Science of the Total Environment 542 (2016) 955–964

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Influence of the water quality improvement on fish population in the Seine River (Paris, France) over the 1990–2013 period Sam Azimi ⁎, Vincent Rocher SIAAP–DDP, 82 Avenue Kleber, F 92 700 Colombes, France

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• In 2013, the dissolved oxygen concentration in the Seine River is slightly modified by the Paris anthropogenic activities. • An emergence of a more sensitive assemblage of fish is observed over the past decades. • The overall fish index of the river Seine has the good quality level downstream of Paris. • Exceedance of the environmental quality standards for mercury and PCBs was evidenced in the fish muscles.

a r t i c l e

i n f o

Article history: Received 4 August 2015 Received in revised form 19 October 2015 Accepted 19 October 2015 Available online 9 November 2015 Editor: D. Barcelo Keywords: River Seine Ecological status Electrofishing Fish species assemblage IBI EROD Dissolved oxygen

⁎ Corresponding author. E-mail address: [email protected] (S. Azimi).

http://dx.doi.org/10.1016/j.scitotenv.2015.10.094 0048-9697/© 2015 Elsevier B.V. All rights reserved.

a b s t r a c t Over the past 20 years, rules concerning wastewater treatment and quality of water discharged into the environment have changed considerably. Huge investments have been made in Paris conurbation to improve waste water treatment processes in accordance with the European Water Framework Directive. The interdepartmental association for sewage disposal in Paris conurbation (SIAAP) carried out a monitoring of both fish assemblages and water quality in the Seine River around the Paris conurbation (France) since the early 90's. The main goal of this study was to estimate the influence of the water quality improvement on fish. On one hand, the study confirmed the improvement of the water quality (dissolved oxygen, ammonia nitrogen, organic matter) in the Seine River, mostly focused downstream of Paris conurbation. On the other hand, an increase of the number of species occurred from 1990 (14) to 2013 (21). Moreover, changes in the river Seine assemblages happened over that 23-year period with emergence of sensitive species (ruffe, scalpin and pikeperch). The improvement of the water quality was also reported with respect to the Index of Biotic Integrity (IBI). However, no variation of pollutant concentrations in roach, eel and chub muscles has been observed. An exceedance of the environmental quality standards have even been reported all over this period as regards mercury and organochlorine. © 2015 Elsevier B.V. All rights reserved.

956

S. Azimi, V. Rocher / Science of the Total Environment 542 (2016) 955–964

1. Introduction The management of aquatic environments in Europe is governed by Water Framework Directive (WFD, 2000/60/CE), transposed into French law by the 2006 Act on water and aquatic environment (2006– 1772 of 30/12/06). These guidelines, aiming at improving and restoring the good ecological status of water, have been backed by the Grenelle Acts 1 and 2 (2009-967 of 03/08/09 and 2010-788 of 12/07/10). For a long time, the watercourse quality assessment has only relied on the analysis of the physical and chemical composition of water. It is now based also on biological components of the aquatic ecosystems as algae (diatoms), macrophytes, benthic macro-invertebrates and fish (O'Farrell et al., 2002; Hering et al., 2003; Hering et al., 2010). Among all these potential bio-indicators, fish are genuine integrators of water quality and, more broadly, of the functioning of aquatic environments because they are at the top of the food chain, Also they are sensitive to both quality of water and health of the physical habitat, they can live for a long time and they have a very mobile nature (Sanchez et al., 2012; Tales, 2009; Polard, 2010; Couillard, 2009). The use of such sentinel species allows to track environmental variations at variable temporal scales, ranging from weeks (invertebrates) to years (fish). Also, since bio-indicators differ in ecological traits such trophic position, they can inform if pollutants have permeated upwards the food-webs (bio-accumulation and bio-magnification risks) (reviewed in Colin et al., 2016). Furthermore, over the past 20 years, rules concerning wastewater treatment and quality of water discharged into the environment have changed considerably. In 1991, the implementation of the European Directive on the collection, treatment and discharge of wastewater, has requested the member states of the European Union to identify areas sensitive to eutrophication, in which discharges of phosphorus and nitrogen should be reduced (it was already the case for organic matter). Specifically, the treatment plants of medium and large sizes located in these areas had to develop treatments that respect maximum concentrations of nitrogen and phosphorus. These regulatory requirements have imposed an upgrade of the Parisian urban wastewater plants in order to provide treatment that can effectively remove carbon, nitrogen and phosphorus from wastewater. So, significant efforts have been made in Paris conurbation since early 90's to integrate water treatment units for chemical treatment of phosphorus and biological treatment of nitrogen. The monitoring of the chemical composition of the river Seine (dissolved oxygen, organic matter, nitrogen, phosphorus, etc.) is done since the beginning of the 20th century (Rocher et al., 2015). In order to collect biological data for the monitoring of watercourse quality, as of 1990, the interdepartmental association for sewage disposal in Paris conurbation (SIAAP) has enlisted support from National Fisheries Board (whose responsibilities were subsequently rolled into those of French National Agency for Water and Aquatic Environments called ONEMA) for carrying out a monitoring of fish assemblages in the river Seine across the Paris conurbation (France). Since the 2000s, that monitoring has been supplemented with analyses of various pollutants (heavy metals, PCBs, pesticides) in the muscles of some species, as well as the EROD activity (7-ethoxyresorufin-O-deethylase) in the chub livers. Both approaches (physico-chemical analysis and bio-indicator species) are complementary for the purpose of aquatic environment monitoring. The former is based on the chemical analysis of a number of pollutants in the environmental matrices and enables the assessment of the contamination of habitats. This approach, however, provides no data about the impact of the chemical molecules and on how natural factors (temperature, hydrological regime) modulate their effects, on the living organisms. The latter combines the occurrence and abundance of bio-indicator species in order to diagnose the quality of water and biocenoses, a posteriori a disturbance, through an examination of the biological effects of the pollutants on the organisms. The aim of this study is to provide information about the possible influence of the improvement of the water quality in the Seine River, with

