Environmental Science and Pollution Research

, Volume 24, Issue 5, pp 4973–4989 | Cite as

Spatio-temporal variability of fluorescent dissolved organic matter in the Rhône River delta and the Fos-Marseille marine area (NW Mediterranean Sea, France)

  • Nicolas Ferretto
  • Marc Tedetti
  • Catherine Guigue
  • Stéphane Mounier
  • Patrick Raimbault
  • Madeleine Goutx
Research Article

Abstract

The spatio-temporal variability of fluorescent dissolved organic matter (FDOM) and its relationships with physical (temperature, salinity) and chemical (nutrients, chlorophyll a, dissolved and particulate organic carbon, nitrogen and phosphorus) parameters were investigated in inland waters of the Rhône River delta and the Fos-Marseille marine area (northwestern Mediterranean, France). Samples were taken approximately twice per month in two inland sites and three marine sites from February 2011 to January 2012. FDOM was analysed using fluorescence excitation-emission matrices (EEMs) coupled with parallel factor analysis (PARAFAC). In inland waters, humic-like components C1 (λExEm: 250 (330)/394 nm) and C3 (λExEm: 250 (350)/454 nm) dominated over one tryptophan-like component C2 (λExEm: 230 (280)/340 nm), reflecting a background contribution of terrigenous material (~67% of total fluorescence intensity, in quinine sulphate unit (QSU)) throughout the year. In marine waters, protein-like material, with tyrosine-like C4 (λExEm: <220 (275)/<300 nm) and tryptophan-like C5 (λExEm: 230 (280)/342 nm), dominated (~71% of total fluorescence intensity, in QSU) over a single humic-like component C6 (λExEm: 245 (300)/450 nm). In inland waters of the Rhône River delta, humic-like components C1 and C3 were more abundant in autumn-winter, very likely due to inputs of terrestrial organic matter from rainfalls, runoffs and wind-induced sediment resuspension. In marine sites, intrusions of the Berre Lagoon and Rhône River waters had a significant impact on the local biogeochemistry, leading to higher fluorescence intensities of humic- and protein-like components in spring-summer. On average, the fluorescence intensities of FDOM components C4, C5 and C6 increased by 33–81% under lower salinity. This work highlights the complex dynamics of FDOM in coastal waters and confirms the link between marine FDOM and the Rhône River freshwater intrusions on larger spatial and temporal scales in the Fos-Marseille marine area.

Keywords

Dissolved organic matter Fluorescence PARAFAC Rhône River Vaccarès pond Bay of Marseille Mediterranean Sea 

Introduction

Marine dissolved organic matter (DOM, <0.7 μm) is one of the greatest reactive carbon pools on Earth (~662 Pg C) (Hansell et al. 2009), comparable to the carbon reservoirs of atmospheric CO2 (~800 Pg C) (Houghton 2007; Dittmar and Stubbins 2014) and of terrestrial biomass (~550 Pg C) and approximately 200 times greater than the carbon reservoir of marine biomass (~3 Pg C) (Siegenthaler and Sarmiento 1993). In the ocean, DOM accounts for more than 95% of the total organic carbon pool (Benner 2002) and is essentially produced by autochthonous biological activity, including extracellular release by phytoplankton, zooplankton-associated processes, cell lysis, solubilization of particles and heterotrophic prokaryote activities (Nagata 2000; Carlson 2002; Jiao et al. 2010).

In the coastal ocean, the DOM cycle is influenced by terrestrial and anthropogenic inputs. Terrestrial inputs of DOM via rivers represent ~0.25 Pg C yr−1 (Hedges et al. 1997; Cauwet 2002; Fichot and Benner 2014) and would be sufficient to renew the oceanic DOM pool in ‘only’ 2800 years (Blough and Del Vecchio 2002). However, the signatures of terrestrial DOM in the open ocean and sediments remain very low (Hedges et al. 1997; Benner and Opsahl 2001), even though measurable amounts of dissolved lignin have been reported throughout the global ocean (Benner et al. 2005; Fichot et al. 2014). The current consensus is that photodegradation plays a central role in the removal of terrestrial DOM in the coastal ocean (Mopper and Kieber 2002; Mopper et al. 2015). In addition to its key role in carbon fluxes, DOM significantly influences the ecology of the coastal ocean through its optical and chemical properties. DOM is the main attenuator of harmful ultraviolet radiation (UVR) (Diaz et al. 2000; Tedetti and Sempéré 2006; Kuwahara et al. 2015), and it attenuates ~50% of photosynthetically available radiation (PAR) in surface waters (Siegel et al. 2005; Nelson and Siegel 2013). Also, DOM interacts with metals and organic pollutants, modifying their toxicity, availability and fate in coastal marine waters (Akkanen et al. 2004; Hirose 2007; Krachler et al. 2015). Although it is well known that DOM is a major pool in the coastal ocean, its biogeochemical cycle is still largely unknown. One of the main reasons for this lack of knowledge is that coastal DOM is an under-sampled reservoir at the spatio-temporal scale (Jaffé et al. 2008).

In recent years, the composition and dynamics of DOM in coastal environments (from freshwaters to coastal ocean waters) have been evaluated through the study of its fluorescent fraction, FDOM (Patel-Sorrentino et al. 2002; Stedmon et al. 2003; Boyd and Osburn 2004; Carstea et al. 2010; Huguet et al. 2010; Osburn et al. 2016). It has been found that FDOM mainly consists of two types of fluorescent components: protein- and humic-like components (see reviews by Hudson et al. 2007; Coble 2007; Fellman et al. 2010; Ishii and Boyer 2012). Protein-like fluorescence, which is issued by aromatic amino-acids (tryptophan and tyrosine) present at the free state or bound in proteins, has been often associated with a bioavailable fraction of DOM derived from autochthonous (marine) biological activities (see review by Stedmon and Cory 2014). Although protein-like components may be produced by phytoplankton activities and serve as labile substrates for heterotrophic prokaryotes (Yamashita and Tanoue 2004; Nieto-Cid et al. 2006), these components were produced from prokaryotic activities in incubation experiments (Cammack et al. 2004; Boyd and Osburn 2004). Prokaryotes can be both a sink and a source for protein-like components, suggesting that the direct link between protein-like fluorescence and bioavailable DOM is not so obvious. In addition, Maie et al. (2007) found that tryptophan-like fluorescent material is very likely related to terrigenous-derived compounds in coastal waters. On the other hand, humic-like materials, made up of higher-molecular weight (HMW) compounds, would be of terrestrial origin, issued from soil or plant organic matter that has been diluted and transformed during transit to and within the ocean (Blough and Del Vecchio 2002; Murphy et al. 2008; Andrew et al. 2013); but these compounds are also of marine origin, released from the prokaryotic degradation of organic matter (Rochelle-Newall and Fisher 2002; Stedmon and Markager 2005a; Yamashita et al. 2010) or from phytoplankton and zooplankton (Steinberg et al. 2004; Ortega-Retuerta et al. 2009; Romera-Castillo et al. 2010; Chari et al. 2013; Bittar et al. 2015). Therefore, in the coastal ocean, the origin of FDOM components is complex and cannot be seen through the prism of the simplistic model ‘autochthonous protein-like components versus allochthonous (terrestrial) humic-like components’.

The Rhône River is the major source of freshwater and terrigenous material to the Mediterranean Sea (Sempéré et al. 2000; Ludwig et al. 2009; The Mermex Group 2011; Guigue et al. 2014). Its presence induces a large transfer of particulate and dissolved materials and enhances the primary production in the northwestern basin due to nutrient and carbon inputs (Moutin et al. 1998; Sempéré et al. 2000). Although observations have shown that the Rhône River plume has a predominant westward direction with an extent and thickness dependent on its discharge, under specific meteorological conditions, the Rhône River plume may travel eastward in the direction of the Gulf of Fos-sur-mer and the Bay of Marseille (southern France) (Pairaud et al. 2011; Fraysse et al. 2014).

So far, only a few studies have investigated the characteristics of FDOM in the Rhône River (Para et al. 2010), the Fos-sur-mer area (Tedetti et al. 2010) and the Bay of Marseille (Para et al. 2010; Tedetti et al. 2010, 2012). Para et al. (2010) found that FDOM in the Bay of Marseille was dominated by protein- and marine humic-like components, very likely derived from phytoplankton activities, and was influenced by episodic events of the eastward extent of the Rhône River plume and wind-induced mixing of the water column. Although their study provided relevant information about factors controlling FDOM transport and production in the vicinity of the Rhône River and the Bay of Marseille, as noted by the authors, a higher-resolution sampling would allow a better understanding of the physical and biogeochemical processes driving the FDOM distribution in this coastal area. In this context, the present work aims to (i) characterize the FDOM composition in the Rhône River delta (Rhône River and Vaccarès Pond) and in the Fos-Marseille marine area (Gulf of Fos-sur-mer and Bay of Marseille) and (ii) assess the spatio-temporal variability of these FDOM components over the course of 1 year in relation to physical and chemical parameters.

Material and methods

Study sites

Five stations were sampled in the coastal northwestern Mediterranean Sea, southern France: two stations in the Rhône River delta (Arles and Vaccarès), hereafter denoted ‘inland waters’, and three stations in the Fos-Marseille marine area (Port-de-Bouc, Couronne and Sofcom), hereafter denoted ‘marine waters’ (Fig. 1; Table 1).
Fig. 1

Location of the five study sites in the Rhône River delta (Arles, Vaccarès) and the Fos-Marseille marine area (Port-de-Bouc, Couronne, Sofcom) (northwestern Mediterranean Sea, southern France), The distance between the Rhône mouth and Port-de-Bouc, Couronne and Sofcom is 13, 16 and 37 km, respectively. The Rhône River is connected to the Vaccarès Pond through a 6-km channel system. The detailed characteristics of these sites are provided in Table 1

Table 1

Characteristics of the study sites located in the Rhône River delta (Arles, Vaccarès) and the Fos-Marseille marine area (Port-de-Bouc, Couronne, Sofcom) (northwestern Mediterranean Sea, southern France), and sampled from February 2011 to January 2012

Station

Abbreviation

Water type

Position

Site depth

(m)

Sampling depth

(m)

Sampling dates

Arles

AR

Freshwaters (Rhône River)

43°40.72′ N, 4°37.27′ E

1–3

0.1

For AR and VA:

17/2, 24/2, 10/3, 25/3, 19/4, 3/5, 17/5, 9/6, 27/6, 5/7, 18/7, 5/9 15/9, 10/10, 24/10, 3/11, 17/11, 1/12 12/12, 9/1, 23/1

Vaccarès

VA

Brackish waters (Vaccarès Pond); Under the influence of the Rhône River and the Mediterranean Sea

