- Split View
-
Views
-
Cite
Cite
Daniela Bănaru, François Carlotti, Aude Barani, Gérald Grégori, Nada Neffati, Mireille Harmelin-Vivien, Seasonal variation of stable isotope ratios of size-fractionated zooplankton in the Bay of Marseille (NW Mediterranean Sea), Journal of Plankton Research, Volume 36, Issue 1, January/February 2014, Pages 145–156, https://doi.org/10.1093/plankt/fbt083
- Share Icon Share
Abstract
Stable isotope ratios of six size fractions of zooplankton (80 to >2000 µm) were analyzed seasonally in 2009–2010 at the SOMLIT site in the Bay of Marseille. Isotopic ratios generally increased with zooplankton size. The highest δ15N values were observed in the 1000–2000 µm fraction. The largest size class (>2000 µm), dominated by gelatinous plankton, had lower δ15N values due to the low isotopic signatures of most of these organisms. In the larger size fractions (>1000 µm), isotopic ratios were measured at the taxon level. Brachyuran, stomatopod, teleost and cephalopod larvae showed the highest δ15N values, and salps and pteropods the lowest ones. Lower values of both δ13C and δ15N were recorded in winter and spring than in summer and autumn for all fractions. Seasonal variations were consistent with the fluctuations of environmental parameters (temperature, nutrients, Chl a concentration) and were related to phytoplankton and seawater particulate organic matter (POM) composition. Stable isotope and flow cytometry analysis of water POM indicated that sewage wastewater particles were mixed with marine phytoplankton at the SOMLIT site and transferred up into the zooplanktonic food web.
INTRODUCTION
The Mediterranean Sea is considered an oligotrophic sea, with low nutrient concentrations and relatively moderate levels of primary production (Estrada, 1996). Fertilization processes that may take place locally are mostly related to wind-driven upwelling events and land runoff due to river discharge and seasonal storms. Coastal environments, highly variable and complex, are exposed to various anthropogenic and land-related influences (sewage discharges, rivers, etc.). The marine coastal area near Marseille, south of France, is severely affected by organic matter inputs and industrial contamination linked to the human activities of this large conurbation (>1.8 × 106 inhabitants) (Syakti et al., 2012). This area is also influenced by the Northern Current (Millot, 1999) and the coastal Huveaune River inputs.
These local environmental conditions influence both phytoplankton and zooplankton production and their seasonal variations (Calbet et al., 2001). Composition, biomass and seasonal variations of zooplankton have been well studied in the NW Mediterranean Sea (Razouls and Kouwenberg, 1993; Champalbert, 1996; Calbet et al., 2001; Gaudy et al., 2003; Saiz et al., 2007; Fanelli et al., 2011). Champalbert (Champalbert, 1996) shows that zooplankton diversity is usually lower in coastal neritic waters than in oceanic waters, whereas biomass is higher due to terrestrial inputs.
Zooplankton play a prominent role in marine pelagic systems due to their trophic position and their role in organic matter transfer from phytoplankton to upper level consumers (Cushing, 1989; Saiz et al., 2007). However, they also represent a major pathway for contaminants in the food web and act as biomagnification agents of most contaminants between low and high trophic levels, as shown in the European hake food web (Cossa et al., 2012; Harmelin-Vivien et al., 2012). In these studies, zooplankton is considered as a unique homogeneous compartment, including all organisms with sizes between 200 and >2000 µm. A size-based approach has been proposed as a useful means to get obtain insight into the structure and function of marine food webs (Kerr and Dickie, 2001) and is now commonly used in plankton studies (Rau et al., 1990; Rolff, 2000; Saiz et al., 2007; Carlotti et al., 2008; among many others). Body size determines potential predator–prey interactions (Cohen et al., 1993), rates of production and natural increase in abundance (Banse and Mosher, 1980), energy requirements and metabolism (Fenchel, 1974; Boudreau et al., 1991) and mortality rates (Hirst and Kiørboe, 2002).
Provided that primary producers have distinct stable isotope ratios, carbon and nitrogen stable isotopes constitute a powerful tool for discriminating among organic matter sources (Vizzini and Mazzola, 2006). Stable isotopes have been used successfully to trace the transfer of organic matter of different origins through aquatic food webs (Fry and Sherr, 1984).
The nitrogen isotope ratio is used to help define the trophic levels of organisms, as δ15N usually increases by 2.5–4.5‰ (mean 3.4‰) from prey to predator (Minagawa and Wada, 1984; Post, 2002). As an increase in δ13C of 0.4–2‰ only occurs from diet to consumer (De Niro and Epstein, 1978; Post, 2002), a consumer's carbon isotope composition can give clues about the sources of its diet, particularly in systems with distinct organic matter sources (Fry and Sherr, 1984). The relative importance of these sources may vary in space and time, and stable isotope ratios at the base of food webs in the marine environment can be highly variable (Vizzini and Mazzola, 2006).
Stable isotope ratios have been widely used to analyze marine food webs and many studies have been performed on isotopic signatures of zooplankton (Rau et al., 1990; Fry and Quiñones, 1994; Rolff, 2000; Kibirige et al., 2002; Sommer and Sommer, 2004; Gentsch et al., 2009; among many others). However, few of them concern the Mediterranean Sea (Koppelmann et al., 2009; Fanelli et al., 2011). Most of these studies analyzed zooplankton stable isotope ratios by size classes (Fry and Quiñones, 1994; Rolff, 2000; Calbet et al., 2001; Kibirige et al., 2002; Koppelmann et al., 2009), but only a few give detailed analysis on taxonomic groups or species (Sommer and Sommer, 2004; Fanelli et al., 2011).
Thus, a size-fractionation study of zooplankton was undertaken within the general framework of the COSTAS (Contaminants in the trophic system: phytoplankton, zooplankton, anchovy, sardine) multidisciplinary program aimed at studying contaminant transfer along the food webs of the two dominant planktivorous teleosts in the Mediterranean Sea, the sardine Sardina pilchardus, and the European anchovy, Engraulis encrasicolus.
In the present study, stable isotope ratios of six different size classes of zooplankton, along with those of particulate organic matter (POM) sources, were analyzed seasonally in the Bay of Marseille and combined with zooplankton taxonomic composition and flow cytometry sorting. Our results were then related to environmental parameters and microphytoplankton composition provided by the SOMLIT program (a long-term observation survey at the sampling site).
The aims of this study were: (i) to determine the influence of season and size class on the C and N isotopic signatures of zooplankton, (ii) to relate the observed variations to zooplankton taxonomy and POM composition, (iii) to determine the influence of environmental parameters on both POM and zooplankton isotopic signatures and eventually (iv) to demonstrate, or not, the integration of sewage-derived POM in zooplankton food webs.
METHOD
Site and sampling procedure
Zooplankton was collected in the Bay of Marseille at the SOMLIT (Service d'Observation en Milieu LITtoral) site (43.24°N; 5.29°E) (Fig. 1). Nutrients, temperature, salinity, oxygen, pH, chlorophyll a and suspended particulate matter (SPM) have been recorded at this site every 2 weeks since 1994 within the frame of a long-term national program on littoral observation (http://somlit.epoc.u-bordeaux1.fr). The SOMLIT site is mainly influenced by the Northern Current (Millot, 1999; Petrenko et al., 2005), but is also subjected to anthropogenic and terrestrial inputs from the Marseille sewage treatment plant located at Cortiou and the coastal Huveaune River (Fig. 1), depending on winds and rain events (Cresson et al., 2012). Thus, three main potential sources of POM are available to zooplankton (seawater, sewage water and river runoff).
Zooplankton was vertically sampled, from 0 to 50 m depth, using a WP2 zooplankton net with a 200 µm mesh size and a PVC cod-end during the day, in spring (March 2009), summer (July 2009), autumn (October 2009) and winter (February 2010). Six zooplankton size classes (80–200, 200–300, 300–500, 500–1000, 1000–2000 and >2000 µm) were separated onboard the vessel with sieves of decreasing mesh size. The first size class (80–200 µm) covers some small organisms caught by the 200 µm net because of clogging. Six to eight tows were performed on each date to collect enough material in all size classes for stable isotope analysis. Surface water samples for POM analysis were collected at the SOMLIT site in each season and in the Cortiou sewage plume only in summer and winter for logistical reasons. The Huveaune River flow is deflected to the Marseille sewage outfall most of the time, but flows directly to the sea through its natural bed after heavy rain events. Stable isotope ratios from Huveaune River POM during the study period were taken from Cresson et al. (Cresson et al., 2012).