respect to the WFD, on fish population. The first section deals with the estimates of fish assemblage in 8 localities from upstream to downstream of the Paris conurbation. The results describe both counting and ecotype of the various identified species. The full dataset was used for calculating the Index of Biotic Integrity (IBI), a suitable indicator for the biological quality of water bodies. The second section of the survey presents the micropollutant contents (metals, PCBs and pesticides) in the muscles of eel, chub and roach and the description of the EROD index as determined in the chub livers. Before studying those topics, a first paragraph will focus on the water quality improvement of the Seine River. 2. Material and methods 2.1. Survey sites Fishing sites have been selected and distributed from upstream to downstream of Paris conurbation, so that the SIAAP's intervention range can be encompassed. They have been chosen to provide a representativeness of aquatic habitats in urban areas (Fig. 1). Until 1999, the monitoring network comprised 4 sampling stations lying along the river Seine in Villeneuve-Saint-Georges, Paris, Levallois/Asnières and Epinay sur Seine. The network was extended to 7 stations in 2000 to include Le Pecq, Poissy and Triel-sur-Seine stations. In 2013, the Choisy-le-Roi station was added to the network to encompass all 4 Paris conurbation wastewater treatment plants that directly discharge treated water into the river Seine. The surveys are conducted in partnership with ONEMA. These stations are intended to allow keeping a close look at the potential impact of releases from most significant structures in urban environments. Specific fishing campaigns have been set up for the purpose of micropollutant analysis at the Villeneuve-St.-Georges, Levallois/ Asnières and Triel-sur-Seine sites. These sites encompass areas lying upstream, near downstream and far downstream of the capital, respectively. 2.2. Applied protocols The sampling protocol was not uniformly applied throughout the survey period, since the national protocol was amended in 2005. From 1990 to 2004, the sampling was conducted using the per-habitats fishing method. That method consists in discretely identifying and exploring some fifteen key habitats that are both typical to the locality and very attractive to fish. The caught fish are classified per habitat, the final result being the sum of caught fish in relation to the sum of acreages of the explored habitats. Hence, it is not an assessment of assemblage densities, but a catch per unit effort (CPUE). That method is considered as very reliable for catching the various species occurring at the station, and relatively reliable for assessing the proportion of the different species occurring there (Benejam et al., 2012). The sampling method per Abundance Grab Sampling (AGS) was applied from 2005 onwards. This method, based on the partial exploration of the stations allows to provide a representative sample of the population with respect to richness, composition and abundance in large watercourses (Tales, 2009; Le Pichon et al., 2012). Thus, the fishing was performed on one hundred points within each sampling station, representing nearly a coverage area of 1250 m2 per station. Although the observed species richness in a single pass generally underestimated the species present in a specific stream reach, richness and composition metrics (such as density of tolerant individuals used in this paper) displayed much less methodological variations (sampling crews, equipment, etc.) and are much easier to estimate precisely than CPUE (Benejam et al., 2012). It has been shown recently that although CPUE estimates depended on crew (and are likely to be difficult to calibrate), observed species richness and species composition were barely affected (Benejam et al.,

S. Azimi, V. Rocher / Science of the Total Environment 542 (2016) 955–964

957

Fig. 1. Monitoring network for evolution of fish assemblage in the river Seine.

2012). These results make possible cross-period comparisons during the 1990–2013 period. The fishing method followed a European standardized method (CEN 14011). Fishing were performed by a single-pass electrofishing using a portable unit which generated up to 200 V and 3 A pulsed D.C. in an upstream direction, covering the whole wetted width of the 100-m long reaches surveyed at each location. Benejam et al. (2012) suggest that although capture probability depends heavily on a number of factors (such as species, size, and sampling site) and needs careful consideration, the effect of electrofishing crew is negligible for assessment of species richness and composition but considerable for fish abundance. Captured fish are sorted by species, counted and classified by size (total length) for survey purposes. Their weights are then determined using charts that make it possible to reduce the capture time and increase their post-release survivability. During the first half of the survey, from 1990 to 1999, the fishing operations took place twice a year, in spring and autumn. From year 2000 onwards, fishing was carried out only once a year, during the spring period.

density of invertivorous individuals, density of omnivorous individuals, and total density of individuals. The IBI concept is widely adaptable, but that metrics must be chosen to reflect regional differences in fish distribution and assemblage structure (Miller et al., 1988; Simon and Lyons, 1995; Oberdorff and Hughes, 1992; Belliard and Roset, 2006). During the fishing trips, each metric has been evaluated and compared with what it would be if the environment was not affected by anthropic pressure. The IBI has a zero score when the assemblage being studied fully matches the expected assemblage in a reference situation. As the environment degradation proceeds, the score increases. The IBI is the sum of all these differences. The watercourses are classified into five categories depending on their rating (Table 1). In this work, the calculation of the IBI is based on the recognition of 34 species or groups of species, providing the best possible picture of the habitats in metropolitan France. Details concerning the development of

Table 1 Ecological status classes according to the Index of Biotic Integrity.