43°31.44′ N, 4°38.14′ E

0.5–1.5

0.1

Port-de-Bouc

PB

Harbour marine waters (Gulf of Fos-sur-mer); Under the influence of the Berre Lagoon and potentially under the influence of the Rhône River

43°23.75′ N, 4°59.20′ E

12

0.1

For PB, CO and SO: 10/2, 15/2, 17/3, 28/3, 8/4, 9/5, 31/5, 7/6, 9/6, 21/6, 28/6, 7/7, 19/7, 7/9, 21/9, 14/10, 21/10, 22/11, 9/12, 11/1, 25/1

Couronne

CO

Marine waters (between Rhône mouth/Gulf of Fos-sur-mer and the Bay of Marseille); Potentially under the influence of the Rhône River

43°16.50′ N, 5°02.04′ E

90

0.1

Sofcom

SO

Marine waters (Bay of Marseille); Potentially under the influence of the Rhône River

43°14.50′ N, 5°17.50′ E

60

0.1

Arles station (AR, 1–3 m depth) is located close to the Compagnie Nationale du Rhône (CNR; http://www.cnr.tm.fr/fr/) along the Rhône River, which is the largest French river in term of water discharge (interannual average discharge: ~1700 m3 s−1). It is 812 km long and has a drainage basin of approximately 97,800 km2. Its delta has a surface area of ~1750 km2, composed mainly of wetlands, ponds and salt marshes. The central part of the delta, known as the ‘Camargue’ (~750 km2), is composed of two parts: farmlands in the northern part and salted ponds, where particular ecosystems are developed in the southern part. Camargue is an area that hosts horse and bull breeding, rice cultivation and salt exploitation. Because of its ecological importance to wildlife and fisheries, Camargue has been protected for decades.

The Vaccarès Pond (VA) is the largest pond of Camargue and covers 66 km2, with a mean water depth of 1.4 m and a maximum depth of 2.1 m. It is the main element of the control system of the Rhône delta waters. Seasonal water level fluctuations are generally limited to ±0.3 m (Chauvelon 1998). Freshwater supply from the catchment (317 km2) is mainly provided by four drainage channels connected to the Rhône River that are used for rice cultivation irrigation from April to September. VA is indirectly connected to the Mediterranean Sea through adjacent shallow ponds, and exchanges of water are controlled by sluices (Millet et al. 2010). The bottom sediments of the pond are covered with a 35 to 300 mm layer of unconsolidated fine particles (Vaquer and Heurteaux 1989). Although Camargue is a national wildlife park, VA may be subjected to biocides and hydrocarbons released from nearby agricultural and industrial activities (Comoretto et al. 2007; Roche et al. 2009; Guigue et al. 2014).

Port-de-Bouc (PB, 12 m depth) is a harbour situated in the Gulf of Fos-sur-mer and is surrounded by the Marseille-Fos petrochemical complex, which includes several chemical, petroleum and steel-work plants. PB continuously receives freshwaters from the Berre Lagoon through the Caronte channel (Ulses et al. 2005) and may episodically receive freshwaters from the Rhône River. Thus, the salinity at PB is always lower than that of Mediterranean waters. Furthermore, PB is positioned on the route of oil cargo ships entering the Gulf of Fos-sur-mer or going to the Berre Lagoon.

Couronne (CO, 90 m depth) is a marine nearshore site located 6 km off of PB between the Rhône River mouth/Gulf of Fos-sur-mer and the Bay of Marseille. Sofcom (SO, 60 m depth), situated in the Bay of Marseille (7 km off Marseille), is the nearshore observation station of the Mediterranean Institute of Oceanography (SOMLIT; http://www.domino.u-bordeaux.fr/somlit_national/) and has been sampled twice per month since 1995 for the measurement of hydrological and chemical parameters. CO and SO may receive freshwaters from the Rhône River depending on the Rhône discharge and wind conditions (Para et al. 2010; Pairaud et al. 2011; Fraysse et al. 2014).

Sampling and in situ measurements

During the study period (February 2011 to January 2012), meteorological parameters, i.e. air temperature, solar irradiance and rainfall intensity, were recorded every hour by the Frioul Island meteorological station, located in the Bay of Marseille. The Rhône River discharge was measured every hour by the Compagnie Nationale du Rhône.

At each site, surface water was collected at a 0.1 m depth approximately twice per month from February 2011 to January 2012 in the morning between 8:00 am and 12:00 pm (Table 1). Inland water samples (AR and VA) were taken directly from the edge, while marine samples (PB, CO, SO) were collected from the R/V Antédon 2. Inland water samples were taken directly with 4 L Nalgene® polycarbonate bottles. Marine samples were collected with a 5 L Niskin bottle equipped with silicon ribbons and a Vitton o-ring (to avoid organic contaminations) and then transferred to Nalgene® bottles. The bottles were washed with 1 M hydrochloric acid (HCl) and pure water (i.e. Milli-Q water) before use, were rinsed three times with the respective sample before filling and were stored in the dark in the cold (4–8 °C).

During the collection of marine samples (PB, CO, SO), in situ measurements of temperature, salinity and total chlorophyll a (TChl a) concentration along the water column were performed with a 19plus conductivity temperature depth (CTD) profiler (SeaBird Electronics Inc., Bellevue, USA) equipped with a WETStar Chl a fluorometer (WETLabs, USA). Because the deployment of the CTD profiler was not possible in shallow inland waters, the salinity at AR and VA was measured on the 0.1-m-depth discrete samples using a refractometer (MASTER-S/Millα, Atago, Tokyo).

Back in the laboratory, the samples were immediately filtered under a low vacuum (<50 mmHg) through pre-combusted (500 °C, 4 h) glass fibre GF/F filters (25-mm diameter, Whatman) using all-glass filtration systems. The latter were washed twice with 100 mL of pure water and once with 50 mL of the sample of interest. The GF/F filtered water was transferred into pre-combusted glass bottles for FDOM, dissolved organic carbon (DOC), dissolved organic phosphorus (DOP) and dissolved organic nitrogen (DON) analyses and in polypropylene Nalgene® bottles for nitrate (NO3), nitrite (NO2) and phosphate (PO43−) analyses. The FDOM samples were stored at 4 °C in the dark and were analysed within 24 h. The samples for DOC, DON, DOP, NO3, NO2 and PO43− were stored at −20 °C before analysis. 0.5 to 2 L of water was filtered for TChl a, particulate organic carbon (POC), particulate organic nitrogen (PON) and particulate organic phosphorus (POP) analyses, and the GF/F filters were stored in glass tubes at −20 °C before analysis.

Chemical analyses

Nutrient concentrations were measured with an AutoAnalyseur III Seal Bran Luebbe (Mequon, USA) using the method described in a previous study (Tréguer and LeCorre 1975). The quality of the measurements was assured by the use of standards and their comparison to commercial products (OSIL). The detection limits of NO3, NO2 and PO43− were respectively 0.05, 0.05 and 0.02 μM.

The TChl a concentration was measured according to Raimbault et al. (2004). Two hundred millilitres of seawater was filtered through 25-mm Whatman filters. The filters were placed in a glass tube and kept frozen (−20 °C) until analysis. At the laboratory, 5 mL of pure methanol was added in the glass tube. Following 20–30 min of extraction, the fluorescence of the extract was determined on a Turner Fluorometer 110 equipped with the Welschmeyer kit to avoid chlorophyll b interference (Welschmeyer, 1994). Because the monochromatic fluorescence method cannot separate divinyl chlorophyll from chlorophyll a, the results are given in terms of total chlorophyll a concentration (TChl a), i.e. the sum of Chl a and divinyl Chl a. The blank ‘methanol + filter’ was close to zero, giving a very low detection limit of approximately 0.01 μg L−1. Calibrations were made using a pure Sigma Chl a standard.

Samples for particulate organic carbon (POC) and total particulate nitrogen (TPN) and phosphorus (TPP) analyses were filtered onto a glass fibre Whatman GF/F (25 mm in diameter, 0.7-μm pore size) precombusted at 500 °C for 4 h. Between 250 and 1200 mL of sample was filtered depending on the quantity of particulate matter in the sample. The filters were then stored frozen in 25-mL Schott glass bottles for subsequent laboratory analysis. Persulfate wet-oxidation was used to digest the organic matter collected on the filters, and the inorganic end-products were determined by colorimetry according to Raimbault et al. (1999). The oxidation filters had been washed with 100 μL of H2SO4 (0.5 N) to remove inorganic carbon. Then, 20 mL of DIW and 2.5 mL of oxidizing reagent were added in the bottles. Blank filters were prepared for each set of samples by washing the filter with 20 mL of 0.2-μm-filtered seawater. POC, TPN and TPP have detection limits of 0.50, 0.10 and 0.02 μM, respectively.

Samples for total organic matter determination (TOC, TN, TP) were collected directly from the Niskin bottles into 50-mL glass Schott bottles. The samples were frozen and stored for subsequent analysis at the laboratory. Prior to oxidation, the samples were immediately acidified with 100 μL H2SO4 0.5 N and bubbled with a high-purity oxygen/nitrogen gas stream for 15 min. Persulfate wet-oxidation was used to digest the organic matter, and the inorganic end-products were determined by colorimetry according to Raimbault et al. (1999). All reagents and sample blanks were prepared using fresh Millipore Milli-Q plus® water.

Total organic nitrogen (TON) and total organic phosphorus (TOP) were calculated as TN and TP minus dissolved inorganic nitrogen (nitrate + nitrite + ammonium) or phosphate measured in the same samples. Dissolved organic carbon (DOC), dissolved organic nitrogen (DON) and dissolved organic phosphorus (DOP) were calculated from these total organic fractions by subtracting values of POC, PON and POP obtained from the >GF/F fractions (see above). The detection limits of DOC, DON and DOP were 10, 0.1 and 0.05 μM, respectively. Deep Sargasso Sea reference water (Hansell Laboratory, Bermuda Biological Station for Research) was used to check the analytical procedure for the DOC determination. The results obtained on five comparative samples were close to the nominal value of 45 μM (i.e. 44.8 ± 5.8 μM). For each chemical analysis, the number of sample replicates was equal to 1.

FDOM analysis and data processing

FDOM analyses were performed with a Hitachi F-7000 spectrofluorometer (Tokyo, Japan), which measures fluorescence from 200 to 750 nm for both excitation (Ex) and emission (Em). This instrument is equipped with a 150 W xenon short-arc lamp with a self-deozonating compartment, two stigmatic concave diffraction gratings with 900 lines/mm brazed at 300 (Ex) and 400 nm (Em) as single monochromators, and Hamamatsu R3788 (185–750 nm) photomultiplier tubes (PMTs) as reference and sample detectors (fluorescence measurements acquired in signal over reference ratio mode). Spectral corrections for Ex and Em were executed according to the manufacturer (Hitachi F-7000 Instruction Manual). They are fully described elsewhere (Tedetti et al. 2012).