Sample processing
Water samples were prefiltered on a 200 µm mesh-sieve to remove large detritus. POM samples were obtained by filtering 6 L of surface water, on pre-weighed Whatman GF/F glass micro-fibre filters pre-combusted for 4 h at 500°C, to obtain a minimum of six filters per season. They were then dried at 70°C, weighed to quantify SPM content and kept in dessicators until analysis. Zooplankton samples were analyzed in the laboratory under a binocular microscope 2 h after sampling to determine the taxonomic composition of each size class. For the two largest size classes (1000–2000 and >2000 µm), organisms were separated into broad taxonomic groups to analyze them separately. The time lag between sampling and sorting allowed gut clearance. In the smaller size classes animals were pooled together for isotopic analysis. Subsequently samples were frozen at −20°C and freeze-dried.
Stable isotope analysis
Salt was separated from the freeze-dried samples which were ground into a fine powder using an agate mortar and pestle. As POM and zooplankton may contain carbonates, an acidification step using the drop-by-drop technique was used to remove 13C-enriched carbonates (De Niro and Epstein, 1978). Thus, a subsample received 1% HCl treatment before rinsing and drying, and was used for δ13C determination, whereas the other untreated subsample was used for δ15N analysis. POM was collected by scraping the surface of acidified and non-acidified filters. Three replicates were performed on POM and on each zooplankton size class per season for both δ13C and δ15N.
Stable isotope measurements were performed with a continuous-flow isotope-ratio mass spectrometer (Delta V Advantage, Thermo Scientific, Bremen, Germany) coupled to an elemental analyzer (Flash EA1112 Thermo Scientific, Milan, Italy). Results are expressed in δ notation relative to Vienna PeeDee Belemnite and atmospheric N2 for δ13C and δ15N, respectively, according to the equation: δX (‰) = [(Rsample/Rstandard) − 1] × 1000, where X is 13C or 15N and R is the isotope ratio 13C/12C or 15N/14N, respectively. For both δ13C and δ15N, measurement precision is <0.1‰ (replicate measurements of internal laboratory standards, acetanilide).
Flow cytometry analysis
Water samples from the SOMLIT site, Cortiou plume and Huveaune River were analyzed at the PRECYM flow cytometry platform (http://precym.com.univ-mrs.fr) to quantify the type and size of particles <20 µm present in POM.
Cell counts were assessed on a FACSCalibur using the CellQuest software (both from BD Biosciences), while sorting was performed with a BD Influx Mariner (BD Biosciences) cell sorter equipped with a 488 nm laser (200 mW, Coherent Sapphire). A posteriori analysis of ultraphytoplankton groups was performed with the SUMMIT v4.3 software (Beckman Coulter).
Standard protocols were used to enumerate phytoplankton (Marie et al., 2001). The red fluorescence (>640 nm or 630LP) related to chlorophyll a was used as trigger signal. Cells were enumerated according to their SSC and FSC properties (related to cell size and structure, respectively), and their orange fluorescence (564–606 nm—PE or 580/30 nm). TruCount beads (BD Biosciences®) were added to the samples to determine the volume analyzed, and for both analysis and sorting, 2 µm beads were used to discriminate picoplankton (<2 µm) from nanoplankton (>2 µm) populations, and 6 µm beads to estimate the nanoplankton particle size (both Fluoresbrite YG, Polyscience). All data were collected in log scale and three sorting gates were determined: picoplankton cells (Pico), nanoplankton with low (Nano low) and higher chlorophyll a content (Nano high). Data were stored in list mode using the BD FACS™ software (BD Biosciences).
Sorted populations were then prepared for scanning transmission electron microscopy (STEM, Hitachi S570) in order to determine their nature.
Statistical analysis
Statistical analysis was with the Statististica 9.1 software. Two-way ANOVAs were performed to assess the effect of season and size on C and N stable isotope ratios of zooplankton after testing for assumptions. One-way ANOVA was used to test for differences in isotopic signatures of water POM at the three sites sampled. Post hoc comparisons of means were performed with Student–Newman–Keuls tests (significance level P-value < 0.05). When possible, differences in mean stable isotope values of individualized zooplankton groups between size classes (1000–2000 and >2000 µm) were tested by the t-test.
RESULTS
Zooplankton composition by size class
Zooplankton composition differed between size classes (Table I). The first (80–200 µm) was dominated by detritus and small copepod stages (eggs, nauplii, copepodits), but phytoplankton and small larvae were also abundant. Copepods (>80%) and cladocerans largely dominated the 200–300 and 300–500 µm size classes. Copepods were always the dominant organisms in the 500–1000 and 1000–2000 µm size classes, but teleost eggs, crustacean larvae and gelatinous organisms such as appendicularians, siphonophores, salps and chaetognaths, increased in abundance in larger size classes. The largest size class (>2000 µm) was dominated by gelatinous organisms (siphonophores, salps, chaetognaths), pteropods and macrurid larvae, whereas large copepods were less abundant.
Size class (µm) . | Main groups . |
---|---|
80–200 | Detritus, copepods (eggs, nauplii and copepodits), phytoplankton (diatoms, dinobionts), larvae (gastropods, annelids) |
200–300 | Copepods, cladocerans, larvae (gastropods, bivalves, annelids, brachyurids) |
300–500 | Copepods, cladocerans, larvae (gastropods, bivalves, annelids, brachyurids), appendicularians |
500–1000 | Copepods, teleost eggs, larvae (brachyurids, macrurids, bivalves), euphausiids, appendicularians, chaetognaths, pteropods |
1000–2000 | Copepods, pteropods, chaetognaths, siphonophores, macrurid larvae, brachyurid larvae, euphausiids, amphipods, teleosts larvae and eggs, appendicularians |
>2000 | Siphonophores, salps, chaetognaths, pteropods, macrurid larvae, brachyurid larvae, teleost larvae, copepods |
Size class (µm) . | Main groups . |
---|---|
80–200 | Detritus, copepods (eggs, nauplii and copepodits), phytoplankton (diatoms, dinobionts), larvae (gastropods, annelids) |
200–300 | Copepods, cladocerans, larvae (gastropods, bivalves, annelids, brachyurids) |
300–500 | Copepods, cladocerans, larvae (gastropods, bivalves, annelids, brachyurids), appendicularians |
500–1000 | Copepods, teleost eggs, larvae (brachyurids, macrurids, bivalves), euphausiids, appendicularians, chaetognaths, pteropods |
1000–2000 | Copepods, pteropods, chaetognaths, siphonophores, macrurid larvae, brachyurid larvae, euphausiids, amphipods, teleosts larvae and eggs, appendicularians |
>2000 | Siphonophores, salps, chaetognaths, pteropods, macrurid larvae, brachyurid larvae, teleost larvae, copepods |
Dominant groups are indicated in bold characters by decreasing order of importance.
Size class (µm) . | Main groups . |
---|---|
80–200 | Detritus, copepods (eggs, nauplii and copepodits), phytoplankton (diatoms, dinobionts), larvae (gastropods, annelids) |
200–300 | Copepods, cladocerans, larvae (gastropods, bivalves, annelids, brachyurids) |
300–500 | Copepods, cladocerans, larvae (gastropods, bivalves, annelids, brachyurids), appendicularians |
500–1000 | Copepods, teleost eggs, larvae (brachyurids, macrurids, bivalves), euphausiids, appendicularians, chaetognaths, pteropods |
1000–2000 | Copepods, pteropods, chaetognaths, siphonophores, macrurid larvae, brachyurid larvae, euphausiids, amphipods, teleosts larvae and eggs, appendicularians |
>2000 | Siphonophores, salps, chaetognaths, pteropods, macrurid larvae, brachyurid larvae, teleost larvae, copepods |
Size class (µm) . | Main groups . |
---|---|
80–200 | Detritus, copepods (eggs, nauplii and copepodits), phytoplankton (diatoms, dinobionts), larvae (gastropods, annelids) |
200–300 | Copepods, cladocerans, larvae (gastropods, bivalves, annelids, brachyurids) |
300–500 | Copepods, cladocerans, larvae (gastropods, bivalves, annelids, brachyurids), appendicularians |
500–1000 | Copepods, teleost eggs, larvae (brachyurids, macrurids, bivalves), euphausiids, appendicularians, chaetognaths, pteropods |
1000–2000 | Copepods, pteropods, chaetognaths, siphonophores, macrurid larvae, brachyurid larvae, euphausiids, amphipods, teleosts larvae and eggs, appendicularians |
>2000 | Siphonophores, salps, chaetognaths, pteropods, macrurid larvae, brachyurid larvae, teleost larvae, copepods |
Dominant groups are indicated in bold characters by decreasing order of importance.