2.3. Index of Biotic Integrity (IBI) The Index of Biotic Integrity (IBI) is a standardized tool (French standard NF T 90-344) being used by ONEMA for classifying and comparing the metropolitan rivers with each other. It is also included in the decree of 25 January 2010 relating to the assessment of the ecological status of surface water. The IBI is based on the measurement of the gap between the composition of the registered fish assemblage and the composition of the assemblage naturally occurring in the watercourse in the absence of disturbances induced by human activities (Karr, 1981; Karr, 1991; Scott and Hall, 1997; Oberdorff et al., 2002). This index is calculated from 7 metrics which are total number of species, number of rheophilic species, number of lithophilic species, density of tolerant individuals,

Watercourse are classified into five categories depending on their rating. Colors are used to describe each class with respect to the ecological statues of the WFD.

958

S. Azimi, V. Rocher / Science of the Total Environment 542 (2016) 955–964

the IBI used in this paper are available in Oberdorff et al. (2001) and Oberdorff et al. (2002). 2.4. Analysis of fish muscle Three families of micropollutants have been looked for, namely metals, PCBs and pesticides (Table 2). Dioxins (PCDD), furans (PCDF) and dioxin-like PCBs (PCB-DL) were first looked for in 2007. The contamination level in the muscle of fish from the river mainly concerns three fish species, namely roach, eel and chub. In particular, the latter is required for the implementation of the EROD protocol and may also become a substitute species when insufficient numbers of roach are caught. The number of individuals per fishing trip and site ranges from 8 to 15. The quantities of fish are adjusted depending on the catches in order to obtain uniform size ranges while allowing to get enough muscle for the assays. In order to minimize the sources of biological and analytical variability, the protocol chosen for muscle was adopted from the procedure used by the German Federal Environment Agency, which consists in making several analyses on one sample, the average of the results becoming the station's official value. To do this, a given amount (30 g) of muscles is taken from each fish. All samples are mixed together and divided into three sub-samples and the mean value of micro-pollutant concentrations of the three replicates is kept. When the concentrations are below the quantification threshold in one or two of the three replicates, the mean value is calculated using the quantification value. 2.5. EROD biomarker Since 2004, the analysis of the enzyme activity of the Ethoxyresorufin-O-deethylase (EROD) has been done. The EROD activity is a biomarker giving information on effects of pollutants on the sentinel species and reflects the fish exposure to several families of toxic substances (Sanchez and Porcher, 2009). Other biomarkers of effects, such as DNA injuries or acetyl cholinesterase are specific to some pollutants while the EROD activity is a nonspecific pollutant biomarker (Vasseur and Cossu-Leguille, 2003). The EROD index was calculated from the analysis of protein concentration in the chub livers. After sacrificing each fish and determining its gender, the liver samples (10) were enclosed in a dry condition within polypropylene tubes and frozen in liquid nitrogen before being dispatched to the laboratory of the French National Research Institute of Science and Technology for Environment and Agriculture (IRSTEA) to be assayed. The livers were homogenized in the laboratory (ball mill). 2.6. Seine River quality monitoring Water quality parameters, such as dissolved oxygen, ammonia nitrogen (NH+ 4 ) and biochemical oxygen demands (BOD) have been measured at 12 stations from upstream to downstream of the Paris Table 2 List of investigated parameters. Family of pollutant

Type

Metals Organochlorines

Mercurya, zinc PCB: indicator (28, 52, 101, 138, 153, 180) and dioxin-typea (12 congeners) Dioxinsa (7 congeners) and furansa (10 congeners) 1,2,4 trichlorobenzene, 1,2,4,5 tetrachlorobenzene, DDT (4 compounds), total aldrin, dieldrin, endrin, endosulfan, heptachlorine and epoxidea, total hexachlorobenzenea, hexachlorobutadienea, hexachlorocyclohexane, lindane, metoxychlore, oxychlordane, pentachlorobenzene

Pesticides (19 compounds)

In bold characters: priority substances as specified in “Directive 2013/60/EC of 12 August 2013 amending the Directives 2000/60/EC and 2008/105/EC on priority substances”. a Compounds having an Environmental Quality Standard (EQS) for the biota under Directive 2013/60/CE of 12 August 2013.