One hour before fluorescence analyses, the samples were allowed to reach room temperature in the dark. They were then transferred into a 1-cm pathlength far UV silica quartz cuvette (170–2600 nm; LEADER LAB®), thermostated at 20 °C in the cell holder by an external circulating water bath. Excitation-emission matrices (EEMs) were generated for Ex wavelengths (λEx) between 200 and 500 nm in 5-nm intervals and for Em wavelengths (λEm) between 280 and 550 nm in 2-nm intervals, with a 5-nm slit width on both Ex and Em, a scan speed of 1200 nm min−1, a time response of 0.5 s, and a PMT voltage of 700 V (Ferretto et al. 2014). Blanks (pure water) and solutions of quinine sulphate dihydrate (Fluka, purum for fluorescence) in 0.05 M sulphuric acid (H2SO4) from 0.5 to 50 μg L−1 were run with each set of samples. For each sample, successive analyses (i.e. 5 EEMs) were carried out, from which an average EEM was determined and further processed (see below). Because the pH variation was very low within samples (7.6–8.2), no pH adjustment was carried out for the EEM measurements.

To correct the average EEMs for inner filtering effects, absorbance measurements were performed between 200 to 550 nm in a 1-cm pathlength quartz cuvette with a Shimadzu UV-Vis 2450 spectrophotometer. The reference used was pure water for continental waters and a salt solution (pure water with pre-combusted NaCl; Sigma-Aldrich) for marine waters. Each EEM was multiplied by a correction matrix calculated for each wavelength pair from the sample absorbance, assuming a 0.5-cm pathlength of Ex and Em light in a 1-cm cuvette (Ohno 2002). From absorbance measurements, we also determined the specific ultraviolet absorbance (SUVA254, in L mg-C−1 m−1) by dividing the absorption coefficient at 254 nm (a254 in m−1) by the DOC concentration (mg L−1) according to Weishaar et al. (2003) and Helms et al. (2008). The EEMs were then blank-corrected by subtracting the pure water EEM (i.e. the average of the filtered pure water EEMs generated before the analyses of a set of samples). Finally, the EEMs were converted into quinine sulphate unit (QSU), 1 QSU corresponding to the fluorescence of 1 μg L−1 quinine sulphate in 0.05 M H2SO4 at λExEm of 350/450 nm (5 nm slit width). The conversion into QSU was made by dividing the EEM fluorescence data by the slope of the quinine linear regression. The detection and quantification limits of the fluorescence measurement were 0.10 and 0.40 QSU, respectively. The Raman scatter peak of pure water was monitored at λExEm of 275/303 nm during the whole study period. Because its variability was very low (coefficient of variation =6%, n = 65), normalization of the fluorescence intensities was not applied to our dataset. The Raman scatter peak was integrated from λEm 380 to 426 nm at λEx of 350 nm for ten pure water samples. The average value was used to establish the conversion factor between the QSU and Raman units (RU, nm−1), based on the Raman-area normalized slope of the quinine linear regression (Murphy et al. 2010). The conversion factor was 0.012 RU per QSU.

PARAFAC analysis

PARAFAC is a multi-way statistical method based on an alternating least squares algorithm and is used to decompose the complex measured EEM signal into its underlying individual fluorescence profiles (components). Here, PARAFAC models were created and validated according to the method by Stedmon and Bro (2008). The three important assumptions for the PARAFAC application are (1) variability: components cannot have identical Ex and Em spectra, nor concentrations that are perfectly correlated; (2) trilinearity: Em spectra do not vary across λEx, Ex spectra do not vary across λEm and fluorescence intensities are not affected by inner filtering effects; and (3) additivity: fluorescence is due to the linear superposition of a fixed number of components (Murphy et al. 2014).

For a better representation of the FDOM composition within samples, inland and marine water datasets were processed separately due to their differences in salinity and fluorescence intensities (Stedmon and Bro 2008). The EEM wavelength ranges used for the PARAFAC analyses were 210–500 nm (Ex) and 280–550 nm (Em). The number of samples was 42 (inland samples) and 94 (marine samples). PARAFAC was operated using the DOMFluor toolbox v1.6. running under MATLAB® 7.10.0 (R2010a). Two to eight components were tested. With regard to sample and wavelength leverages, no outlier was initially present in the datasets. The validation of the two PARAFAC models (running with the non-negativity constraint) and the determination of the correct number of components were achieved through the examination of (1) the percentage of explained variance, (2) the shape of the residuals, (3) split half analysis and (4) random initialization using the Tucker Congruence Coefficients (Stedmon and Bro 2008).

Statistical analysis

For linear regressions and principal component analyses (PCA), spearman’s (rank-order) correlations were preferred to Pearson’s correlations because of the high amplitudes of some variables and their non-normal distribution (Joliffe 1986). In the same way, for the comparison of two independent data groups, the Mann-Whitney non-parametric test (U-test) was used rather than analysis of variance (ANOVA). The normality test (Kolmogorov–Smirnov test), linear regressions, PCA and U-test were performed with XLSTAT 2013.5.01 (Microsoft Excel add-in program). For the different analyses and tests, the significance threshold was set at p < 0.05.

Results and discussion

Meteorological conditions and Rhône River discharge

In the whole study area during the sampling period (February 2011–January 2012), the air temperature ranged from 4.6 (31 January) to 29.5 °C (24 August) and the solar irradiance varied from 4.4 (5 November) to 391.2 W m−2 (30 June) at solar noon (supplementary material 1). On average, the rainfall intensity was higher in the autumn-winter period (October–March) than in the spring-summer period (April–September) (U-test, p < 0.05), even though unusual elevated intensities were recorded in July (Fig. 2a). A rainfall intensity >20 mm h−1 was observed several times during the year (14 February, 29 April, 27 July, 4 September), with up to 60 mm h−1 on 5 November. Samples were collected under rainfall conditions or a day after rainfall on the following dates: 17 February, 9 June, 5 July, 5 September and 3 November for inland waters, and 15 February, 17 and 28 March, 31 May, 7, 9 June and 19 July for marine waters (Fig. 2a).
Fig. 2

a Rainfall intensity (mm h−1) in the study area and b Rhône River discharge (m3 s−1) during the sampling period (February 2011–January 2012 for rainfall intensity and February 2011–December 2011 for Rhône River discharge). Sampling dates for inland waters (AR, VA: red circles) and marines waters (PB, CO, SO: blue triangle) are superimposed on each figure

The Rhône River discharges for 2011–2012 were, on average, 1098 ± 625 m3 s−1, with higher values in winter (1488 ± 580 m3 s−1) than in summer (839 ± 251 m3 s−1) (U-test, p < 0.05) (Fig. 2b). Minimal values were observed in May, August, September and October, while discharges ≥2000 m3 s−1 were recorded on 14–18 March, 18, 19 July, 3–9 November, 8, 9 and 12–29 December. During the study period, the Rhône River discharge remained under critical values, except from 4 to 7 November when a flood occurred (discharge >3800 m3 s−1). Apart from this event, no obvious match was found between rainfall intensity and the Rhône discharge (Fig. 2a, b). Sampling was carried out just before the flood (on 3 November for inland waters) and several days after (on 17 and 22 November for inland and marine waters, respectively). Several samples were collected under discharge conditions ≥2000 m3 s−1 in inland waters (on 18 July, 3 November, 12 December) and marine waters (on 17 March, 19 July, 9 December) (Fig. 2b).

Spectral characteristics and identification of PARAFAC components

Three PARAFAC components from each group of modelled samples (inland and marine sites) were validated: components C1–C3 for the inland sample dataset (supplementary material 2) and components C4–C6 for the marine sample dataset (supplementary material 3). The fluorescence maxima of each component are summarized in Table 2. These components have been already detected in the aquatic environment and belong to the humic-like (C1, C3, C6) and protein-like (C2, C4, C5) types. C1, which presents maximal absorption in the UVC and UVA domains, corresponds to humic peaks A + M in Coble’s (1996) classification and to component 3 in the Ishii and Boyer’s (2012) classification (Table 2; supplementary material 2). C3 also displays maximal absorption in the UVC and UVA domains and corresponds to humic peaks A + C in Coble’s (1996) classification and to component 2 in Ishii and Boyer’s (2012) classification. C3 is very likely of higher molecular weight, is more hydrophobic, and is more photodegradable compared to C1, as indicated by its longer Ex and Em wavelengths (Table 2; supplementary material 2). C6 is the only humic-like component found in marine waters (PB, CO, SO sites). Although it stricto senso corresponds to component 3 in Ishii and Boyer’s (2012) classification (as C1), this component is clearly intermediate between peak A + M and peak A + C (and, thus, between C1 and C3, being close to C1 in terms of λEx and close to C3 with regard to λEm). Thus, C6 mainly absorbs in the UVC and UVB domains (Table 2; supplementary material 3). C2 and C5 correspond to a tryptophan-like component (peaks T1 + T2), whereas C4 is associated with a tyrosine-like component (peaks B1 + B2) (Coble 1996; Hudson et al. 2007) (Table 2; supplementary materials 2 and 3).
Table 2

Excitation and emission maxima (λEx and λEm max) of the six components validated by the PARAFAC models applied on inland water samples of the Rhône River delta (C1-C3 components) and on marine water samples of the Fos-Marseille area (C4-C6 components), and identification of these components (type/nature and possible origins) by comparison with the literature data

This study

Previous studies

Identification

Comp.

λExaEm max (nm)

area

Comp.

λExaEm (nm)

Area

Ref.