Low seasonal differences in zooplankton composition were recorded in the three smaller size classes. In the 80–200-µm size class, detritus was more important in winter, whereas between 200 and 500 µm copepods largely dominated the zooplankton in all seasons. Copepods dominated the 500–1000-µm size class, except in summer and autumn when invertebrate larvae were more abundant. In the two largest size classes, crustacean larvae and gelatinous organisms were dominant. Siphonophores and teleost eggs were more abundant in winter, whereas invertebrate and teleost larvae along with chaetognaths and pteropods were relatively more numerous in summer and autumn.
Effect of season and size on zooplankton stable isotope ratios
Significant differences in isotopic signatures occurred between zooplankton size classes with generally lower values in smaller size classes (from 80 to 500 µm) than in the larger ones (>1000 µm), whatever the season (Tables II and III). However, the increase in isotopic ratios was not regular since the highest mean δ15N value was recorded in the 1000–2000-µm size class (4.08 ± 0.68‰). The 1000–2000 and >2000-µm size classes were highly heterogeneous in terms of taxonomic composition (Table I). They were composed of groups of organisms which exhibited different stable isotope ratios. Salps and pteropods had the lowest δ15N (<2‰), whereas larvae of stomatopods, brachyurans, teleosts and cephalopods exhibited the highest δ15N (>4‰) (Table IV). Euphausiids, brachyuran and macruran larvae and chaetognaths had higher δ15N in the >2000 µm compared with the 1000–2000-µm size class, but this difference was not significant due to high standard deviations (t-tests, P > 0.05). Nevertheless, a decrease in δ15N values was observed in the largest size class (>2000 µm) as a whole, because it was dominated by organisms with low nitrogen isotopic signatures (Table IV).
. | Spring . | Summer . | Autumn . | Winter . | Mean . |
---|---|---|---|---|---|
δ13C (‰) | |||||
80–200 µm | −24.79 ± 0.13 | −22.73 ± 0.12 | −19.30 ± 0.17 | −23.13 ± 0.03 | −22.49 ± 2.30 |
200–300 µm | −25.58 ± 0.33 | −22.93 ± 0.03 | −20.30 ± 0.20 | −23.52 ± 0.02 | −23.08 ± 2.18 |
300–500 µm | −26.00 ± 0.13 | −22.70 ± 0.04 | −20.18 ± 0.10 | −23.12 ± 0.37 | −23.00 ± 2.38 |
500–1000 µm | −25.24 ± 0.25 | −23.77 ± 0.11 | −20.20 ± 0.48 | −22.98 ± 0.40 | −23.05 ± 2.12 |
1000–2000 µm | −23.04 ± 0.38 | −22.44 ± 0.01 | −19.00 ± 0.31 | −22.26 ± 0.29 | −21.69 ± 1.82 |
>2000 µm | −22.59 ± 0.31 | −21.77 ± 0.28 | −18.70 ± 0.14 | −21.70 ± 0.40 | −21.19 ± 1.71 |
Mean | −24.54 ± 0.99 | −22.72 ± 1.42 | −19.61 ± 1.42 | −22.79 ± 1.57 | −22.42 ± 2.00 |
δ15N (‰) | |||||
80–200 µm | 1.38 ± 0.07 | 3.94 ± 0.04 | 2.08 ± 0.47 | 2.27 ± 0.04 | 2.42 ± 1.08 |
200–300 µm | 1.29 ± 0.13 | 4.22 ± 0.10 | 2.29 ± 0.10 | 2.08 ± 0.04 | 2.47 ± 1.24 |
300–500 µm | 1.08 ± 0.16 | 4.21 ± 0.16 | 2.32 ± 0.06 | 2.45 ± 0.43 | 2.52 ± 1.29 |
500–1000 µm | 1.78 ± 0.08 | 4.51 ± 0.08 | 3.44 ± 0.53 | 2.82 ± 0.03 | 3.14 ± 1.14 |
1000–2000 µm | 3.36 ± 0.41 | 4.83 ± 0.23 | 3.68 ± 0.35 | 4.46 ± 0.77 | 4.08 ± 0.68 |
>2000 µm | 2.74 ± 0.34 | 4.20 ± 0.26 | 2.56 ± 0.32 | 2.08 ± 0.44 | 2.90 ± 0.91 |
Mean | 1.94 ± 1.03 | 4.32 ± 0.84 | 2.73 ± 0.88 | 2.69 ± 0.90 | 2.92 ± 1.12 |
. | Spring . | Summer . | Autumn . | Winter . | Mean . |
---|---|---|---|---|---|
δ13C (‰) | |||||
80–200 µm | −24.79 ± 0.13 | −22.73 ± 0.12 | −19.30 ± 0.17 | −23.13 ± 0.03 | −22.49 ± 2.30 |
200–300 µm | −25.58 ± 0.33 | −22.93 ± 0.03 | −20.30 ± 0.20 | −23.52 ± 0.02 | −23.08 ± 2.18 |
300–500 µm | −26.00 ± 0.13 | −22.70 ± 0.04 | −20.18 ± 0.10 | −23.12 ± 0.37 | −23.00 ± 2.38 |
500–1000 µm | −25.24 ± 0.25 | −23.77 ± 0.11 | −20.20 ± 0.48 | −22.98 ± 0.40 | −23.05 ± 2.12 |
1000–2000 µm | −23.04 ± 0.38 | −22.44 ± 0.01 | −19.00 ± 0.31 | −22.26 ± 0.29 | −21.69 ± 1.82 |
>2000 µm | −22.59 ± 0.31 | −21.77 ± 0.28 | −18.70 ± 0.14 | −21.70 ± 0.40 | −21.19 ± 1.71 |
Mean | −24.54 ± 0.99 | −22.72 ± 1.42 | −19.61 ± 1.42 | −22.79 ± 1.57 | −22.42 ± 2.00 |
δ15N (‰) | |||||
80–200 µm | 1.38 ± 0.07 | 3.94 ± 0.04 | 2.08 ± 0.47 | 2.27 ± 0.04 | 2.42 ± 1.08 |
200–300 µm | 1.29 ± 0.13 | 4.22 ± 0.10 | 2.29 ± 0.10 | 2.08 ± 0.04 | 2.47 ± 1.24 |
300–500 µm | 1.08 ± 0.16 | 4.21 ± 0.16 | 2.32 ± 0.06 | 2.45 ± 0.43 | 2.52 ± 1.29 |
500–1000 µm | 1.78 ± 0.08 | 4.51 ± 0.08 | 3.44 ± 0.53 | 2.82 ± 0.03 | 3.14 ± 1.14 |
1000–2000 µm | 3.36 ± 0.41 | 4.83 ± 0.23 | 3.68 ± 0.35 | 4.46 ± 0.77 | 4.08 ± 0.68 |
>2000 µm | 2.74 ± 0.34 | 4.20 ± 0.26 | 2.56 ± 0.32 | 2.08 ± 0.44 | 2.90 ± 0.91 |
Mean | 1.94 ± 1.03 | 4.32 ± 0.84 | 2.73 ± 0.88 | 2.69 ± 0.90 | 2.92 ± 1.12 |
Mean values were calculated using three replicates per size class and season.