conurbation in the Seine River. This monitoring has been performed from 1985 to 2013. The dissolved oxygen have been measured in situ while a 1 L sample have been taken each month at the SIAAP certified laboratory for analysis, in accordance with the official French standards (AFNOR). 3. Results and discussion 3.1. Seine River quality over the past 20 years The Seine River water quality has been monitored from 1985 to 2013 by the SIAAP from upstream to downstream of Paris conurbation and the 9th percentile for two pairs of years have been reported, namely years 1985/1986 and 2012/2013 (Fig. 2). This monitoring shows that the water quality of the Seine River is slightly modified by the Paris anthropogenic activity. Thus, looking at the 2012–13 period, even if the water quality level change after the 73th km downstream from Paris, the status still remain good for the dissolved oxygen and the BOD, only ammonia nitrogen passes into the fair quality level for the next 20 km. This variation is mostly induced by the discharge of the SAV treatment plant. However, this does not influence dissolved oxygen concentration which never goes under 7 mg O2·L−1 (9th percentile). These results show that the recent development and modernization of wastewater treatment processes considerably improved the water quality of the Seine River even after the Paris conurbation. Looking at the 1985/86 period, the water quality level reached the very poor status, mostly after the SAV treatment plant discharge. The 9th percentile of the dissolved oxygen concentration at the station situated 73 km downstream of Paris City was measured at 1.4 mg O2·L−1. As a result, ammonia nitrogen and BOD flows dumped by the SIAAP treatment plants have been estimated at around 4.9 t of N and 21 t of O2 respectively in 2014 while there were estimated at around 78.7 tons of N and 81 tons of O2 in 1997. 3.2. Temporal variation of fish assemblage 3.2.1. Temporal variation of the number of individuals and species As for the water quality, the number of species inhabiting the Seine River has been monitored from 1990 to 2013 (Fig. 3). Two major periods can be distinguished on this plot. During the first period, from 1990 to 2000, the number of identified species gradually increased, rising from 12 to 20. Over the second period, i.e. since 2001, the number of species range from 16 to 21. The mean values obtained for the first period (16) is significantly different than the mean value calculated for the second period (19). The difference has been statistically tested with a Mann–Whitney test (U-value of 16, critical U-value of 37, p ≤ 0.05). The total number of species follows the same pattern with an increase and one newly reported species every year during the 90s. Then, since 2000, after a fruitful year in 2001 with four newly reported species, this figure remains nearly constant and reaches 32 species. Moreover, the rise in the number of species has been observed while the frequency of sampling has decreased (spring and autumn for the period 1990–1999 and only spring for 2000–2013). The global increase of individuals during the survey period may reflect the changes made to the wastewater management and the gradual expansion of the treatment plants' capacity. A sharp improvement of the river quality and a diversification of fish fauna could be achieved through these steps. Indeed, anthropogenic eutrophication and seasonal hypoxia can affect fish growth, reproduction and behavior and thereby affect population dynamics and community composition (Arend et al., 2011). The banks of the river Seine, however, are highly anthropized and are not suitable for creating relevant quiet zones. This is the reason why the number of individuals or even the number of species no longer increases every year. The total number of caught fish varies greatly depending on the years. This figure is related to the environmental conditions throughout

S. Azimi, V. Rocher / Science of the Total Environment 542 (2016) 955–964

959

Fig. 2. 9th percentile of the concentration of dissolved oxygen (mg O2.L−1), ammonia nitrogen (mg N·L−1) and Biochemical Oxygen Demand (mg O2·L−1) in the Seine River from upstream to downstream of the Paris conurbation during two periods (1985/86 and 2012/13). The 8 fishing sampling sites have been reported (arrows).

the year, as well as to the events during the spawning period. 1996, for instance, was featured by an upsurge in the number of juvenile bleaks. This event excluded, the number of caught individuals varied from 381 to 2587 in the last 20 years. The mean value is 1324 (±575) with two population peaks and a strong decline seen 9 years apart. The first

peak, in 1998, was induced by the occurrence of numerous juvenile bleaks and roaches. The second one, in 2007, mainly comprises juvenile roaches. The decline, occurred in 2002, was linked to the lack of bleak and roach (less than 60 and 40 fished respectively). Since 2010, the number of caught individuals is above 1100 each year. This amount is

Fig. 3. Number of species and individuals reported from the river Seine per fishing year and cumulative figures over the survey period (1990–2013).

960

S. Azimi, V. Rocher / Science of the Total Environment 542 (2016) 955–964

in the same order of magnitude than the number of individuals caught before 1999 when fishing operations took place twice a year. 3.2.2. Species composition Two categories have been distinguished among the listed species, namely the limnophilic species and the rheophic species. The species in the first category live and spawn in rocky habitats and prefer standing water, whereas those in the second category swim in streams. As regards their diets, two categories have been distinguished as well. The carnivorous species feed on living organisms, whereas the omnivorous species enjoy all sorts of food. For instance, the carnivorous limnophilic or carnivorous rheophilic species are sensitive to the habitat quality (sloughing, suspended solids, etc.) and have a selective diet. On the other hand, the omnivorous limnophilic species are relatively undemanding with respect to the habitat. As regards the omnivorous rheophilic species, they are somewhat demanding with respect to water aeration and flow (Belliard et al., 1999; Arend et al., 2011). In order to get a more precise characterization of the assemblages, the species have been categorized according to their living and feeding patterns (Fig. 4). As a rule, over the last 20 years, the prevailing species in the river Seine are limnophilic and omnivorous, hence undemanding in terms of water quality and diets. This group accounts on average for 72 (± 19)% of the reported species. 23 (± 16)% of the species tend to be limnophilic and carnivorous. Since 1997, the species from that family are gradually increasing in number, but they receded in 2011. That increase is linked to the growth of the populations of eel, which is quite common in the river Seine. This finding is in line with the results recorded by Beslagic et al. (2013) who have determined that the present (in 2010) fish associations around the capital are represented by a larger proportion of limnophilic species. As regards the rheophilic and omnivorous species, the percentage fluctuates from year to year, being substantially influenced by the yearly reported number of gudgeon, chub and dace. Their share, however, has substantially increased because of the continuous establishment of the golden orfe since 2009. The rheophilic and carnivorous species still remain marginal in the river. Since 2006, however, the sculpin repeatedly occurs, accounting for nearly 2.5% of the individuals in 2013. Thus, the fish species assemblage has changed during the 23-year survey period. Over the years, an emergence of more sensitive species has been observed, mostly since 2009. 3.3. Index of Biotic Integrity (IBI) The Index of Biotic Integrity (IBI) has been calculated based on all the above-described fishing stations. For the sake of clarity, only those data that have been collected upstream and downstream of the capital have been displayed (Fig. 5).

Fig. 4. Distribution of the occurrence of individuals according to their ways of life (limnophilic (L)/rheophilic (R)) and diets (carnivorous(C)/omnivorous (O)).