Type/Nature

Possible origin in aquatic media

C1

250(330)/394

Rhône River delta

C3

C4

C3

C4

C2

<240–260(295–380)/374–450

<240(320)/400

250(310)/400

250(325)/416

235(300)/404

-

Atlantic and Pacific oceans

South Atlantic Bight

Danish estuary catchment

Lake Taihu, China

1

2

3

4

5

UVC + UVA humic-like

Peaks A + Mb

Oxidized quinone-likec

Autochthonous (heterotrophic prokaryotes or phytoplankton), terrestrial or anthropogenic

C2 and C5

230(280)/340, 342

Rhône River delta, Fos-Marseille marine area

C5

C7

C2

C4

P7

<240(280)/368

280/344

<230(270)/346

≤225(285)/344

280/342

Danish estuary catchment

Danish estuary catchment

Bay of Marseille, France

Yungui Plateau lakes, China

Atlantic and Pacific oceans

4

6

7

8

9

Tryptophan-like

Peaks T1 + T2b

Amino-acid free or bound in proteins, LMW DOM

Autochthonous (heterotrophic prokaryotes or phytoplankton) or anthropogenic

C3

250(350)/454

Rhône River delta

C2

C3

C3

C5

C1

<240–275(339–420)/434–520

270(360)/478

250(355)/461

255(365)/474

255(350)/471

-

Danish estuary catchment

Lake Taihu, China

Bay of Marseille, France

Yungui Plateau lakes, China

1

4

5

7

8

UVC + UVA humic-like

Peaks A + Cb

Reduced quinone-likec

HMW DOM

Terrestrial or autochthonous (heterotrophic prokaryotes)

C4

<220(275)/<300

Fos-Marseille marine area

C8

C1

C3

P5

C4

274/304

<230(275)/306

≤225(285)/322

270/310

275/306

Danish estuary catchment

Bay of Marseille, France

Yungui Plateau lakes, China

Atlantic and Pacific oceans

Okhotsk Sea

6

7

8

9

10

Tyrosine-like

Peaks B1 + B2b

Amino-acid free or bound in proteins, LMW DOM

Autochthonous (heterotrophic prokaryotes or phytoplankton) or anthropogenic

C6

245(300)/450

Fos-Marseille marine area

C3

C3

P1

<240–260(295–380)/374–450

250(310)/400

<260(310)/414

-

South Atlantic Bight

Atlantic and Pacific oceans

1

3

9

UVC + UVB/UVA humic-like

Peaks A + M/Cb

Autochthonous (heterotrophic prokaryotes or phytoplankton), terrestrial, photoproduct

References (Ref): 1. Ishii and Boyer (2012); 2. Jørgensen et al. (2011); 3. Kowalczuk et al. (2009); 4. Stedmon et al. (2003); 5. Yao et al. (2011); 6. Stedmon and Markager (2005b); 7. Tedetti et al. (2012); 8. Zhang et al. (2010); 9. Murphy et al. (2008); 10. Yamashita et al. (2010)

aThe secondary Ex maximum is given in brackets; b Coble’s (1996) classification; c Cory and McKnight (2005)

Spatial variability of physical/chemical parameters and FDOM components

The inland waters of the Rhône River delta (AR and VA) and the PB marine site, for which the shallower water column was more sensitive to air temperature, displayed higher surface temperature variations (8.2–23.1, 8.0–21.8 and 8.7–22.5 °C, respectively) than offshore marine sites CO and SO (12.5–20.6 and 13.0–19.0 °C, respectively) (Fig. 3a). Surface salinity exhibited great variations between sites. While freshwater dominated at AR, the salinity ranged from fresh to brackish water at VA (0–14) as a result of hydraulic exchanges between the Camargue freshwater hydrosystem and the Mediterranean Sea (Fig. 3b). Concerning the Fos-Marseille marine area, salinity showed a west-east gradient (from the Gulf of Fos-sur-mer to the Bay of Marseille), with a significant increase from PB (24.8–36.8) to CO (34.7–38.1) and SO (37.7–38.2) (Fig. 3b; U-test, p < 0.05). These salinity values indicate freshwater intrusions from the Rhône River and the Berre Lagoon in PB, as well as, to a much lesser extent, in CO.
Fig. 3

Temporal variation of a temperature (T, in °C), b salinity (S), c total chlorophyll a (TChl a, in μg L−1), d nitrates plus nitrites (NO3 + NO2, in μM), e phosphates (PO43−, in μM), f specific ultraviolet absorbance at 254 nm (SUVA254, in L mg-C−1 m−1), g dissolved organic carbon (DOC, in μM), h dissolved organic nitrogen (DON, in μM), i dissolved organic phosphorus (DOP, in μM), j particulate organic carbon (POC, in μM), k particulate organic nitrogen (PON, in μM) and l particulate organic phosphorus (POP, in μM) in inland (AR, VA) and marine waters (PB, CO, SO) from February 2011 to January 2012. The shaded area corresponds to the period spring-summer (April–September) and the non-shaded one to the period autumn-winter (October–March). All graphics present a secondary Y-axis, except T and SUVA254

The concentrations of TChl a, nutrients (NO3 + NO2, PO43−), dissolved organic matter (DOC, DON, DOP) and particulate organic matter (POC, PON, POP), as well as FDOM intensities and SUVA254 index values, were much higher in the Rhône River delta (AR, VA) than in the Fos-Marseille marine waters (PB, SO, CO) (U-test, p < 0.05) (Fig. 3c–l, Fig. 4a–f). Besides this quantitative aspect, the Rhône River delta waters showed a much higher proportion of humic-like material within the FDOM pool (61–90%), with C1 and C3 components, compared to marine waters (16–45%) (Fig. 4a–f), which only displayed one (C6) humic-like component. The FDOM composition at AR and VA was thus typical of aquatic systems strongly influenced by (highly aromatic) terrestrial organic matter originating from higher land plants and soils and transported through rain waters and runoffs. However, clear differences also appeared within inland waters. VA presented significantly higher concentrations of TChl a, and dissolved and particulate organic matter. VA also presented significantly higher FDOM fluorescence intensities and significantly lower nutrient concentrations and lower SUVA254 values compared to AR (U-test, p < 0.05, except for DOP) (Fig. 3c–l, Fig. 4a–c). For many river systems as AR, dissolved organic matter and, thus, FDOM may be issued from watershed soil and higher plants, as well as autochthonous biological activity (Rostan and Cellot 1995; Cauwet 2002; Panagiotopoulos et al. 2012). The higher concentrations in organic matter and the higher FDOM fluorescence intensities recorded at VA could be explained by a more elevated autochthonous biological activity (as revealed by TChl a concentrations), which would be favoured (compared to AR) by still waters and by more-important runoff inputs, including those from rice paddies, that are discharged directly into the Vaccarès pond through two drainage channels (Chauvelon et al. 2003). In addition, VA is subjected to a strong wind forcing. Interestingly, recent studies have highlighted the dominant role of wind velocity in controlling the concentrations of suspended matter in this shallow pond (Banas et al. 2005; Millet et al. 2010; Boutron et al. 2015). Hence, the wind-induced input of particulate organic matter into the water column could contribute to FDOM as well.
Fig. 4

Temporal variation of PARAFAC components (fluorescence intensity in QSU): a C1, b C2 and c C3 for inland waters (AR, VA), and d C4, e) C5 and f C6 for marine waters (PB, CO, SO) from February 2011 to January 2012. The shaded area corresponds to the period spring-summer (April–September) and the non-shaded one to the period autumn-winter (October–March). All graphics present a secondary Y-axis, except C4

In the Fos-Marseille marine waters (PB, CO, SO), the FDOM fluorescence intensities were much lower than in inland waters and were dominated by protein-like material, which accounted for 55–84% of the FDOM pool (Fig. 4d–f). This FDOM composition reflects a higher contribution of autochthonous biological activity and a lower contribution of terrestrial material compared to inland waters. As for salinity, a clear west-east (from Fos to Marseille) gradient emerged for chemical parameters and FDOM, with a decrease in the concentrations of TChl a, nutrients, dissolved and particulate organic matter and FDOM fluorescence intensities (Fig. 3b–l, Fig. 4d–f). Actually, all the chemical parameters (TChl a, nutrients, organic matter) and FDOM components C4-C6 presented a significant negative linear correlation with salinity (r = −0.40 ˗ −0.82, n = 48, p < 0.05), demonstrating the influence of freshwater inputs on the biogeochemistry in this coastal marine area, especially at PB. This site is known to be strongly impacted by the Rhône River and the Berre Lagoon. The latter is connected to the Gulf of Fos-sur-mer, in the vicinity of Port-de-Bouc, through the Caronte channel (Ulses et al. 2005) (Fig. 1). The Berre Lagoon is considered a eutrophic system, with TChl a concentrations as high as 33 μg L−1 (Delpy et al. 2012). Therefore, the relatively high FDOM fluorescence intensities recorded at PB very likely mirrored more-elevated autochthonous biological activity and terrestrial influence compared to CO and SO. It should be noted that despite this proximity to the Fos petrochemical complex, the FDOM at PB did not present any ‘hydrocarbon-like component’, as reported by Ferretto et al. (2014) in the same area.

Temporal variability of physical/chemical parameters and FDOM components in the Rhône River delta

The Rhône River

In the Rhône River (AR), humic-like materials C1 and C3 tended to increase in the autumn-winter period (October–March), with fluorescence intensities ranging from 10.1 to 34.8 and 10.3 to 29.1 QSU, respectively, compared to the spring-summer period (April–September), when they ranged from 9.4 to 12.0 and 10.0 to 15.1 QSU (U-test, p < 0.05) (Fig. 4a, c). The highest C1 and C3 fluorescence intensities were recorded on 9 January and 12 December (Fig. 4a, c, Fig. 5a, b). Even though the Rhône River discharge was not available for the first date, it was relatively high on 12 December (Fig. 2b). The highest concentrations of POC, PON and POP were also observed on 9 January at AR (Fig. 3j–l). Since high river discharges are usually associated with an increase in particulate load, we may assume that the Rhône River discharge was also elevated on 9 January. The same trend was observed for NO3 + NO2 and DOC concentrations, which varied within the ranges 77.4–126.6 and 114.4–284.3 μM, respectively, in autumn-winter and 69.9–107.3 and 75.9–150.3 μM in spring-summer (U-test, p < 0.05) (Fig. 3d, g). C1 and C3 were strongly correlated with each other (r = 0.94, n = 16, p < 0.05) and with SUVA254 (r = 0.66–0.79, n = 16, p < 0.05) but were not correlated with tryptophan-like component C2 (r = 0.00–0.19, n = 16, p > 0.05) (Fig. 5a). Indeed, the fluorescence intensity of C2 was higher in spring-summer (7.5–14.3 QSU) than in autumn winter (5.9–10.8 QSU) (U-test, p < 0.05) (Fig. 4b) and was significantly correlated with temperature (r = 0.55, n = 16, p < 0.05) (Fig. 3a, Fig. 5a). The highest fluorescence intensity for C2 was measured on 9 June. To a lesser extent, this trend was also found for the TChl a concentration, which varied from 1.7 to 14.7 μg L−1 in spring-summer and from 0.40 to 16.0 μg L−1 in autumn-winter (Fig. 3c, Fig. 5a). SUVA254 is a relevant index of dissolved aromatic carbon content in aquatic systems (Weishaar et al. 2003). Thus, the increase in humic-like components, DOC, and SUVA254 in autumn-winter, which was marked by higher rainfall intensities and Rhône discharges (Fig. 5a, b), may reflect a higher contribution of terrestrial DOM via runoffs. While humic-like components C1 and C3 seemed related to terrestrial DOM, the tryptophan-like component C2, closer to temperature and TChl a concentration (i.e. phytoplankton biomass) in terms of seasonal pattern (Fig. 5a), was very likely more associated with Rhône River autochthonous biological activity (Rostan and Cellot 1995; Cauwet 2002; Panagiotopoulos et al. 2012).
Fig. 5