. | Spring . | Summer . | Autumn . | Winter . | Mean . |
---|---|---|---|---|---|
δ13C (‰) | |||||
80–200 µm | −24.79 ± 0.13 | −22.73 ± 0.12 | −19.30 ± 0.17 | −23.13 ± 0.03 | −22.49 ± 2.30 |
200–300 µm | −25.58 ± 0.33 | −22.93 ± 0.03 | −20.30 ± 0.20 | −23.52 ± 0.02 | −23.08 ± 2.18 |
300–500 µm | −26.00 ± 0.13 | −22.70 ± 0.04 | −20.18 ± 0.10 | −23.12 ± 0.37 | −23.00 ± 2.38 |
500–1000 µm | −25.24 ± 0.25 | −23.77 ± 0.11 | −20.20 ± 0.48 | −22.98 ± 0.40 | −23.05 ± 2.12 |
1000–2000 µm | −23.04 ± 0.38 | −22.44 ± 0.01 | −19.00 ± 0.31 | −22.26 ± 0.29 | −21.69 ± 1.82 |
>2000 µm | −22.59 ± 0.31 | −21.77 ± 0.28 | −18.70 ± 0.14 | −21.70 ± 0.40 | −21.19 ± 1.71 |
Mean | −24.54 ± 0.99 | −22.72 ± 1.42 | −19.61 ± 1.42 | −22.79 ± 1.57 | −22.42 ± 2.00 |
δ15N (‰) | |||||
80–200 µm | 1.38 ± 0.07 | 3.94 ± 0.04 | 2.08 ± 0.47 | 2.27 ± 0.04 | 2.42 ± 1.08 |
200–300 µm | 1.29 ± 0.13 | 4.22 ± 0.10 | 2.29 ± 0.10 | 2.08 ± 0.04 | 2.47 ± 1.24 |
300–500 µm | 1.08 ± 0.16 | 4.21 ± 0.16 | 2.32 ± 0.06 | 2.45 ± 0.43 | 2.52 ± 1.29 |
500–1000 µm | 1.78 ± 0.08 | 4.51 ± 0.08 | 3.44 ± 0.53 | 2.82 ± 0.03 | 3.14 ± 1.14 |
1000–2000 µm | 3.36 ± 0.41 | 4.83 ± 0.23 | 3.68 ± 0.35 | 4.46 ± 0.77 | 4.08 ± 0.68 |
>2000 µm | 2.74 ± 0.34 | 4.20 ± 0.26 | 2.56 ± 0.32 | 2.08 ± 0.44 | 2.90 ± 0.91 |
Mean | 1.94 ± 1.03 | 4.32 ± 0.84 | 2.73 ± 0.88 | 2.69 ± 0.90 | 2.92 ± 1.12 |
. | Spring . | Summer . | Autumn . | Winter . | Mean . |
---|---|---|---|---|---|
δ13C (‰) | |||||
80–200 µm | −24.79 ± 0.13 | −22.73 ± 0.12 | −19.30 ± 0.17 | −23.13 ± 0.03 | −22.49 ± 2.30 |
200–300 µm | −25.58 ± 0.33 | −22.93 ± 0.03 | −20.30 ± 0.20 | −23.52 ± 0.02 | −23.08 ± 2.18 |
300–500 µm | −26.00 ± 0.13 | −22.70 ± 0.04 | −20.18 ± 0.10 | −23.12 ± 0.37 | −23.00 ± 2.38 |
500–1000 µm | −25.24 ± 0.25 | −23.77 ± 0.11 | −20.20 ± 0.48 | −22.98 ± 0.40 | −23.05 ± 2.12 |
1000–2000 µm | −23.04 ± 0.38 | −22.44 ± 0.01 | −19.00 ± 0.31 | −22.26 ± 0.29 | −21.69 ± 1.82 |
>2000 µm | −22.59 ± 0.31 | −21.77 ± 0.28 | −18.70 ± 0.14 | −21.70 ± 0.40 | −21.19 ± 1.71 |
Mean | −24.54 ± 0.99 | −22.72 ± 1.42 | −19.61 ± 1.42 | −22.79 ± 1.57 | −22.42 ± 2.00 |
δ15N (‰) | |||||
80–200 µm | 1.38 ± 0.07 | 3.94 ± 0.04 | 2.08 ± 0.47 | 2.27 ± 0.04 | 2.42 ± 1.08 |
200–300 µm | 1.29 ± 0.13 | 4.22 ± 0.10 | 2.29 ± 0.10 | 2.08 ± 0.04 | 2.47 ± 1.24 |
300–500 µm | 1.08 ± 0.16 | 4.21 ± 0.16 | 2.32 ± 0.06 | 2.45 ± 0.43 | 2.52 ± 1.29 |
500–1000 µm | 1.78 ± 0.08 | 4.51 ± 0.08 | 3.44 ± 0.53 | 2.82 ± 0.03 | 3.14 ± 1.14 |
1000–2000 µm | 3.36 ± 0.41 | 4.83 ± 0.23 | 3.68 ± 0.35 | 4.46 ± 0.77 | 4.08 ± 0.68 |
>2000 µm | 2.74 ± 0.34 | 4.20 ± 0.26 | 2.56 ± 0.32 | 2.08 ± 0.44 | 2.90 ± 0.91 |
Mean | 1.94 ± 1.03 | 4.32 ± 0.84 | 2.73 ± 0.88 | 2.69 ± 0.90 | 2.92 ± 1.12 |
Mean values were calculated using three replicates per size class and season.
. | F . | P-value . | Post hoc . |
---|---|---|---|
δ13C | |||
Season | 58.79 | <0.001 | Autumn > Summer = Winter>Spring |
Size class | 9.52 | <0.001 | 2000 = 1000>500 = 80 = 300 = 200 |
Season × Size class | 0.64 | 0.840 | n.s. |
δ15N | |||
Season | 13.23 | <0.001 | Summer>Autumn = Spring = Winter |
Size class | 4.18 | 0.001 | 1000 > 500 = 2000 = 300 = 200 = 80 |
Season × Size class | 0.56 | 0.732 | n.s. |
. | F . | P-value . | Post hoc . |
---|---|---|---|
δ13C | |||
Season | 58.79 | <0.001 | Autumn > Summer = Winter>Spring |
Size class | 9.52 | <0.001 | 2000 = 1000>500 = 80 = 300 = 200 |
Season × Size class | 0.64 | 0.840 | n.s. |
δ15N | |||
Season | 13.23 | <0.001 | Summer>Autumn = Spring = Winter |
Size class | 4.18 | 0.001 | 1000 > 500 = 2000 = 300 = 200 = 80 |
Season × Size class | 0.56 | 0.732 | n.s. |
F, statistics; P, level of significance; Post hoc, results of NKS post hoc tests of comparison of means. n.s., not significant.
The lower limit only of each size class was indicated for clarity. 80 = 80–200 µm, 200 = 200–300 µm, 300 = 300–500 µm, 500 = 500–1000 µm, 1000 = 1000–2000 µm, 2000 ≥ 2000 µm
. | F . | P-value . | Post hoc . |
---|---|---|---|
δ13C | |||
Season | 58.79 | <0.001 | Autumn > Summer = Winter>Spring |
Size class | 9.52 | <0.001 | 2000 = 1000>500 = 80 = 300 = 200 |
Season × Size class | 0.64 | 0.840 | n.s. |
δ15N | |||
Season | 13.23 | <0.001 | Summer>Autumn = Spring = Winter |
Size class | 4.18 | 0.001 | 1000 > 500 = 2000 = 300 = 200 = 80 |
Season × Size class | 0.56 | 0.732 | n.s. |
. | F . | P-value . | Post hoc . |
---|---|---|---|
δ13C | |||
Season | 58.79 | <0.001 | Autumn > Summer = Winter>Spring |
Size class | 9.52 | <0.001 | 2000 = 1000>500 = 80 = 300 = 200 |
Season × Size class | 0.64 | 0.840 | n.s. |
δ15N | |||
Season | 13.23 | <0.001 | Summer>Autumn = Spring = Winter |
Size class | 4.18 | 0.001 | 1000 > 500 = 2000 = 300 = 200 = 80 |
Season × Size class | 0.56 | 0.732 | n.s. |
F, statistics; P, level of significance; Post hoc, results of NKS post hoc tests of comparison of means. n.s., not significant.