In the light of the results at the upstream station (Villeneuve-St.Georges), the IBI fluctuates between the poor status and the very good ecological status. The data collected since 2011 provide values that are at least equivalent to the good ecological status (as in the late 90's), although the IBI tends towards a fair quality (the IBI value was 16.4 in 2013, the good quality threshold is 16). The unfavorable metrics are the number of rheophic and lithophilic species and the density of individuals. Such behavior has also been reported for other watercourses (Le Pichon et al., 2012). This trend, with a good water quality since 1985 (dissolved oxygen level was around 6 mg O2.L−1), may be induced by the fact that the bank of the river are unattractive to fish fauna in this area. At the downstream site (Triel-sur-Seine, 88 km downstream of Paris City), the IBI varies between the poor and the fair ecological status from 2000 to 2011. The improvement all over the studying period is not obviously observed (coefficient of determination R2 = 0.41, n = 14) while in 2012 and 2013, the IBI level reaches the very good quality level. 3.4. Analysis of muscles 3.4.1. Heavy metal contamination Screening tests of heavy metal – mercury and zinc – are being conducted on eel and roach muscles since 2000 as well as on chub since 2009 at all three stations (upstream, near downstream and far downstream) near Paris (Fig. 6). The results are expressed in mg of metallic element per kilogram of wet weight (WW) of the relevant species. Years for which the results are not displayed correspond to lack of the species during the specific fishing campaign. In the case of mercury, the European Directive 008/105/EC of 16 December 2008 lays down an Environmental Quality Standard (EQS) of 0.02 mg/kg WW in the fish tissues. The mercury concentrations, ranging from 0.02 to 0.43 mg/kg WW, are generally excessive at all the sites, species and for all the years. The concentrations are substantially higher in the muscles of fish that are caught far downstream of Paris (Triel-sur-Seine). From 2003 to 2013, for eel, the concentrations at this site are 26–64% higher than the mean concentration at the other two sites, 38–67% higher for roach (except in 2010) and 28–81% higher for chub (since 2010). The zinc concentrations are similar throughout the Paris conurbation. They range from 5.8 and 27 mg/kg WW for all the years, sites and species. Moreover, there is no significant difference between sites located upstream and downstream of Paris (Student test, α = 0.01). That difference in the behaviors of the two metallic compounds is probably due to the nature of the element being considered. The point is that the non-essential metals (mercury) are toxic, whereas the essential metals tend to be controlled by fish metabolism (Fernandes et al., 2007). Because of the presence of some ligands, e.g. metallothionein, which is involved in sequestration of metals such as Cu, Cd and Zn (Engel and Roesijadi, 1987), liver provides more evidences of the accumulation of these metals than muscles (Ploetz et al., 2007; Uysal et al., 2009). Besides, for reasons of public health or consumption, this survey, as most of studies, has focused on the measurement of the accumulation of metals in the muscle tissues (Storelli et al., 2006; Keskin et al., 2007). 3.4.2. PCB contamination Fig. 7 displays the mean concentrations of indicator PCBs (PCB 28, 52, 101, 138, 153, 180) in eel, roach and chub muscle per kilogram of wet weight and fatty tissue over the 2000–2013 period. The resulting sum for each of the 3 sites (Villeneuve-St.-Georges, Levallois and Trielsur-Seine) has been averaged in order to provide a value for the river Seine and for each species. The threshold values regarding the PCB concentrations in the fish muscle are laid out by the EU Regulation No. 1259/2011 of 02 December 2011 relating to maximum levels in foodstuffs. The limit values (300 μg/kg WW for eel and 125 μg/kg WW for roach and chub) are mentioned in the Figure. As regards the eel muscle, the PCB indicators always exceed the regulatory provisions. The resulting figures range from

S. Azimi, V. Rocher / Science of the Total Environment 542 (2016) 955–964

961

Fig. 5. Temporal variation of the Index of Biotic Integrity (IBI) upstream (Villeneuve-St.-Georges, −10 km from Paris City) and downstream (Triel-sur-Seine, +88 km from Paris City) of the Paris conurbation from 1990 to 2013.

361 to 1369 μg/kg WW over the 2000–2013 period. The time variation even seems to reveal an increase with the highest value observed in 2013 (1369 μg/kg WW). As regards the roach and chub muscle, the prescribed thresholds are not reached until 2012 and the concentrations range from 27 to 39 μg/kg WW. The concentrations measured in 2013 (216 μg/kg WW for chub) exceed the prescribed thresholds and seem to corroborate the increase found in eel muscle. Unfortunately, no roach was caught in 2013 to give an opportunity to make a comparison. As regards fatty tissue, the concentrations are 1000 times higher than in wet weight and range from 1.5 and 17 mg/kg FT (over the 2001–2013 period, the concentrations in 2000 have been excluded). In addition, the differences of concentrations among the species are smaller and highlight similar uptakes of PCBs in the fatty tissues of the various species. The binding of organic pollutants preferentially to the lipids have already been reported by Lazartigues (2010). The concentrations of dioxins, furans and dioxin-like PCBs have been measured in the eel and chub muscle at Villeneuve-St.-Georges, Levallois and Triel-sur-Seine from 2007 to 2013. The results of the sum of these 3 types of pollutants are provided in toxicity equivalent (toxicity calculated in relation to that of the 2,3,7,8-T4CDD molecule with the equivalence factors as defined by the World Health Organization in 2005 per gram of wet weight (WW) and per gram of fatty tissue (FT) (Table 3). Regardless of the measuring station, no variation in time can be seen for the two species being considered. The concentrations range from 14 to 64 pg/g WW in eel and from 1.2 to 9.4 pg/g WW in chub. The concentrations in eels exceed the prescribed threshold in foodstuffs as laid out by the EU Regulation No 1259/2011 of 2 December 2011 (10 pg/g WW). For chub, only the concentration measured at Triel in 2013 has a higher value (threshold set to 6.5 pg/g WW). The results in the eel muscle have a gradient from upstream to downstream. The concentrations measured at Triel are systematically higher than the concentration determined at Villeneuve-St.-Georges. That trend is valid but less visible in chub. The concentrations in the fatty tissue have the same order of magnitude in both species and are 10 times higher than the concentrations in the wet weight. For both species, the concentrations range from 105 and 395 pg/g FT (TEQ 2005) excluding the concentration in chub in 2013 at the Triel-sur-Seine station. The latter sample shows a high level of