Principal component analysis (PCA) based on Spearman’s rank-order correlation matrix computed on the hydrological, chemical and FDOM parameters, applied to each water body: a and b AR, c and d VA (inland waters), e and f PB, g and h CO, i and j SO (marine waters). Correlation circle is displayed along with the projection of samples on the first factorial plane (F1 versus F2). For each water body, some samples (5 or AR, 3 for VA, 1 for PB, 2 for CO and 5 for SO) are not included in PCA because of missing data for TChl a, DOC or SUVA254

The Vaccarès pond

In the Vaccarès pond (VA), the seasonal pattern of salinity (0.0–4.0 in spring-summer, 0.0–15.0 in autumn-winter), NO3 + NO2 concentration (0.9–36.8 μM in spring-summer, 1.7–12.7 μM in autumn-winter) and particulate organic matter (50–278.4 μM in spring-summer, 12.4–136.8 μM in autumn-winter for POC) (Fig. 3b–j-l, Fig. 5c, d) may be explained by hydraulic exchanges between the Camargue hydrosystem and the Mediterranean Sea. Indeed, at VA, rice field irrigation with the Rhône River waters occurred from the beginning of April to the end of September (spring-summer period), which led to a significant salinity decrease during this period (U-test, p < 0.05). Due to the use of fertilizers, the irrigation water pumped from the Rhône River and crossing fields transported a high quantity of NO3 and NO2 and particulate matter (POC, PON, POP) to the pond, which explains the higher values observed for these parameters in spring-summer (U-test, p < 0.05, except for PON) (De Groot and Golterman 1999; Chauvelon et al. 2003; Banas et al. 2005). In contrast, the autumn-winter period was marked by higher salinity as well as lower concentrations in NO3 + NO2 and particulate organic matter due to the cessation of irrigation in addition to Mediterranean water intrusions into the pond (Fig. 3b, d, j–l, Fig. 5c, d). In VA, humic-like components C1 and C3, which were strongly correlated with each other (r = 0.86, n = 18, p < 0.05), displayed quite good correlations with protein-like component C2 (r = 0.61–0.72, n = 18, p < 0.05) (Fig. 5c), contrary to what was observed in the Rhône River. These three components presented higher fluorescence intensities in autumn-winter, with values ranging from 32.4 to 144.5 QSU for C1, 19.3 to 32.9 QSU for C2 and 41.2 to 124.3 QSU for C3, compared to spring-summer (40.0–67.1 QSU for C1, 14.4–36.1 QSU for C2 and 38.2–62.8 QSU for C3) (U-test, p < 0.05, except for C1). The highest fluorescence intensities of C1 and C3 were observed on 1 December and 17 November, while that of C2 was found on 9 June (Fig. 4a–c, Fig. 5c, d). It appears that the high fluorescence intensity of the FDOM components encountered in November–December at VA could be related to inputs from surface runoffs induced by the strong rainfall episode that occurred on 5 November (Fig. 2a, Fig. 5d). The DOM parameters (DOC, DON, DOP) (Fig. 3g–i) also showed higher concentrations in autumn-winter (887.0–1725.8 μM for DOC) than in warmer months (312.9–622.8 μM for DOC) (U-test, p < 0.05 for DOC) and were correlated with FDOM components (r = 0.49–0.52, n = 18, p < 0.05). In contrast to the Rhône River, humic-like components C1 and C3 were not coupled with SUVA254 in VA waters (r = 0.12–0.39, n = 18, p > 0.05) (Fig. 5c). Moreover, the three FDOM components presented an inverse seasonal pattern to that of the TChl a concentration, which was, on average, higher in spring-summer, although the U-test resulted in p > 0.05 (Fig. 3c, Fig. 5c, d). Thus, in the Vaccarès pond, the irrigation of rice fields in the Rhône delta (occurring in spring-summer) was probably a weaker source of C1 and C3 compounds compared to the inputs from the rainfall-induced runoff that occurred in autumn-winter. Besides inputs of terrestrial organic matter from rainfalls and runoffs, this increase in the fluorescence intensities of humic-like materials in autumn-winter at VA (Fig. 5c, d) could be related to inputs from sediment particle resuspension due to strong winter winds (Banas et al. 2005; Millet et al. 2010; Boutron et al. 2015).

The occurrence of humic-like component C1 is widespread, having been found in marine waters, estuaries, lakes and agricultural/forest streams, and is thought to be derived from autochthonous aquatic activities and terrestrial or anthropogenic organic matter (Stedmon and Markager 2005b; Borisover et al. 2009; Zhang et al. 2010; Jørgensen et al. 2011; Yamashita et al. 2011) (Table 2). Humic-like component C3, observed in many aquatic environments, has been associated with terrestrial organic matter (Luciani et al. 2008; Murphy et al. 2008; Kowalczuk et al. 2009) but also with autochthonous aquatic activities (Yamashita et al. 2010; Jørgensen et al. 2011) (Table 2). On the other hand, tryptophan-like components (C2 and C5) are generally related to autochthonous aquatic production (phytoplankton activity) or anthropogenic activities (Parlanti et al. 2000; Lønborg et al. 2010; Romera-Castillo et al. 2010; Yao et al. 2011; Tedetti et al. 2012) (Table 2) but may also be directly associated with humic substances, representing their bioavailable fraction (Stedmon and Cory 2014). Consequently, with regard to their multiple possible origins, it appears very difficult to determine the specific sources of FDOM components in AR and VA waters. Humic-like components C1 and C3 seems to be driven by inputs of terrestrial organic matter from rainfalls, runoffs and wind-induced sediment resuspension (especially for VA) in autumn-winter. Consequently, at AR, they presented high fluorescence intensities when high Rhône discharges occurred (Fig. 4a, c, Fig. 5b, d). However, the seasonal variability of tryptophan-like component C2 was different between the AR and VA sites. C2 was slightly more important in spring-summer at AR, reflecting the seasonal cycle of biological activity (Fellman et al. 2010). In contrast, at VA, C2 was more abundant during the coldest season, as humic-like materials (Fig. 4b, Fig. 5b, d). Hence, it is very likely that this component was subjected to different processes in these two water bodies. In spring-summer, in addition to less-important terrestrial and sediment inputs, the lower fluorescence intensities of humic-like components observed for both AR and VA waters might be explained by photodegradation processes. Indeed, according to their excitation (absorption) wavelengths, C3, and to a lesser extent C1, are submitted to photodegradation under natural solar radiation (Para et al. 2010; Yamashita et al. 2010; Jørgensen et al. 2011).

Temporal variability of physical/chemical parameters and FDOM components in the Fos-Marseille marine area

Port-de-Bouc

At PB, the salinity, which ranged from 24.8 (10 February) to 36.8 (9 December), was lower than the typical salinity values of the Mediterranean Sea (~38). A salinity <30 was recorded on 10 February, 8 April, 9 May, 7, 21 September and 21 October (Fig. 3b; supplementary material 4). Although freshwater intrusions occurred throughout the year, they were more frequent in the spring-summer period (Fig. 3b; supplementary material 4). The highest TChl a concentrations were found in spring-summer (on 9 June and 7 September) and were associated with an increase in temperature and a decrease in salinity (Fig. 3c; supplementary material 4). Additional information from remotely sensed images of surface salinity and TChl a concentrations (supplementary material 5) illustrated Rhône River intrusions at PB on 10 February, corresponding to periods of relatively high Rhône River discharges (Fig. 2b). Conversely, the low salinity event on 8 April at PB was neither related to an exceptional rain event nor to a high Rhône River discharge (Fig. 2a, b). Therefore, this event in April would have been due to freshwater inputs from the Berre Lagoon, as has been shown previously (Ulses et al. 2005). Typically, whereas the nutrient concentrations at PB were within the range of nutrient concentrations of the Mediterranean Sea surface waters in spring-summer (0.55 μM ± 0.13 and 0.04 ± 0.01 μM for NO3 and PO43−, respectively) (Céa et al. 2015), these concentrations increased by a factor 12 and 4 for NO3 + NO2 and PO43− during the Rhône intrusion events on 10 February and 21 October, respectively (Fig. 3d, e, Fig. 5e, f). During the study period, tyrosine-like component C4, tryptophan-like component C5 and humic-like material C6 varied from 1.9 to 12.8, 4.5 to 17.6 and 3.6 to 15.7 QSU, respectively, without displaying any clear seasonal pattern. Nevertheless, their highest values were recorded during strong freshwater intrusion events on 7 and 21 September and 21 October (Fig. 4d–f, Fig. 5e, f). These three FDOM components were significantly correlated with each other (r = 0.66–0.88), although C4 was slightly decoupled from C5 and C6 (Fig. 5e). Also, C4 was significantly positively correlated with temperature (r = 0.53), TChl a (r = 0.70) and particulate organic matter (r = 0.59–0.62), while C5 and C6 were correlated with PO43− (r = 0.53–0.58) and dissolved organic matter (r = 0.53–0.77, n = 17, p < 0.05) (Fig. 5e). The PO43− concentration was inversely correlated with salinity (r = −0.56, n = 17, p < 0.05).

Couronne

The CO site, with salinities between 34.7 (9 December) and 38.1 (14 October), presented several freshwater intrusions (detected when the salinity was <37.7) on 28 March, 31 May, 28 June, 7 July, 7, 21 September, 9 December and 25 January (Fig. 3b; supplementary material 4). Moreover, increases in surface TChl a concentrations were recorded at CO when freshwater intrusion events occurred (May–June, September and December) (Fig. 3c; supplementary material 4). Remotely sensed images (supplementary material 5) illustrated Rhône River intrusions at CO on 9 December, corresponding to a period of relatively high Rhône River discharges (Fig. 2b). Tyrosine-like component C4, tryptophan-like component C5 and humic-like material C6 displayed higher fluorescence intensities in spring-summer (3.4–8.5, 2.3–4.7 and 1.3–3.7 QSU, respectively) than in autumn-winter (1.2–6.9, 1.1–3.1 and 1.1–3.9 QSU, respectively) (U-test, p < 0.05, except for C6) and showed the majority of their highest values during freshwater intrusion events, i.e. on 21 September, 7 July and 9 December (Fig. 4d–f, Fig. 5g, h). These three components were significantly correlated with each other (r = 0.73–0.87 n = 16, p < 0.05) (Fig. 5g). They were all significantly positively correlated with temperature (r = 0.74–0.84), DOC (r = 0.55–0.72) and particulate organic matter (r = 0.52–0.77) and were negatively correlated with salinity (r = −0.57– -0.78 n = 16, p < 0.05) (Fig. 5g).