The lower limit only of each size class was indicated for clarity. 80 = 80–200 µm, 200 = 200–300 µm, 300 = 300–500 µm, 500 = 500–1000 µm, 1000 = 1000–2000 µm, 2000 ≥ 2000 µm
. | n . | δ13C (‰) . | δ15N (‰) . |
---|---|---|---|
Salps | 11 | −22.76 ± 1.29 | 0.35 ± 1.46 |
Pteropods | 13 | −18.82 ± 1.83 | 1.54 ± 0.47 |
Ctenophores | 4 | −22.36 ± 0.39 | 2.67 ± 1.55 |
Siphonophores | 12 | −20.42 ± 0.70 | 3.23 ± 0.53 |
Hyperiid amphipods | 5 | −18.01 ± 1.43 | 3.43 ± 0.45 |
Macruran larvae | 7 | −20.79 ± 0.52 | 3.60 ± 0.60 |
Euphausiids | 9 | −22.68 ± 1.22 | 3.63 ± 1.03 |
Copepods | 6 | −21.45 ± 0.58 | 3.82 ± 0.07 |
Chaetognaths | 15 | −20.78 ± 0.72 | 3.97 ± 0.82 |
Brachyuran larvae | 9 | −22.04 ± 1.59 | 4.07 ± 1.07 |
Stomatopod larvae | 1 | −19.61 ± 0.00 | 4.91 ± 0.00 |
Teleost larvae | 8 | −20.35 ± 0.91 | 4.97 ± 0.65 |
Cephalopod larvae | 1 | −18.83 ± 0.00 | 9.53 ± 0.00 |
. | n . | δ13C (‰) . | δ15N (‰) . |
---|---|---|---|
Salps | 11 | −22.76 ± 1.29 | 0.35 ± 1.46 |
Pteropods | 13 | −18.82 ± 1.83 | 1.54 ± 0.47 |
Ctenophores | 4 | −22.36 ± 0.39 | 2.67 ± 1.55 |
Siphonophores | 12 | −20.42 ± 0.70 | 3.23 ± 0.53 |
Hyperiid amphipods | 5 | −18.01 ± 1.43 | 3.43 ± 0.45 |
Macruran larvae | 7 | −20.79 ± 0.52 | 3.60 ± 0.60 |
Euphausiids | 9 | −22.68 ± 1.22 | 3.63 ± 1.03 |
Copepods | 6 | −21.45 ± 0.58 | 3.82 ± 0.07 |
Chaetognaths | 15 | −20.78 ± 0.72 | 3.97 ± 0.82 |
Brachyuran larvae | 9 | −22.04 ± 1.59 | 4.07 ± 1.07 |
Stomatopod larvae | 1 | −19.61 ± 0.00 | 4.91 ± 0.00 |
Teleost larvae | 8 | −20.35 ± 0.91 | 4.97 ± 0.65 |
Cephalopod larvae | 1 | −18.83 ± 0.00 | 9.53 ± 0.00 |
N, number of samples analyzed.
. | n . | δ13C (‰) . | δ15N (‰) . |
---|---|---|---|
Salps | 11 | −22.76 ± 1.29 | 0.35 ± 1.46 |
Pteropods | 13 | −18.82 ± 1.83 | 1.54 ± 0.47 |
Ctenophores | 4 | −22.36 ± 0.39 | 2.67 ± 1.55 |
Siphonophores | 12 | −20.42 ± 0.70 | 3.23 ± 0.53 |
Hyperiid amphipods | 5 | −18.01 ± 1.43 | 3.43 ± 0.45 |
Macruran larvae | 7 | −20.79 ± 0.52 | 3.60 ± 0.60 |
Euphausiids | 9 | −22.68 ± 1.22 | 3.63 ± 1.03 |
Copepods | 6 | −21.45 ± 0.58 | 3.82 ± 0.07 |
Chaetognaths | 15 | −20.78 ± 0.72 | 3.97 ± 0.82 |
Brachyuran larvae | 9 | −22.04 ± 1.59 | 4.07 ± 1.07 |
Stomatopod larvae | 1 | −19.61 ± 0.00 | 4.91 ± 0.00 |
Teleost larvae | 8 | −20.35 ± 0.91 | 4.97 ± 0.65 |
Cephalopod larvae | 1 | −18.83 ± 0.00 | 9.53 ± 0.00 |
. | n . | δ13C (‰) . | δ15N (‰) . |
---|---|---|---|
Salps | 11 | −22.76 ± 1.29 | 0.35 ± 1.46 |
Pteropods | 13 | −18.82 ± 1.83 | 1.54 ± 0.47 |
Ctenophores | 4 | −22.36 ± 0.39 | 2.67 ± 1.55 |
Siphonophores | 12 | −20.42 ± 0.70 | 3.23 ± 0.53 |
Hyperiid amphipods | 5 | −18.01 ± 1.43 | 3.43 ± 0.45 |
Macruran larvae | 7 | −20.79 ± 0.52 | 3.60 ± 0.60 |
Euphausiids | 9 | −22.68 ± 1.22 | 3.63 ± 1.03 |
Copepods | 6 | −21.45 ± 0.58 | 3.82 ± 0.07 |
Chaetognaths | 15 | −20.78 ± 0.72 | 3.97 ± 0.82 |
Brachyuran larvae | 9 | −22.04 ± 1.59 | 4.07 ± 1.07 |
Stomatopod larvae | 1 | −19.61 ± 0.00 | 4.91 ± 0.00 |
Teleost larvae | 8 | −20.35 ± 0.91 | 4.97 ± 0.65 |
Cephalopod larvae | 1 | −18.83 ± 0.00 | 9.53 ± 0.00 |
N, number of samples analyzed.
C and N stable isotope ratios of zooplankton significantly differed with season and size, and the absence of interaction between factors indicated that the pattern of difference with size was similar in each season (Table III). Mean δ13C values were significantly higher in autumn and lower in spring, whereas mean δ15N values were significantly higher in summer and lower in winter (Table II). As a general pattern, zooplankton isotopic signatures were characterized by higher values of both δ13C and δ15N during the warm season (summer–autumn) compared with the cold season (winter–spring). In all size classes, the highest δ13C was recorded in autumn and the highest δ15N in summer, whereas the lowest values of both elements were recorded in spring, except for δ15N in the >2000 µm size class (winter) (Table II).
Time (season) represented the main factor of isotopic variability in zooplankton (Table III). Differences in stable isotope ratios were higher between seasons (Δδ13C ± 4.93 + 1.21‰, Δδ15N = 2.38 + 0.94‰) than between size classes (Δδ13C = 1.89 + 1.95‰, Δδ15N = 1.66 + 0.88‰).
POM composition and seasonal variation of isotopic ratios
There were strong differences between quantitative and qualitative composition of water POM at the SOMLIT site, Cortiou plume and Huveaune River. A much greater quantity of small particles (2–6 µm) was recorded in the Cortiou plume (212 × 103cells µm L–1) and Huveaune River (117 × 103 cells µm L–1) than in SOMLIT POM (21 cells µm L−1). These small particles were mainly detritus and bacteria, as was observed on STEM photos. They represented 100% of cells found in Huveaune River water, 94% in Cortiou plume and only 25% in SOMLIT water. Excepting these detrital particles, Cortiou plume contained more pico- and nanoplankton cells than SOMLIT water (12 × 103 and 15 cells µm L−1, respectively). However, in spite of such a difference in cell abundance, the composition of the pico- and nanoplankton community was similar at the two sites, with a dominance of the cyanobacteria Prochlorococcus and Synechococcus, and two groups of non-identified picoeukaryotes (picoeukaryotes 1 and 2) (Fig. 2). Thus, while the Huveaune River water was entirely composed of detritus and bacterial particles, the Cortiou plume and SOMLIT POM were composed of both detritus and living phytoplankton cells, but in different proportions. Detritus dominated the sewage plume, whereas living pico- and nanoplankton cells dominated in SOMLIT water.
Stable isotope ratios also differed between these three types of POM. Both δ13C (ANOVA, F = 7.60, P = 0.018) and δ15N (ANOVA, F = 17.75, P = 0.002) differed significantly between POM of the SOMLIT site, Cortiou sewage plume and Huveaune River. Huveaune POM was characterized by low δ13C (−26.25 ± 0.51‰) and high δ15N (4.48 ± 0.41‰), Cortiou POM by low values of both δ13C (−25.50 ± 0.62‰) and δ15N (−0.59 ± 0.82‰) and SOMLIT POM by high δ13C (−23.59 ± 1.37‰) and intermediate δ15N (2.18 ± 1.39‰). Stable isotope ratios of the three POM types and zooplankton at SOMLIT showed similar seasonal variations. They were generally lower during the cold period (winter–spring) than during the warm season (summer–autumn). These seasonal variations of zooplankton at SOMLIT clearly reflected those of POM sources at this site (Fig. 3), indicating the POM integration in zooplankton food webs. At the SOMLIT site, the δ15N difference between POM and zooplankton was higher in winter and spring (mean Δδ15N ± 1.29 + 0.67‰) than in summer and autumn (mean Δδ15N = 0.21 + 0.77‰), suggesting higher fractionation factors along the planktonic food web during the cold season.