contamination induced by the high amount of PCB 126. The increase from upstream to downstream is not as clear as what was seen with the WW. 3.4.3. Pesticide contamination The mean concentrations of total pesticides, including total DDTs and hexachlorobenzene, have been measured per kilogram of wet weight (WW) and per kilogram of fatty tissue (FT) in eel, roach and chub (Table 4). The results for all 3 stations (Villeneuve-St.-Georges, Levallois and Triel-sur-Seine) have been averaged per year over the 2002–2013 period. Eel is more contaminated by pesticides (in wet weight) than chub, as regards the number of quantified compounds (7–12 for eel, 3–4 for chub and roach) as well as the measured concentrations. For instance, the mean concentrations of total pesticides range from 9 to 600 μg/kg WW for eel whereas they range from 0.2 to 15 μg/kg WW for roach and chub. Even in the fatty tissue, the concentrations in eel are higher than in the other two species, except in years 2009 and 2012. Among the pesticides, DDTs are the main compounds occurring in any considered species. The DDT share in the total pesticides ranges from 47 to 100%, except in years 2010 and 2012 in chub. The EQSs are complied with for hexachlorobenzene (10 μg/kg WW) in all three 3 stations. Likewise, hexachlorobutadiene, a compound that has an EQS (55 μg/kg WW), has never been detected during the campaigns (QL = 2 μg/kg WW). 3.5. EROD biomarker The EROD biomarker has been measured in chub livers at all three sites of the survey from 2004 to 2013 (Fig. 8). The EROD activities of the females in the survey have been corrected according to the gonadosomatic index (GSI). That correction is used for enhancing the values in the females whose EROD activity is partly inhibited by the sexual activity. The EROD index changes during the ten years of the survey between the good and the poor ecological status. No variation can be found from upstream to downstream of the Paris conurbation. Apart from years 2009 and 2013, the index fluctuates between the poor status and the

962

S. Azimi, V. Rocher / Science of the Total Environment 542 (2016) 955–964

Fig. 6. Variation of mercury and zinc concentrations in the eel, roach and chub muscle at Villeneuve-St.-Georges, Levallois/Paris and Triel-sur-Seine over the 2000–2013 period.

very good status. The upstream site has a poor index (88 pmol/mn/mg protein). In 2013, all three sites have poor indices with geometric averages of 181, 101 and 113 pmol/mn/mg protein respectively in

Villeneuve-St.- Georges, Levallois and Triel-sur-Seine. However, since this biomarker is a one-off indicator, the results in 2013 can be due to fairly outstanding hydromorphological variations during the fishing

Fig. 7. Evolution of mean indicator PCB contamination per kg of wet weight (WW) and per kg of fatty tissue (FT) in the eel, roach and chub muscle in the river Seine over the 2000–2013 period. The contamination thresholds as laid out by the EU Regulation No. 1259/2011 are mentioned for eel (300 μg/kg WW) and chub (125 μg/kg WW)).

S. Azimi, V. Rocher / Science of the Total Environment 542 (2016) 955–964

963

Table 3 PCB concentration (sum of the dioxins, furans and dioxin-like PCBs) in toxicity equivalent (TEQ 2005) per kg wet weight (WW) and per kg of fatty tissue (FT) at all 3 monitoring stations around Paris (Villeneuve-St.-Georges, Levallois and Triel-sur-Seine) over the 2007– 2013 period.

pg/g WW (TEQ 2005) Villeneuve St. G Eel Chub Levallois Eel Chub Triel-sur-Seine Eel Chub pg/g FT (TEQ 2005) Villeneuve St. G Eel Chub Levallois Eel Chub Triel-sur-Seine Eel Chub

2007

2008

2009

2010

2011

2012

24.6

20.3 31.4

25.2 2.4 38.5

37.4

64.2

36.1 5.1

31.4

22.7 1.9 21.4 1.2 32.0 1.9

14.2

21.3

26.9 2.2 31.2

263

317

153

116

195

288

115 150 142

134 251 395

135 244

286

302 184 135 173 189 218

2013 23.0 3.5 23.4 2.5 27.9 9.4

18.3 1.5 23.5 2.7

219 164 117 153 235

150 219 105 165 234 1002

Fig. 8. Evolution of EROD index (pmol/mn/mg protein) at the sites of the survey (Villeneuve-St.-Georges, Levallois and Triel-sur- Seine) over the 2004–2013 period.

remain week since the concentration is higher than 7 mg O2·L−1 90% of the time. Such changes in the water quality have an influence on the fish species assemblage. Thus, the emergence of a more sensitive assemblage is observed with an increasing number of individuals in such species as bleak and chub and the increasing presence of such species as ruffe, scalpin and pike-perch, slowly settling in the habitats. Moreover, the IBI of the river Seine has reach the good quality level over the last few years, mostly downstream of the Paris conurbation. However, the water quality improvement seems to have had no influence on the pollutant concentrations in the fish muscles. Irrespective of the considered pollutant, no temporal variation has been observed while an exceedance of the environmental quality standards for mercury and indicator PCBs were evidenced. The study of the EROD induction reveals a sharp increase over the 2012–2013 period, but it seems it does not necessarily reflect the overall status of the river.