Sofcom

The salinity values at SO were typical of Mediterranean marine waters, varying from 37.7 to 38.2. Thus, this site did not undergo freshwater intrusions (Fig. 3b; supplementary material 4). Tyrosine-like component C4, tryptophan-like component C5 and humic-like material C6 fell within the ranges 1.1–7.1, 1.1–4.1 and 1.3–2.4 QSU, respectively. Only C4 presented higher fluorescence intensities in spring-summer (2.8–7.1 QSU) compared to autumn winter (1.0–4.6 QSU) (U-test, p < 0.05) (Fig. 4d–f, Fig. 5i, j). C5 and C6 were correlated with each other (r = 0.63, n = 15, p < 0.05) but were not correlated with tyrosine-like component C4 (r = 0.14–0.34, n = 16, p > 0.05). The latter was significantly correlated with temperature (r = 0.56, n = 15, p < 0.05), which presented, along with particulate organic matter, higher values during the spring–summer period (Fig. 5i, j).

During the study period (February 2011–January 2012), we observed several freshwater intrusion events at Port-de-Bouc and Couronne sites, but none were detected at Sofcom. While these episodes were due to lower-salinity waters from both the Berre Lagoon and the Rhône River at Port-de-Bouc, they were due to Rhône River inputs at Couronne. Previous works have shown that after southeast winds, lower-salinity waters resulting from an eastward extension of the Rhône River plume can enter the Fos-Marseille marine area and reach the Bay of Marseille up to Sofcom station (Pairaud et al. 2011; Fraysse et al. 2014). These events were particularly observed in spring-summer, which is in line with our observations. The FDOM we determined in this study was in accordance with that investigated previously in the same area in terms of composition and fluorescence intensities (Para et al. 2010; Tedetti et al. 2010, 2012). Here, we found that freshwater intrusion events significantly influenced the seasonal pattern of FDOM at the PB and CO sites. On average, the fluorescence intensities of components C4, C5 and C6 respectively increased by 33, 48 and 32% at PB and by 33, 39 and 81% at CO under lower salinity (salinity <30 for PB and <37.7 for CO).

Protein-like components (tyrosine and tryptophan) in marine coastal waters may originate from autochthonous biological activities—either phytoplankton or prokaryotic activities—and from terrestrial organic matter and anthropogenic inputs (Determann et al. 1998; Parlanti et al. 2000; Stedmon et al. 2003; Cammack et al. 2004; Lønborg et al. 2010; Romera-Castillo et al. 2010; Yamashita et al. 2010; Tedetti et al. 2012; Stedmon and Cory 2014) (Table 2). Humic-like component C6, intermediate between humic-like components C1 and C3 in terms of λEx and λEm, may be of terrestrial origin or may be derived from the degradation of organic matter by marine heterotrophic prokaryotes (Kowalczuk et al. 2009; Lønborg et al. 2009; Jørgensen et al. 2011; Andrew et al. 2013) (Table 2). Therefore, as for inland waters, highlighting the specific origin of the FDOM components in marine waters without the support of molecular proxies is not clear (Osburn et al. 2016). However, we may observe from our results that in marine waters impacted (PB, CO) or not (SO) by the Rhône River, tyrosine-like component C4 was strongly or moderately decoupled (i.e. presented a weaker correlation) from tryptophan- and humic-like components C5 and C6. In addition, the C4 compound was rather higher in spring-summer and was positively correlated with temperature and particulate organic matter (Fig. 5e–j). This suggests that tyrosine-like fluorescence displayed a stronger link with seasonal autochthonous biological activity than components C5 and C6, which were more influenced/related to freshwater intrusion events.

Conclusions

This study highlights the composition and temporal variability of FDOM in the Rhône River delta and the Fos-Marseille marine area. Inland waters were dominated by humic-like components, whereas marine waters were dominated by protein-like fluorescence. In the Rhône delta, the fluorescence intensities of humic-like components were higher in autumn-winter, very likely due to inputs of terrestrial organic matter from rainfalls, runoffs and wind-induced sediment resuspension. In marine coastal sites, intrusions of the Berre Lagoon and the Rhône River waters had a significant impact on the local biogeochemistry, leading to higher fluorescence intensities of humic- and protein-like components in spring-summer. This work emphasizes the complex dynamics of FDOM in coastal waters and, on the basis of the work by Para et al. (2010), allowed us to establish a link between marine FDOM and Rhône River intrusions on larger spatial and temporal scales. Distinct temporal trends among FDOM components would be a reflection of the fact that FDOM components displaying similar spectral characteristics are likely composed of multiple chemically distinct DOM moieties from different sources. The use of molecular analyses in future works (lignin-derived phenols, for instance) would help better distinguish the terrestrial/autochthonous origin of these FDOM components.

Notes

Acknowledgements

We are grateful to the captain and crew of the R/V Antédon 2 for their help during the sampling. We warmly thank C. Pinazo for providing satellite maps and Chritophe Yohia (OSU Pytheas) for providing Meteorological data. We acknowledge the MOOSE program (Mediterranean Oceanic Observing System on Environment) for additional rain fall and Rhône flow data. We thank D. Lefèvre and A. Robert for the use of the Shimadzu spectrophotometer as well as the core parameter analytical platform (PAPB) of the Mediterranean Institute of Oceanography (MIO) for performing chemical analyses. Two anonymous Reviewers are acknowledged for their relevant comments and corrections, which contributed to improve the quality of this manuscript. This study is part of the ‘IBISCUS’ research project that was funded by the Agence Nationale de la Recherche (ANR)—ECOTECH program (project ANR-09-ECOT-009-01). This work also contributes to the Work Package 3 of the CNRS-INSU MISTRALS ‘MERMEX’ project.

Supplementary material

11356_2016_8255_MOESM1_ESM.docx (2.5 mb)
ESM 1(DOCX 2570 kb.)