Seasonal variations of environmental parameters and microphytoplankton composition
The environmental parameters measured at the SOMLIT site showed high seasonal variations (Table V), but two main periods could be highlighted. The cold season (winter–spring) was characterized by temperatures ∼13°C and high SPM, nutrient (NO3 and NO2) and Chl a content. In contrast, the warm season (summer–autumn) had temperatures ∼21–22°C and low SPM, nutrient and Chl a concentrations. Precipitation was highest during winter and lowest in summer.
Parameters . | Spring . | Summer . | Autumn . | Winter . | Data sources . |
---|---|---|---|---|---|
SPM mg L−1 | 0.53 ± 0.0 | 0.40 ± 0.1 | 0.50 ± 0.1 | 6.5 ± 0.6 | SOMLIT national 2009–2010; this study |
Chl aμg L−1 | 0.57 ± 0.1 | 0.24 ± 0.1 | 0.30 ± 0.2 | 0.50 ± 0.0 | SOMLIT national 2009–2010 |
NO3μmol L−1 | 1.26 ± 1.0 | 0.43 ± 0.4 | 0.27 ± 0.2 | 1.54 ± 0.2 | SOMLIT national 2009–2010 |
NO2μmol L−1 | 0.12 ± 0.1 | 0.13 ± 0.1 | 0.06 ± 0.1 | 0.20 ± 0.0 | SOMLIT national 2009–2010 |
T°C surface | 13.3 ± 0.1 | 22.0 ± 3.1 | 20.9 ± 1.3 | 12.5 ± 0.1 | SOMLIT national 2009–2010 |
Salinity | 38.09 ± 0.0 | 37.92 ± 0.1 | 38.13 ± 0.0 | 38.02 ± 0.0 | SOMLIT national 2009–2010 |
Rain mm | 51 ± 21 | 1 ± 0.1 | 83.5 ± 0.5 | 121 ± 34 | www.meteofrance.fr (2009–2010 Martigues-Toulon) |
Parameters . | Spring . | Summer . | Autumn . | Winter . | Data sources . |
---|---|---|---|---|---|
SPM mg L−1 | 0.53 ± 0.0 | 0.40 ± 0.1 | 0.50 ± 0.1 | 6.5 ± 0.6 | SOMLIT national 2009–2010; this study |
Chl aμg L−1 | 0.57 ± 0.1 | 0.24 ± 0.1 | 0.30 ± 0.2 | 0.50 ± 0.0 | SOMLIT national 2009–2010 |
NO3μmol L−1 | 1.26 ± 1.0 | 0.43 ± 0.4 | 0.27 ± 0.2 | 1.54 ± 0.2 | SOMLIT national 2009–2010 |
NO2μmol L−1 | 0.12 ± 0.1 | 0.13 ± 0.1 | 0.06 ± 0.1 | 0.20 ± 0.0 | SOMLIT national 2009–2010 |
T°C surface | 13.3 ± 0.1 | 22.0 ± 3.1 | 20.9 ± 1.3 | 12.5 ± 0.1 | SOMLIT national 2009–2010 |
Salinity | 38.09 ± 0.0 | 37.92 ± 0.1 | 38.13 ± 0.0 | 38.02 ± 0.0 | SOMLIT national 2009–2010 |
Rain mm | 51 ± 21 | 1 ± 0.1 | 83.5 ± 0.5 | 121 ± 34 | www.meteofrance.fr (2009–2010 Martigues-Toulon) |
Mean values were calculated using eight samples per season. SPM, suspended particulate matter; Chl a, chlorophyll a; T, surface water temperature.
Parameters . | Spring . | Summer . | Autumn . | Winter . | Data sources . |
---|---|---|---|---|---|
SPM mg L−1 | 0.53 ± 0.0 | 0.40 ± 0.1 | 0.50 ± 0.1 | 6.5 ± 0.6 | SOMLIT national 2009–2010; this study |
Chl aμg L−1 | 0.57 ± 0.1 | 0.24 ± 0.1 | 0.30 ± 0.2 | 0.50 ± 0.0 | SOMLIT national 2009–2010 |
NO3μmol L−1 | 1.26 ± 1.0 | 0.43 ± 0.4 | 0.27 ± 0.2 | 1.54 ± 0.2 | SOMLIT national 2009–2010 |
NO2μmol L−1 | 0.12 ± 0.1 | 0.13 ± 0.1 | 0.06 ± 0.1 | 0.20 ± 0.0 | SOMLIT national 2009–2010 |
T°C surface | 13.3 ± 0.1 | 22.0 ± 3.1 | 20.9 ± 1.3 | 12.5 ± 0.1 | SOMLIT national 2009–2010 |
Salinity | 38.09 ± 0.0 | 37.92 ± 0.1 | 38.13 ± 0.0 | 38.02 ± 0.0 | SOMLIT national 2009–2010 |
Rain mm | 51 ± 21 | 1 ± 0.1 | 83.5 ± 0.5 | 121 ± 34 | www.meteofrance.fr (2009–2010 Martigues-Toulon) |
Parameters . | Spring . | Summer . | Autumn . | Winter . | Data sources . |
---|---|---|---|---|---|
SPM mg L−1 | 0.53 ± 0.0 | 0.40 ± 0.1 | 0.50 ± 0.1 | 6.5 ± 0.6 | SOMLIT national 2009–2010; this study |
Chl aμg L−1 | 0.57 ± 0.1 | 0.24 ± 0.1 | 0.30 ± 0.2 | 0.50 ± 0.0 | SOMLIT national 2009–2010 |
NO3μmol L−1 | 1.26 ± 1.0 | 0.43 ± 0.4 | 0.27 ± 0.2 | 1.54 ± 0.2 | SOMLIT national 2009–2010 |
NO2μmol L−1 | 0.12 ± 0.1 | 0.13 ± 0.1 | 0.06 ± 0.1 | 0.20 ± 0.0 | SOMLIT national 2009–2010 |
T°C surface | 13.3 ± 0.1 | 22.0 ± 3.1 | 20.9 ± 1.3 | 12.5 ± 0.1 | SOMLIT national 2009–2010 |
Salinity | 38.09 ± 0.0 | 37.92 ± 0.1 | 38.13 ± 0.0 | 38.02 ± 0.0 | SOMLIT national 2009–2010 |
Rain mm | 51 ± 21 | 1 ± 0.1 | 83.5 ± 0.5 | 121 ± 34 | www.meteofrance.fr (2009–2010 Martigues-Toulon) |
Mean values were calculated using eight samples per season. SPM, suspended particulate matter; Chl a, chlorophyll a; T, surface water temperature.
Seasonal variations of microphytoplankton composition at the SOMLIT site (Table VI) showed high abundance of large phytoplanktonic cells (20–200 µm) in summer and spring, and lower abundance in winter and particularly in autumn. Flagellates and crytophyceans dominated in spring and winter, whereas summer was the only season when diatoms were abundant. In autumn, the phytoplankton community was largely dominated by crytophyceans, along with some dinobionts.