period. The EROD inductions have been correlated to the dioxine and dioxin-like PCB concentrations because these two families can be the cause for EROD activity. However, the trend has not shown anything, probably because EROD activity produces synergy/inhibition effects together with other micropollutants. 4. Conclusions Over the past 20 years, considerable investments have been made in Paris conurbation to improve waste water treatment processes in accordance with the European Framework Directive. These efforts have allowed nearly achieving the good ecological state for the Seine River in 2014. As an example, ammonia nitrogen and BOD flows dumped by the waste water treatment plants of SIAAP have been estimated at around 4.9 tons of N and 21 tons of O2 respectively. Moreover, the impact of these contributions on the dissolved oxygen in the Seine River

Table 4 Concentrations of total pesticides, including total DDTs and hexachlorobenzene per kg of wet weight (WW) and per kg of fatty tissue (FT) for roach, chub and eel in Paris (mean concentrations at the Villeneuve-St.-Georges, Levallois and Paris stations) over the 2002–2013 period.

μg/kg WW Eel Total pesticides Total DDTs Hexachlorobenzene Roach Total pesticides Total DDTs Hexachlorobenzene Chub Total pesticides Total DDTs Hexachlorobenzene μg/kg FT Eel Total pesticides Total DDTs Hexachlorobenzene Roach Total pesticides Total DDTs Hexachlorobenzene Chub Total pesticides Total DDTs Hexachlorobenzene

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

600 577 3.7

64 49 2.6

42 40 1.4

45 42 2.7

78 77 1.1

29 25 2.4

95 80 5.1

78 67 0.4

9 4 1.3

24 16 2.5

3.6 3.6 0.0

0.9 0.8 –

8.9 8.7 0.1

7.5 7.0 0.5

9.4 8.8 0.3

1.5 1.5 0.1

1.6 1.4 0.1

2.6 2.5 0.1

– – –

–

–

–

–

–

–

–

1.70 1.60 0.10

2012

2012

2013

46 35 10.3

52

66

– – –

1.3 1.2 0.1

9.8

–

0.13 0.03 0.03

0.19 0.16

1.60 0.43 0.10

11.90

14.76

1.8

0.24

2026 2019 33

973 950 9

703 696 23

432 427 23

660 660 10

228 215 12

563 552 26

338 317 9

77 77 11

176 171 15

269 280 59

300 –

524 518 6

133 108 –

449 431 19

316 311 5

617 600 –

163 155 8

135 84 6

99 93 6

– – –

– – –

130 116 14

1140

177 153

136 113 23

338

15

437 117 117

– – – –

964

S. Azimi, V. Rocher / Science of the Total Environment 542 (2016) 955–964

Acknowledgments The authors are grateful to Tiphaine Vérité and Alain Verger (SIAAP) for their assistance in this work. We thank two anonymous reviewers for helpful comments on the manuscript.

References Arend, K., Beletsky, D., DePinto, J., Ludsin, S.A., Roberts, J.J., Rucinski, D.K., Scavia, D., Schwab, D.J., Höök, T.O., 2011. Seasonal and interannual effects of hypoxia on fish habitat quality in central Lake Erie. Freshw. Biol. 56, 366–383. Belliard, J., Roset, N., 2006. Form of presentation and use of Fish Reference Index (FRI). National Office for Water and Aquatic Environments Internal Report (24 pp.). Belliard, J., Berrebi, R., Monnier, D., 1999. Fish communities and river alteration in the Seine Basin and nearby coastal streams. Hydrobiologia 400, 155–166. Benejam, L., Alcaraz, C., Benito, J., Caiola, N., Casals, F., Maceda-Veiga, A., de Sosta, A., Garcia-Berthou, E., 2012. Fish catchability and comparison of four electrofishing crews in Mediterranean streams. Fish. Res. 123-124, 9–15. Beslagic, S., Delaigue, O., Belliard, J., 2013. Long term evolution of fish populations on the Seine basin (in French). PIREN Seine research Program internal Report (10 pp.). Colin, N., Porte, C., Fernandes, D., Barata, C., Padros, F., Carrason, M., Monroy, M., CanoRocabayera, O., de Sostoa, A., Pina, B., Maceda-Veiga, A., 2016. Ecological relevance of biomarkers in monitoring studies of macro-invertebrates and fish in Mediterranean rivers. Sci. Total Environ. 540, 307–323. Couillard, C., 2009. Utilisation des poissons pour évaluer les effets biologiques des contaminants dans l'Estuaire du Saint-Laurent et le fjord du Saguenay. J. Water Sci. 22 (2), 291–314. Engel, D.W., Roesijadi, G., 1987. Metallothionein: a monitoring tool. In: Vernberg, F.J., Calabrese, A., Thurberget, F.P., Vernberg, W.B. (Eds.), Pollution Physiology of Estuarine Organisms. University of South Carolina Press, pp. 42l–438. Fernandes, C., Fontaínhas-Fernandes, A., Peixoto, F., Salgado, M.A., 2007. Bioaccumulation of heavy metals in Liza saliens from the Esmoriz−Paramos coastal lagoon, Portugal. Ecotoxicol. Environ. Saf. 66, 426–431. Hering, D., Buffagni, A., Moog, O., Sandin, L., Sommerhauser, M., Stubauer, I., Fald, C., Johnson, R., Pinto, P., Skoulikidis, N., Zahradkova, Z., 2003. The development of a system to assess the ecological quality of streams based on macroinvertebrates — design of the sampling programme within the AQEM project. Int. Rev. Hydrobiol. 88 (3–4), 345–361. Hering, D., Borja, A., Carstensen, J., Carvalho, L., Elliott, M., Feld, C.K., Heiskanen, A., Johnson, R.K., Moe, J., Pont, D., Solheim, A.L., van de Bund, W., 2010. The European Water Framework Directive at the age of 10: a critical review of the achievements with recommendations for the future. Sci. Total Environ. 408, 4007–4019. Karr, J.R., 1981. Assessment of biotic integrity using fish communities. Fisheries 6, 21–27. Karr, J.R., 1991. Biological integrity: a long-neglected aspect of water resource managment. Ecol. Appl. 1, 66–84. Keskin, Y., Baskaya, R., Ozyaral, O., Yurdun, T., Luleci, N.E., Hayran, O., 2007. Cadmium, lead, mercury and copper in fish from the Marmara Sea, Turkey. Bull. Environ. Contam. Toxicol. 78, 258–261.