References

  1. Akkanen J, Vogt RD, Kukkonen JVK (2004) Essential characteristics of natural dissolved organic matter affecting the sorption of hydrophobic organic contaminants. Aquat Sci 66:171–177CrossRefGoogle Scholar
  2. Andrew AA, Del Vecchio R, Subramaniam A, Blough NV (2013) Chromophoric dissolved organic matter (CDOM) in the equatorial Atlantic Ocean: optical properties and their relation to CDOM structure and source. Mar Chem 148:33–43CrossRefGoogle Scholar
  3. Banas D, Grillas P, Auby I, Lescuyer F, Coulet E, Moreteau JC, Millet B (2005) Short time scale changes in underwater irradiance in a wind-exposed lagoon (Vaccarès lagoon, France): efficiency of infrequent field measurements of water turbidity or weather data to predict irradiance in the water column. Hydrobiologia 551:3–16CrossRefGoogle Scholar
  4. Boutron O, Bertrand O, Fiandrino A, Höhener P, Sandoz A, Chérain Y, Coulet E, Chauvelon P (2015) An unstructured numerical model to study wind-driven circulation patterns in a managed coastal Mediterranean wetland: the Vaccarès lagoon system. Water 7:5986–6016CrossRefGoogle Scholar
  5. Benner R (2002) Chemical composition and reactivity. In: Hansell DA, Carlson CA (eds) Biogeochemistry of marine dissolved organic matter. Academic Press, San Diego, pp. 59–90CrossRefGoogle Scholar
  6. Benner R, Louchouarn P, Amon RMW (2005) Terrigenous dissolved organic matter in the Arctic Ocean and its transport to surface and deep waters of the North Atlantic. Glob Biogeochem Cycles 19:GB2025. doi:10.1029/2004GB002398 CrossRefGoogle Scholar
  7. Benner R, Opsahl S (2001) Molecular indicators of the sources and transformations of dissolved organic matter in the Mississippi river plume. Org Geochem 32:597–611CrossRefGoogle Scholar
  8. Bittar TB, Vieira AAH, Stubbins A, Mopper K (2015) Competition between photochemical and biological degradation of dissolved organic matter from the cyanobacteria Microcystis aeruginosa. Limnol Oceanogr 60:1172–1194CrossRefGoogle Scholar
  9. Blough NV, Del Vecchio R (2002) Chromophoric DOM in the coastal environment. In: Hansell DA, Carlson CA (eds) Biogeochemistry of marine dissolved organic matter. Academic Press, San Diego, pp. 509–546CrossRefGoogle Scholar
  10. Borisover M, Laor Y, Parparov A, Bukhanovsky N, Lado M (2009) Spatial and seasonal patterns of fluorescent organic matter in Lake Kinneret (sea of galilee) and its catchment basin. Water Res 43:3104–3116CrossRefGoogle Scholar
  11. Boyd TJ, Osburn CL (2004) Changes in CDOM fluorescence from allochthonous and autochthonous sources during tidal mixing and bacterial degradation in two coastal estuaries. Mar Chem 89:189–210CrossRefGoogle Scholar
  12. Cammack WKL, Kalff J, Prairie YT, Smith EM (2004) Fluorescent dissolved organic matter in lakes: relationships with heterotrophic metabolism. Limnol Oceanogr 49:2034–2045CrossRefGoogle Scholar
  13. Carlson CA (2002) Production and removal processes. In: Hansell DA, Carlson CA (eds) Biogeochemistry of marine dissolved organic matter. Academic Press, San Diego, pp. 91–152CrossRefGoogle Scholar
  14. Carstea EM, Baker A, Bieroza M, Reynolds D (2010) Continuous fluorescence excitation-emission matrix monitoring of river organic matter. Water Res 44:5356–5366CrossRefGoogle Scholar
  15. Cauwet G (2002) DOM in the coastal zone. In: Hansell DA, Carlson CA (eds) Biogeochemistry of marine dissolved organic matter. Academic Press, San Diego, pp. 579–609CrossRefGoogle Scholar
  16. Céa B, Lefèvre D, Chirurgien L, Raimbault P, Garcia N, Charrière B, Grégori G, Ghiglione JF, Barani A, Lafont M, Van Wambeke F (2015) An annual survey of bacterial production, respiration and ectoenzyme activity in coastal NW Mediterranean waters: temperature and resource controls. Environ Sci Pollut Res 22:13654–13668CrossRefGoogle Scholar
  17. Chari NVHK, Keerthi S, Sarma NS, Rao Pandi S, Chiranjeevulu G, Kiran R, Koduru U (2013) Fluorescence and absorption characteristics of dissolved organic matter excreted by phytoplankton species of western Bay of Bengal under axenic laboratory condition. J Exper Mar Biol Ecol 445:148–155CrossRefGoogle Scholar
  18. Chauvelon P (1998) A wetland managed for agriculture as an interface between the Rhône river and the Vaccarès lagoon (Camargue, France): transfers of water and nutrients. Hydrobiologia 373(374):181–191CrossRefGoogle Scholar
  19. Chauvelon P, Tournoud MG, Sandoz A (2003) Integrated hydrological modelling of a managed coastal Mediterranean wetland (Rhone delta, France): initial calibration. Hydrol Earth Syst Sci Discuss 7:123–132CrossRefGoogle Scholar
  20. Coble PG (1996) Characterization of marine and terrestrial DOM in seawater using excitation emission matrix spectroscopy. Mar Chem 51:325–346CrossRefGoogle Scholar
  21. Coble PG (2007) Marine optical biogeochemistry – the chemistry of ocean color. Chem Rev 107:402–418CrossRefGoogle Scholar
  22. Comoretto L, Arfib B, Chiron S (2007) Pesticides in the Rhône river delta (France): basic data for a field-based exposure assessment. Sci Tot Environ 380:124–132CrossRefGoogle Scholar
  23. Cory RM, McKnight DM (2005) Fluorescence spectroscopy reveals ubiquitous presence of oxidized and reduced quinones in dissolved organic matter. Environ Sci Technol 39:8142–8149CrossRefGoogle Scholar
  24. De Groot CJ, Golterman H (1999) Le risque d’eutrophisation de l'Etang de Vaccarès et des marais de la Camargue (Delta du Rhône, France). Ecologie 30:91–100Google Scholar
  25. Delpy F, Pagano M, Blanchot J, Carlotti F, Thibault-Botha D (2012) Man-induced hydrological changes, metazooplankton communities and invasive species in the Berre Lagoon (Mediterranean Sea, France). Mar Pollut Bull 64:1921–1932CrossRefGoogle Scholar
  26. Determann S, Lobbes JM, Reuter R, Rullkötter J (1998) Ultraviolet fluorescence excitation and emission spectroscopy of marine algae and bacteria. Mar Chem 62:137–156CrossRefGoogle Scholar
  27. Diaz SB, Morrow JH, Booth CR (2000) UV physics and optics. In: de Mora S, Demers S, Vernet M (eds) The effects of UV radiation in the marine environment. Cambridge University Press, Cambridge, pp. 35–71CrossRefGoogle Scholar
  28. Dittmar T, Stubbins A (2014) Dissolved organic matter in aquatic systems. Treatise on Geochemistry, Second edition, pp. 125–156Google Scholar
  29. Fellman JB, Hood E, Spencer RGM (2010) Fluorescence spectroscopy opens new windows into dissolved organic matter dynamics in freshwater ecosystems: a review. Limnol Oceanogr 55:2452–2462CrossRefGoogle Scholar
  30. Ferretto N, Tedetti M, Guigue C, Mounier S, Redon R, Goutx M (2014) Identification and quantification of known polycyclic aromatic hydrocarbons and pesticides in complex mixtures using fluorescence excitation-emission matrices and parallel factor analysis. Chemosphere 107:344–353CrossRefGoogle Scholar
  31. Fichot CG, Benner R (2014) The fate of terrigenous dissolved organic carbon in a river-influenced ocean margin. Glob Biogeochem Cycles 28:300–318CrossRefGoogle Scholar
  32. Fichot CG, Lohrenz SE, Benner R (2014) Pulsed, cross-shelf export of terrigenous dissolved organic carbon to the Gulf of Mexico. J Geophys Res Oceans:119. doi:10.1002/2013JC009424
  33. Fraysse M, Pairaud I, Ross ON, Faure VM, Pinazo C (2014) Intrusion of Rhone River diluted water into the Bay of Marseille: generation processes and impacts on ecosystem functioning. J Geophys Res 119:6535–6556CrossRefGoogle Scholar
  34. Guigue C, Tedetti M, Ferretto N, Garcia N, Méjanelle L, Goutx M (2014) Spatial and seasonal variabilities of dissolved hydrocarbons in surface waters from the northwestern Mediterranean Sea: results from one year intensive sampling. Sci Tot Environ 466–467:650–662CrossRefGoogle Scholar
  35. Hansell DA, Carlson CA, Repeta DJ, Schlitzer R (2009) Dissolved organic matter in the ocean. A controversy stimulates new insights Oceanography 22:52–61Google Scholar
  36. Hedges JI, Keil RG, Benner R (1997) What happens to terrestrial organic matter in the ocean? Org Geochem 27:195–212CrossRefGoogle Scholar
  37. Helms JR, Stubbins A, Ritchie JD, Minor EC, Kieber DJ, Mopper K (2008) Absorption spectral slopes and slope ratios as indicators of molecular weight, source, and photobleaching of chromophoric dissolved organic matter. Limnol Oceanogr 53:955–969CrossRefGoogle Scholar
  38. Hirose K (2007) Metal–organic matter interaction: ecological roles of ligands in oceanic DOM. Appl Geochem 22:1636–1645CrossRefGoogle Scholar
  39. Houghton RA (2007) Balancing the global carbon budget. Annu Rev Earth Planet Sci 35:313–347CrossRefGoogle Scholar
  40. Hudson N, Baker A, Reynolds D (2007) Fluorescence analysis of dissolved organic matter in natural, waste and polluted waters—a review. River Res Appl 23:631–649CrossRefGoogle Scholar
  41. Huguet A, Vacher L, Saubusse S, Etcheber H, Abril G, Relexans S, Ibalot F, Parlanti (2010) New insights into the size distribution of fluorescent dissolved organic matter in estuarine waters. Org Geochem 41:595–610CrossRefGoogle Scholar
  42. Ishii SKL, Boyer TH (2012) Behavior of reoccurring PARAFAC components in fluorescent dissolved organic matter in natural and engineered systems: a critical review. Environ Sci Technol 46:2006–2017CrossRefGoogle Scholar
  43. Jaffé R, McKnight DM, Maie N, Cory RM, McDowell WH, Campbell JL (2008) Spatial and temporal variations in DOM composition in ecosystems: the importance of long-term monitoring of optical properties. J Geophys Res 113:G04032. doi:10.1029/2008JG000683 CrossRefGoogle Scholar
  44. Jiao N, Herndl GJ, Hansell DA, Benner R, Kattner G, Wilhelm SW, Kirchman DL, Weinbauer MG, Luo T, Chen F, Azam F (2010) Microbial production ofrecalcitrant dissolved organic matter: long-term carbon storage in the global ocean. Nat Rev Microbiol 8:593–599CrossRefGoogle Scholar
  45. Joliffe IT (1986) Principal component analysis. Springer-Verlag, p. 271Google Scholar
  46. Jørgensen L, Stedmon CA, Kragh T, Markager S, Middelboe M, Søndergaard M (2011) Global trends in the fluorescence characteristics and distribution of marine dissolved organic matter. Mar Chem 126:139–148CrossRefGoogle Scholar
  47. Kowalczuk P, Durako MJ, Young H, Kahn AE, Cooper WJ, Gonsior M (2009) Characterization of dissolved organic matter fluorescence in the South Atlantic bight with use of PARAFACmodel: interannual variability. Mar Chem 113:182–196CrossRefGoogle Scholar
  48. Krachler R, Krachler RF, Wallner G, Hann S, Laux M, Cervantes Recalde MF, Jirsa F, Neubauer E, von der Kammer F, Hofmann T, Keppler BK (2015) River-derived humic substances as iron chelators in seawater. Mar Chem 174:85–93CrossRefGoogle Scholar
  49. Kuwahara VS, Nozaki S, Nakano J, Toda T, Kikuchi T, Taguchi S (2015) 18-year variability of ultraviolet radiation penetration in the mid-latitude coastal waters of the western boundary Pacific. Estuar Coast Shelf Sci 160:1–9CrossRefGoogle Scholar
  50. Lønborg C, Álvarez-Salgado XA, Davidson K, Martínez-García S, Teira E (2010) Assessing the microbial bioavailability and degradation rate constants of dissolved organic matter by fluorescence spectroscopy in the coastal upwelling system of the Ría de Vigo. Mar Chem 119:121–129CrossRefGoogle Scholar
  51. Lønborg C, Álvarez-Salgado XA, Davidson K, Miller AEJ (2009) Production of bioavailable and refractory dissolved organic matter by coastal heterotrophic microbial populations. Estuar Coast Shelf Sci 82:682–688CrossRefGoogle Scholar
  52. Luciani X, Mounier S, Paraquetti HHM, Redon R, Lucas Y, Bois A, Lacerda LD, Raynaud M, Ripert M (2008) Tracing of dissolved organic matter from the Sepetiba Bay (Brazil) by PARAFAC analysis of total luminescence matrices. Mar Environ Res 65:148–157CrossRefGoogle Scholar