Phytoplankton groups | Spring (%) | Summer (%) | Autumn (%) | Winter (%) | Main genera |
Diatoms | 0.2 | 40.2 | 0 | 0.9 | Chaetoceros |
Dinobionts (dinoflagellates) | 4.5 | 2.8 | 15.7 | 13.5 | Gymnodinium, Heterocapsa |
Dinobionts (flagellates) | 74.1 | 45 | 0 | 49.9 | |
Cryptophyceans | 21.2 | 2.8 | 84.3 | 35.5 | Cryptomonas |
Prasinophyceans | 0 | 4.1 | 0 | 0 | |
Monades | 0 | 5.1 | 0 | 0 | |
Other groups | 0 | 0 | 0 | 0.2 | Silicoflagellates |
Number of cells L−l | 147 863 | 231 600 | 2235 | 36 727 |
Phytoplankton groups | Spring (%) | Summer (%) | Autumn (%) | Winter (%) | Main genera |
Diatoms | 0.2 | 40.2 | 0 | 0.9 | Chaetoceros |
Dinobionts (dinoflagellates) | 4.5 | 2.8 | 15.7 | 13.5 | Gymnodinium, Heterocapsa |
Dinobionts (flagellates) | 74.1 | 45 | 0 | 49.9 | |
Cryptophyceans | 21.2 | 2.8 | 84.3 | 35.5 | Cryptomonas |
Prasinophyceans | 0 | 4.1 | 0 | 0 | |
Monades | 0 | 5.1 | 0 | 0 | |
Other groups | 0 | 0 | 0 | 0.2 | Silicoflagellates |
Number of cells L−l | 147 863 | 231 600 | 2235 | 36 727 |
Phytoplankton groups | Spring (%) | Summer (%) | Autumn (%) | Winter (%) | Main genera |
Diatoms | 0.2 | 40.2 | 0 | 0.9 | Chaetoceros |
Dinobionts (dinoflagellates) | 4.5 | 2.8 | 15.7 | 13.5 | Gymnodinium, Heterocapsa |
Dinobionts (flagellates) | 74.1 | 45 | 0 | 49.9 | |
Cryptophyceans | 21.2 | 2.8 | 84.3 | 35.5 | Cryptomonas |
Prasinophyceans | 0 | 4.1 | 0 | 0 | |
Monades | 0 | 5.1 | 0 | 0 | |
Other groups | 0 | 0 | 0 | 0.2 | Silicoflagellates |
Number of cells L−l | 147 863 | 231 600 | 2235 | 36 727 |
Phytoplankton groups | Spring (%) | Summer (%) | Autumn (%) | Winter (%) | Main genera |
Diatoms | 0.2 | 40.2 | 0 | 0.9 | Chaetoceros |
Dinobionts (dinoflagellates) | 4.5 | 2.8 | 15.7 | 13.5 | Gymnodinium, Heterocapsa |
Dinobionts (flagellates) | 74.1 | 45 | 0 | 49.9 | |
Cryptophyceans | 21.2 | 2.8 | 84.3 | 35.5 | Cryptomonas |
Prasinophyceans | 0 | 4.1 | 0 | 0 | |
Monades | 0 | 5.1 | 0 | 0 | |
Other groups | 0 | 0 | 0 | 0.2 | Silicoflagellates |
Number of cells L−l | 147 863 | 231 600 | 2235 | 36 727 |
DISCUSSION
The results of the present study showed seasonal variations in stable isotope ratios of zooplankton size groups (80 to >2000 µm) according to their composition and among taxonomic groups. The original approach of this paper resides in the relationships between these variations and those of POM sources, environmental parameter fluctuations (temperature, nutrients, Chl a concentration) and phytoplankton and seawater POM composition in Marseille Bay.
Influence of composition and size on zooplankton stable isotope ratios
Zooplankton composition at the SOMLIT site differed among size classes. Detritus and small copepod stages dominated the smallest size class (80–200 µm). The dominance of detritus in this size class could be related to the proximity of the large city of Marseille, the constant inputs of Cortiou sewage wastewaters and sporadic river water runoff. Rolff (Rolff, 2000) showed that phytoflagellates and ciliates dominated in the 50–100-µm size class, and ciliates, copepod nauplii, rotifers and diatoms in the 100–200 µm size class. The three following larger size classes (200–300, 300–500, 500–1000 μm) were dominated by copepods at SOMLIT and by cladocerans to a lesser extent, as already observed in different regions (Champalbert, 1996; Rolff, 2000; Calbet et al., 2001; Saiz et al., 2007; Koppelmann et al., 2009). Copepods usually account for 45–95% of these size classes in the Mediterranean Sea (Razouls and Kouwenberg, 1993; Champalbert, 1996). The 1000–2000 and >2000-µm size classes were very diverse and composed, along with large copepods, of pteropods, gelatinous organisms (appendicularians, salps, siphonophores, chaetognaths), euphausiids and large crustacean, cephalopod and teleost larvae.
Both δ13C and δ15N of zooplankton generally increased with size at the SOMLIT site, though not following a constant pattern. The lowest δ13C was observed in the 200–300-µm size class, while the abundance of terrestrial detritus in the 80–200-µm size class should have resulted in lower δ13C values, as observed in front of the Rhone River (Harmelin-Vivien et al., 2008). However, this might be due to a high proportion of 13C-enriched exuviae and feces as indicated by Checkley and Entzeroth (Checkley and Entzeroth, 1985) and Klein Breteler et al. (Klein Breteler et al., 2002). In addition, δ15N of the largest size class (>2000 µm) was lower than that of the 1000–2000 and 500–1000-µm size classes. In the literature, an increase in zooplankton δ15N with size is generally recorded (e.g. Rau et al., 1990; Malej et al., 1993; Waite et al., 2007; Koppelmann et al., 2009; among others). Enrichment in zooplankton δ15N with size is usually explained as reflecting size-related consumption patterns in marine plankton food webs (Rolff, 2000). Several authors report a large degree of trophic level overlap among zooplankton size classes (Fry and Quiñones, 1994; Koppelmann et al., 2009). A large range of δ15N was recorded among the different groups present in the two largest size classes at SOMLIT, which was related to feeding differences. Microplankton filter feeders (salps, pteropods) had much lower δ15N than predators (chaetognaths, large crustacean larvae, teleost and cephalopod larvae). Wide differences of isotopic signatures, particularly δ15N, among zooplankton groups or even species have been recorded in many studies, and are related to differences in diet (Schell et al., 1998; Hobson et al., 2002; Strzelecki et al., 2007; Waite et al., 2007; Fanelli et al., 2011). The lowest isotope ratios are found in filter feeders (numerous copepods, ostracods, pteropods, salps and thaliaceans) which feed on POM and small planktonic cells, and the highest δ15N values were recorded in carnivores (chaetognaths, large crustaceans, micronekton) (Fanelli et al., 2011). Costalago et al. (Costalago et al., 2012) analyzed the stable isotope ratios of some zooplankton groups in the Gulf of Lions, close from Marseille Bay. The comparison of their results from the summer season with our data shows similar δ15N and higher δ13C values for microplankton, copepods and cladocerans (dominant in 200–500-µm size classes).
Effect of composition and environmental parameters on POM isotopic signatures
Although located a few miles offshore, the SOMLIT site can be influenced by material of terrestrial and anthropogenic origin washed into the sea by the Cortiou sewage outfall and the Huveaune River (Fig. 1), as shown by hydrodynamic studies performed in the Bay of Marseille (Pradal and Millet, 2006). This is particularly the case after heavy rains and with strong winds blowing from the south-east. Detritus, along with some bacteria, were the only components of the Huveaune River POM when it was flowing to the sea. POM collected in the Cortiou sewage plume was a mixture of wastewater POM and marine phytoplankton, but was largely dominated by anthropogenic detrital material (>75%). In contrast, SOMLIT water POM was mainly composed of phytoplankton, but also included a considerable proportion of small detritus particles. Cresson et al. (Cresson et al., 2012) demonstrated with mixing models that POM sampled further inshore in the Bay of Marseille is composed of POM of various origins, including Cortiou sewage and Huveaune River POM. However, coastal and offshore marine phytoplankton was dominant in seawater POM. These authors recorded higher POM δ15N in winter as their sites are more influenced by the Huveaune River than Cortiou sewage waters. The lower δ15N recorded in SOMLIT POM in winter indicated a strong influence of Cortiou sewage waters (−1.17 + 0.3 ‰) during winter sampling that occurred just after a heavy rainy event. Higher quantities of POM with lower δ15N from Cortiou mixed with marine POM and decreased the δ15N of the SOMLIT POM in winter.
Thus, both flow cytometry analysis and seasonal variations of POM isotopic signatures strongly suggest that the detrital particles observed in SOMLIT water POM originated from Cortiou sewage inputs.
