisches Ried. Due to the further use of terbutryn in other applications, such as biocides in paint products, the degree of river pollution remained on a high level. Run-off from painted roofs and outside walls has to be considered as a possible source of diffuse surface water pollution and also sewage waters from paint factories should be considered as possible point source for terbutryn, as shown in one example. Due to the adverse effect of terbutryn on aquatic organisms, the substance should be restricted for all uses which result in discharges, emissions, or losses to the environment similar to the restriction of nonylphenols and its ethoxylates (since 2005), which proved to be very effective.

Further investigations are required to find explanations for the ongoing high concentrations of TCEP in river waters. Since TCEP has been replaced as a flame retardant by TCPP already in the 1990s, evaporation of TCEP from “old” products impregnated with this flame retardant has to be checked as a possible emission pathway. Due to the observed persistence of TCEP, it has to be clarified whether the currently in high abundance applied flame retardant TCPP (with similar properties in the environment) is the appropriate substitute for TCEP.

Acknowledgements

This study was supported by grants of the Hessian Ministry for Science and the Arts. The authors would like to thank all participants of the INTAFERE project and the Hessian State Office for the Environment and Geology for support.

Copyright information

© Springer-Verlag 2009