  53. Ludwig W, Dumont E, Meybeck M, Heussner S (2009) River discharges of water and nutrients to the Mediterranean and Black Sea: major drivers for ecosystem changes during past and future decades? Progr Oceanogr 80:199–217CrossRefGoogle Scholar
  54. Maie N, Scully NM, Pisani O, Jaffé R (2007) Composition of a protein-like fluorophore of dissolved organic matter in coastal wetland and estuarine ecosystems. Water Res 41:563–570CrossRefGoogle Scholar
  55. Millet B, Robert C, Grillas P, Coughlan C, Banas D (2010) Numerical modelling of vertical suspended solids concentrations and irradiance in a turbid shallow system (Vaccares, Se France). Hydrobiologia 638:161–179CrossRefGoogle Scholar
  56. Mopper K, Kieber DJ (2002) Photochemistry and the cycling of carbon, Sulphur, nitrogen and phosphorus. In: Hansell DA, Carlson CA (eds) Biogeochemistry of marine dissolved organic matter. Academic Press, San Diego, pp. 455–507CrossRefGoogle Scholar
  57. Mopper K, Kieber DJ, Stubbins A (2015) Marine photochemistry: processes and impacts. In: Hansell DA, Carlson CA (eds) Biogeochemistry of marine dissolved organic matter, 2nd edn. Academic Press, Burlington, pp. 389–450CrossRefGoogle Scholar
  58. Moutin T, Raimbault P, Golterman HL, Coste B (1998) The input of nutrients by the Rhône river into the Mediterranean Sea: recent observations and comparison with earlier data. Hydrobiologia 373(374):237–246CrossRefGoogle Scholar
  59. Murphy KR, Bro R, Stedmon CA (2014) Chemometric analysis of organic matter fluorescence. In: Coble PG, Lead J, Baker A, Reynolds DM, Spencer RGM (eds) Aquatic Organic Matter Fluorescence. Cambridge University Press, New York, pp. 339–375CrossRefGoogle Scholar
  60. Murphy KR, Butler KD, Spencer RGM, Stedmon CA, Boehme JR, Aiken GR (2010) Measurement of dissolved organic matter fluorescence in aquatic environments: an interlaboratory comparison. Environ Sci Technol 44:9405–9412CrossRefGoogle Scholar
  61. Murphy KR, Stedmon CA, Waite TD, Ruiz GM (2008) Distinguishing between terrestrial and autochthonous organic matter sources in marine environments using fluorescence spectroscopy. Mar Chem 108:40–58CrossRefGoogle Scholar
  62. Nagata T (2000) Production mechanisms of dissolved organic matter. In: Kirchman DL (ed) Microbial ecology of the oceans. Wiley-Liss, New York, pp. 121–152Google Scholar
  63. Nelson NB, Siegel DA (2013) The global distribution and dynamics of chromophoric dissolved organic matter. Annu Rev Mar Sci 5:447–476CrossRefGoogle Scholar
  64. Nieto-Cid M, Álvarez-Salgado XA, Pérez FF (2006) Microbial and photochemical reactivity of fluorescent dissolved organic matter in a coastal upwelling system. Limnol Oceanogr 51:1391–1400CrossRefGoogle Scholar
  65. Ohno T (2002) Fluorescence inner-filtering correction for determining the humification index of dissolved organic matter. Environ Sci Technol 36:742–746CrossRefGoogle Scholar
  66. Ortega-Retuerta E, Frazer TK, Duarte CM, Ruiz-Halpern S, Tovar-Sanchez A, Arrieta JM, Reche I (2009) Biogeneration of chromophoric dissolved organic matter by bacteria and krill in the Southern Ocean. Limnol Oceanogr 54:1941–1950CrossRefGoogle Scholar
  67. Osburn CL, Boyd TJ, Montgomery MT, Bianchi TS, Coffin RB, Paerl HW (2016) Optical proxies for terrestrial dissolved organic matter in estuaries and coastal waters. Front Mar Sci 2:127. doi:10.3389/fmars. 2015.00127 CrossRefGoogle Scholar
  68. Pairaud IL, Gatti J, Bensoussan N, Verney R, Garreau P (2011) Hydrology and circulation in a coastal area off Marseille: validation of a nested 3D model with observations. J Mar Syst 88:20–33CrossRefGoogle Scholar
  69. Panagiotopoulos C, Sempéré R, Para J, Raimbault P, Rabouille C, Charrière B (2012) The composition and flux of particulate and dissolved carbohydrates from the Rhone River into the Mediterranean Sea. Biogeoscience 9:1827–1844CrossRefGoogle Scholar
  70. Para J, Coble PG, Charrière B, Tedetti M, Fontana C, Sempéré R (2010) Fluorescence and absorption properties of chromophoric dissolved organic matter (CDOM) in coastal surface waters of the northwestern Mediterranean Sea (bay of Marseilles, France). Biogeosciences 7:4083–4103CrossRefGoogle Scholar
  71. Parlanti E, Wo K, Geo L, Lamotte M (2000) Dissolved organic matter fluorescence spectroscopy as a tool to estimate biological activity in a coastal zone submitted to anthropogenic inputs. Org Geochem 31:1765–1781CrossRefGoogle Scholar
  72. Patel-Sorrentino N, Mounier S, Benaim JY (2002) Excitation–emission fluorescence matrix to study pH influence on organic matter fluorescence in the Amazon basin rivers. Water Res 36:2571–2581CrossRefGoogle Scholar
  73. Raimbault P, Lantoine F, Neveux J (2004) Dosage rapide de la chlorophylle a et des phéopigments a par fluorimétrie après extraction au méthanol. Comparaison avec la méthode classique d’extraction à l'acétone. Océanis 30:189–205Google Scholar
  74. Raimbault P, Pouvesle W, Sempere R (1999) Wet-oxidation and automated colorimetry for simultaneous determination of organic carbon, nitrogen and phosphorus dissolved in seawater. Mar Chem 66:161–169CrossRefGoogle Scholar
  75. Roche H, Vollaire Y, Martin E, Rouer C, Coulet E, Grillas P, Banas D (2009) Rice fields regulate organochlorine pesticides and PCBs in lagoons of the nature Reserve of Camargue. Chemosphere 75:526–533CrossRefGoogle Scholar
  76. Rochelle-Newall EJ, Fisher TR (2002) Production of chromophoric dissolved organic matter fluorescence in marine and estuarine environments: an investigation into the role of phytoplankton. Mar Chem 77:7–21CrossRefGoogle Scholar
  77. Romera-Castillo C, Sarmento H, Alvarez-Salgado XA, Gasol JM, Marrase C (2010) Production of chromophoric dissolved organic matter by marine phytoplankton. Limnol Oceanogr 55:446–454CrossRefGoogle Scholar
  78. Rostan JC, Cellot B (1995) On the use of UV spectrophotometry to assess dissolved organic carbon origin variations in the upper Rh6ne river. Aquatic Sci 57:70–80CrossRefGoogle Scholar
  79. Sempéré R, Charrière B, Van Wambeke F, Cauwet G (2000) Carbon inputs of the Rhone River to the Mediterranean Sea: biogeochemical implications. Glob Biogeochem Cy 14:669–681CrossRefGoogle Scholar
  80. Siegel DA, Maritorena S, Nelson NB, Behrenfeld MJ (2005) Independence and interdependencies of global ocean color properties: reassessing the bio-optical assumption. J Geophys Res 110:C07011. doi:10.1029/2004JC002527 CrossRefGoogle Scholar
  81. Siegenthaler U, Sarmiento J (1993) Atmospheric carbon dioxide and the ocean. Nature 365:119–125CrossRefGoogle Scholar
  82. Stedmon CA, Bro R (2008) Characterizing dissolved organic matter fluorescence with parallel factor analysis: a tutorial. Limnol Oceanogr Meth 6:572–579CrossRefGoogle Scholar
  83. Stedmon CA, Cory RM (2014) Biological origins and fate of fluorescent dissolved organic matter in aquatic environments. In: Coble PG, Lead J, Baker A, Reynolds DM, Spencer RGM (eds) Aquatic Organic Matter Fluorescence. Cambridge University Press, New York, pp. 278–299CrossRefGoogle Scholar
  84. Stedmon CA, Markager S (2005a) Tracing the production and degradation of autochthonous fractions of dissolved organic matter by fluorescence analysis. Limnol Oceanogr 50:686–697CrossRefGoogle Scholar
  85. Stedmon CA, Markager S (2005b) Resolving the variability of dissolved organic matter fluorescence in a temperate estuary and its catchment using PARAFAC analysis. Limnol Oceanogr 50:686–697CrossRefGoogle Scholar
  86. Stedmon CA, Markager S, Bro R (2003) Tracing dissolved organic matter in aquatic environments using a new approach to fluorescence spectroscopy. Mar Chem 82:239–254CrossRefGoogle Scholar
  87. Steinberg DK, Nelson N, Carlson C, Prusak AC (2004) Production of chromophoric dissolved organic matter (CDOM) in the open ocean by zooplankton and the colonial cyanobacterium Trichodesmium spp. Mar Ecol Prog Ser 267:45–56CrossRefGoogle Scholar
  88. Tedetti M, Guigue C, Goutx M (2010) Utilization of a submersible UV fluorometer for monitoring anthropogenic inputs in the Mediterranean coastal waters. Mar Pollut Bull 60:350–362CrossRefGoogle Scholar
  89. Tedetti M, Longhitano R, Garcia N, Guigue C, Ferretto N, Goutx M (2012) Fluorescence properties of dissolved organic matter in coastal Mediterranean waters influenced by a municipal sewage effluent (Bay of Marseilles, France). Environ Chem 9:438–449CrossRefGoogle Scholar
  90. Tedetti M, Sempéré R (2006) Penetration of ultraviolet radiation in the marine environment. A review. Photochem Photobiol 82:389–397CrossRefGoogle Scholar
  91. The MerMex group (2011) Marine ecosystems’ responses to climatic and anthropogenic forcings in the Mediterranean. Prog Oceanogr 91:97–166CrossRefGoogle Scholar
  92. Tréguer P, LeCorre P (1975) Manuel d’analyses des sels nutritifs dans l’eau de mer: Utilisation de l’Autoanalyser II Technicon, 2nd edn. Université de Bretagne Occidentale, BrestGoogle Scholar
  93. Ulses C, Grenz C, Marsaleix P, Schaaff E, Estournel C, Meulé S, Pinazo C (2005) Circulation in a semi-enclosed bay under influence of strong freshwater input. J Mar Syst 56:113–132CrossRefGoogle Scholar
  94. Vaquer A, Heurteaux P (1989) Modifications récentes de la végétation aquatique de l’étang du Vaccarès (Camargue, France) liées aux perturbations anthropiques. Ann Limnol 25:25–38CrossRefGoogle Scholar
  95. Weishaar JL, Aiken GR, Bergamaschi BA, Fram MS, Fujii R, Mopper K (2003) Evaluation of specific ultraviolet absorbance as an indicator of the chemical composition and reactivity of dissolved organic carbon. Environ Sci Technol 37:4702–4708CrossRefGoogle Scholar
  96. Welschmeyer NA (1994) Fluorometric analysis of chlorophyll a in the presence of chlorophyll b and pheopigments. Limnol Ocean 39:1985–1992CrossRefGoogle Scholar
  97. Yamashita Y, Cory RM, Nishioka J, Kuma K, Tanoue E, Jaffé R (2010) Fluorescence characteristics of dissolved organic matter in the deep waters of the Okhotsk Sea and the northwestern North Pacific Ocean. Deep-Sea Res Pt II 57:1478–1485CrossRefGoogle Scholar
  98. Yamashita Y, Panton A, Mahaffey C, Jaffé R (2011) Assessing the spatial and temporal variability of dissolved organic matter in Liverpool Bay using excitation–emission matrix fluorescence and parallel factor analysis. Ocean Dyn 61:569–579CrossRefGoogle Scholar
  99. Yamashita Y, Tanoue E (2004) In situ production of chromophoric dissolved organic matter in coastal environments. Geophys Res Lett 31:1–4. doi:10.1029/2004GL019734 CrossRefGoogle Scholar
  100. Yao X, Zhang Y, Zhu G, Qin B, Feng L, Cai L, Gao G (2011) Resolving the variability of CDOM fluorescence to differentiate the sources and fate of DOM in Lake Taihu and its tributaries. Chemosphere 82:145–155CrossRefGoogle Scholar
  101. Zhang Y, Zhang E, Yin Y, Dijk MA, Van Feng L, Shi Z (2010) Characteristics and sources of chromophoric dissolved organic matter in lakes of the Yungui plateau, China, differing in trophic state and altitude. Limnol Oceanogr 55:2645–2659CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2016

Authors and Affiliations

  • Nicolas Ferretto
    • 1
  • Marc Tedetti
    • 1
  • Catherine Guigue
    • 1
  • Stéphane Mounier
    • 2
  • Patrick Raimbault
    • 1
  • Madeleine Goutx
    • 1
  1. 1.Aix Marseille Université, CNRS/INSU, Université de Toulon, IRD, Mediterranean Institute of Oceanography (MIO)MarseilleFrance
  2. 2.Laboratoire PROTEEUniversité de ToulonLa Garde CedexFrance

Personalised recommendations