The low δ13C values (<−25‰) recorded in Huveaune and Cortiou waters agree with their continental origin, as terrestrial plant detritus exhibits much lower δ13C than marine phytoplankton (Fry and Sherr, 1984; Riera and Richard, 1996; Harmelin-Vivien et al., 2008). Schell et al. (Schell et al., 1998) also found depleted isotopic values in coastal waters due to the Mackenzie River inputs of terrestrially derived carbon and nitrogenous nutrients with low 13C and 15N values. The low δ13C observed in SOMLIT POM in winter (−24.00‰) and spring (−25.16‰) confirmed the influence of terrestrial matter at this site during these seasons. While the high δ15N of the Huveaune River water (>4‰) agrees with those found in other rivers (Kendall et al., 2001; Harmelin-Vivien et al., 2010), the very low δ15N value recorded in the Cortiou plume (∼0‰) is surprising and could be related to wastewater treatment and denitrifying bacteria which may be responsible for these low δ15N ratios (Kendall et al., 2007). Blooms of diatoms and their diazotrophic cyanobacterial symbiont responsible of N2 fixation (diazotrophy) were noticed in some plume areas where organic matter containing recently fixed nitrogen had a δ15N = −1‰ (Yeung et al., 2012). However, the flow cytometry analysis did not confirmed the presence of these organisms in our study area. Previously, Darnaude et al. (Darnaude et al., 2004) found similar stable isotope ratios for marine phytoplankton in Marseille Bay (δ13C = −22.36 + 0.24‰; δ15N = 2.33 + 0.11‰). In the Gulf of Lions, in the western part of Marseille, Harmelin-Vivien et al. (Harmelin-Vivien et al., 2008) found higher stable isotope ratios for phytoplankton (δ13C = −20.08 + 0.78‰; δ15N = 4.45 + 0.75‰) that in SOMLIT POM. This proves that even these areas are rather close (∼200 km), the local baseline of a food web is very important.
Seasonal variations of stable isotope ratios of POM were similar at the three sites, with generally lower values during the cold season (winter–spring) and higher values during the warm season (summer–autumn), except for high δ15N in the Huveaune River in winter. Environmental parameters also differed between these two main climatic periods. Both seasonal variations of environmental parameters and POM composition could explain the observed seasonal differences in POM isotopic signatures. The cold season is characterized by high Chl a and nutrient levels (Table V). Temperature seems to influence stable isotope ratios and δ13C values decrease with decreasing temperatures according to Goericke and Fry (Goericke and Fry, 1994). However, in our case the mean δ13C values of zooplankton in winter and summer were not statistically different.
Higher quantities of 13C depleted terrestrial inputs and 15N depleted sewage wastewater in winter, due to rainy events, might explain in part the lower δ13C and δ15N observed. The abundance of small planktonic cells (flagellates and dinoflagellates) during the cold season also contributed to low stable isotope values, as dinobionts have lower stable isotope values than diatoms (Fry and Wainwright, 1991; Sommer and Sommer, 2004). In addition, Grégori et al. (Grégori et al., 2001) observed a massive bloom of picophytoplankton cells (0.2–2 µm) in spring in the Bay of Marseille, and small cells are markedly 13C depleted compared with larger microphytoplankton particles (Rau et al., 1990).
In contrast, the warm season was characterized by low nutrient and Chl a content. The growth of phytoplankton in an impoverished pool of nitrates, where only 15N-enriched compounds remain, may be responsible for the high δ15N values observed for POM in summer as shown by Savoye et al. (Savoye et al., 2003) and Montoya (Montoya, 2007). Fry and Wainwright (Fry and Wainwright, 1991) also report a 13C enrichment of phytoplankton at the end of a bloom. During summer, the bloom of large diatoms was probably responsible for nutrient consumption and limitation, resulting in an increase in stable isotope values. Moreover, diatoms are isotopically heavier than dinobionts (Sommer and Sommer, 2004; Waite et al., 2007). Thus, both composition of the phytoplankton community (diatoms) and nutrient limitation contributed to increase POM stable isotope ratios in summer along with high temperatures. As observed by Cresson et al. (Cresson et al., 2012), the fact that high concentrations of Chl a (winter–spring) did not match with the maximum abundance of large microphytoplankton, and particularly diatoms (summer), also reflects the predominance of small phytoplankton cells at the SOMLIT site, as at many other sites in the Mediterranean Sea (Arin et al., 2005).
Seasonal variation of isotopic signatures of zooplankton and trophic functioning
Mean isotopic signatures of zooplankton at SOMLIT showed the same seasonal tendency in δ13C and δ15N as seawater POM, with higher values in summer and autumn, and lower values in winter and spring. Seasonal variations of zooplankton isotopic signatures are observed by many authors with also generally higher δ13C in summer or autumn (Wainright and Fry, 1994; Bouillon et al., 2000; Kibirige et al., 2002; Fanelli et al., 2011; among others). Kibirige et al. (Kibirige et al., 2002) also recorded minimum values of stable isotopes in winter and maximum values in summer for both food sources and dominant zooplankton species.
Seasonal variations in isotopic signatures of zooplankton size classes reflect the seasonal variation of different parameters, (i) temperature and environmental parameters, (ii) composition and isotopic ratios of phytoplankton, (iii) zooplankton composition and (iv) most likely variation in zooplankton diet. Differences in stable isotope values among zooplankton size classes at the SOMLIT site were more dispersed during the cold than the warm season, particularly for δ15N. A Δδ15N of 2.4 and 2.3‰ was observed among zooplankton size classes in winter and spring respectively, and a difference of 0.9 and 1.6‰ in summer and autumn, respectively. In addition, differences between POM and zooplankton were higher during the cold than the warm season. These results suggest that food overlap between zooplankton size classes could be higher in summer–autumn than in winter–spring. A higher nitrogen fractionation was particularly evident in the 1000–2000-µm size class during the cold season, suggesting more specific predatory diets in winter, while lower fractionation in summer indicated wider omnivorous diets. This would imply that food sources available to zooplankton were more diversified and abundant during the cold season, while a food shortage occurred during the warm season with higher food overlap, omnivory and probably competition for resources. Such a hypothesis agrees with higher SPM and Chl a contents in the environment in winter and spring. Poulet et al. (Poulet et al., 1986) demonstrated that POM has distinct chemical characteristics and nutritional values varying with categories, size fractions and biogenic or terrestrial origin of the particles. Recently, Cresson et al. (Cresson et al., 2012) demonstrated that the biochemical and nutritional value of POM further inshore in the Bay of Marseille varies with season, with lowest values being recorded in autumn. Marine trophic pathways differ according to water characteristics and thus seasons (Azam et al., 1983). In the Mediterranean Sea, they range from the prevalence of microbial loop in oligotrophic environments to the classical food chain in meso- or eutrophic environments like neritic zones (Christaki et al., 1996; Saiz et al., 2007). At the SOMLIT site, the food web is probably an intermediate situation between classical food chain and microbial loop (Legendre and Rassoulzadegan, 1999). A strong coupling between the microbial food webs and upper trophic levels is mediated by ciliates and other microheterotrophs that are preyed on by small copepod species and juvenile stages (Calbet et al., 2001; Saiz et al., 2007). To validate or refute the hypothesis of a difference in zooplankton diet with season at the SOMLIT site, more accurate studies at specific and size level, including diet description, taxonomic and biochemical composition, would be necessary.
Thus, from the present study, it can be concluded that a complex zooplankton food web was present at the SOMLIT site in the Bay of Marseille, intermediate between microbial and herbivorous food webs. The small detritus and bacterial particles brought to the sea by the Cortiou sewage outfall and the Huveaune River were mixed with natural marine phytoplankton populations, dominated most of the year by pico- and nanoplankton cells, and consumed by zooplankton populations. It is important that this transfer of highly contaminated organic matter (Wafo et al., 2006; Guigue et al., 2011) be taken into account in modeling approaches and estimation of pelagic food web contamination.
ACKNOWLEDGEMENTS
Thanks are expressed to the crew of the R.V. Antedon II for help in sampling, to Béatrice Becker for microphytoplankton identification, to Chantal Bézac for STEM photos, to Patrick Raimbault in charge of the SOMLIT long-term survey in Marseille and to Michael Paul for improvement of the English. Stable isotope analysis was performed at the Scottish Crop Research Institute (UK). This study was part of the national programs POTOMAC (Rôle du plancton dans le transfert trophique des contaminants en Méditerranée nord occidentale. EC2CO-PNEC-2009) and COSTAS (Contaminants dans le système trophique: phytoplancton, zooplancton, anchois, sardine ANR- AA-PPPP-007).
REFERENCES
Author notes
Corresponding editor: Roger Harris