Lazartigues, A. Pesticides et polyculture d'étang: de l'épandage sur le bassin versant aux résidus dans la chair de poisson. Ph.D. Dissertation, Lorraine Polytechnique National Institut, France, 2010. Le Pichon, C., Tales, E., Belliard, J., Gorges, G., Zahm, A., Clement, F., 2012. La distribution spatiale des peuplements de poissons dans les petits bassins versants. Sciences Eaux et Territoires 24-33. Miller, D.L., Leonard, P.M., Hughes, R.M., Karr, J.R., Moyle, P.B., Schrader, L.H., Thompson, B.A., Daniels, R.A., Fausch, K.D., Fitzhugh, G.A., Gammon, J.R., Halliwell, D.B., Angermeier, P.L., Orth, D.J., 1988. Regional applications of an index of biotic integrity for water resource management. Fisheries 13, 12–30. Oberdorff, T., Hughes, R.M., 1992. Modification of an index of biotic integrity based on fish assemblages to characterize rivers of the Seine Basin, France. Hydrobiologia 228, 117–130. Oberdorff, T., Pont, D., Hugueny, B., Chessel, D., 2001. A probabilistic model characterizing riverine fich communities of French rivers: a framework for environmental assessment. Freshw. Biol. 46, 399–415. Oberdorff, T., Pont, D., Hugueny, B., Porcher, J.P., 2002. Development and validation of a fish-based index for the assessment of ‘river health’ in France. Freshw. Biol. 47, 1720–1734. O'Farrell, I., Lombardo, R.J., Tezanos Pinto, P., Loez, C., 2002. The assessment of water quality in the lower Lujan River (Buenos Aires, Argentina): phytoplankton and algal bioassays. Environ. Pollut. 120, 207–218. Ploetz, D.M., Fitts, B.E., Rice, T.M., 2007. Differential accumulation of heavy metals in muscles and liver of a marine fish (King Mackerel, Scomberomorus cavalla, Cuvier) from the Northern Gulf of Mexico, USA. Bull. Environ. Contam. Toxicol. 78, 134–137. Polard, T. Caractérisation des effets génotoxiques sur poissons de produits phytosanitaires en période de crue. Ph.D. Dissertation, Toulouse University, France, 2010. Rocher, V., Paffoni, C., Azimi, S., Gonçalves, A., Rousselot, O., Marcel, C., Gasperi, J., Lestel, L., Flipo, N., Mouchel, J.M. Evolution de la qualité de la Seine en lien avec les progrès de l'assainisssement de 1970 à aujourd'hui (in French). Eds AESN, 2015, (in press). Sanchez, W., Porcher, J.-M., 2009. Utilisation des biomarqueurs pour la caractérisation de l'état écotoxicologiques des masses d'eau. Techniques Sciences et Méthodes 6, 29–38. Sanchez, W., Bado-Nilles, A., Porcher, J.-M., 2012. Biomarqueurs chez le poisson: un outil d'intérêt pour le contrôle d'enquête. La Houille Blanche 10, 49–55. Scott, M.C., Hall Jr., L.W., 1997. Fish assemblages as indicators of environmental degradation in Maryland coastal plain streams. Trans. Am. Fish. Soc. 126, 349–360. Simon, T.P., Lyons, J., 1995. Application of the index of biotic integrity to evaluate water resource integrity in freshwater ecosystems. In: Davis, W.S., Simon, T.P. (Eds.), Biological Assessment and Criteria — Tools for Water Resource Planning and Decision Making. Lewis Press, Boca Raton, FL, U.S.A., pp. 245–262. Storelli, M.M., Barone, G., Storelli, A., Marcotrigiano, G.O., 2006. Trace metals in tissues of Mugilids (Mugil auratus, Mugil capito, and Mugil labrosus) from the Mediterranean Sea. Bull. Environ. Contam. Toxicol. 77, 43–50. Tales, E., 2009. Tendances d'évolution des peuplements de poissons de la Seine en réponse à la variabilité hydroclimatique. Hydroécologie appliquée 16, 29–52. Uysal, K., Köse, E., Bülbül, M., Dönmez, M., Erdögan, Y., Koyun, M., Omeröglu, C., Ozmal, F., 2009. The comparison of heavy metal accumulation ratios of some fish species in Enne Dame Lake (Kütahya/Turkey). Environ. Monit. Assess. 157, 355–362. Vasseur, P., Cossu-Leguille, C., 2003. Biomarkers and community indices as complementary tools for environmental safety. Environ. Int. 28, 711–717.