Organic matter distribution, composition and its possible fate in the Chilean North-Patagonian estuarine system

Organic matter distribution, composition and its possible fate in the Chilean North-Patagonian estuarine system

Science of the Total Environment 657 (2019) 1419–1431 Contents lists available at ScienceDirect Science of the Total Environment journal homepage: w...

NAN Sizes 0 Downloads 0 Views

Science of the Total Environment 657 (2019) 1419–1431

Contents lists available at ScienceDirect

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

Organic matter distribution, composition and its possible fate in the Chilean North-Patagonian estuarine system Humberto E. González a,b,⁎, Jorge Nimptsch a, Ricardo Giesecke a,b, Nelson Silva c a b c

Universidad Austral de Chile, Instituto de Ciencias Marinas y Limnológicas, Valdivia, Chile Centro FONDAP de Investigación de Ecosistemas Marinos de Altas Latitudes (IDEAL), Universidad Austral de Chile, Valdivia, Chile Escuela de Ciencias del Mar, Pontificia Universidad Católica de Valparaíso, Valparaíso, Chile

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• The distribution, composition and transport of both DOC and POC, were studied across a terrestrial-marine transition system. • A fDOM land-ocean gradient from humic-C1 terrigenous-derived to tyrosine-like C3 autochthonous fDOM was observed. • The UVC to UVA humic ratio was correlated with salinity, highlighting the high variability in fDOM chemical characteristics. • Climate warming and anthropogenic practices boost the mobilization of terrestrial carbon pools.

a r t i c l e

i n f o

Article history: Received 25 June 2018 Received in revised form 26 November 2018 Accepted 29 November 2018 Available online 04 December 2018 Editor: Daniel Wunderlin Keywords: North-Patagonia Fjord DOC POC Fluorescent DOM Organic matter

a b s t r a c t The distribution, composition, and transport of both dissolved and particulate organic carbon (DOC and POC) were studied across a terrestrial - marine transition system in the Chilean North-Patagonia (41°S). At the land-fjord boundary we reported: (i) high concentrations of both silicic acid (up to 100 μM) and integrated chlorophyll a (62 mg m−2), (ii) dominance of nanophytoplankton (63%), humic-, terrigenous-derived, and protein-like DOC (19 and 36%, respectively), and (iii) a shallow photic zone (12 m depth). In contrast, the estuarine-ocean boundary was characterized by (i) high concentrations of nitrate and phosphate (20 and 2 μM respectively) and low chlorophyll a concentration (11 mg m−2), (ii) dominance of microphytoplankton (59%) and tyrosine-like C3 autochthonous DOC (34%), and (iii) a deep photic zone (29 m depth). Allochthonous DOC input at the fjord head and the ocean accounted for 60% and 10% of total DOC, respectively. The input of humic-like substances was enhanced by intense forestry and agriculture activity around the Puelo River watershed, contributing from 50% to 14% of total DOC along the fjord - ocean transect. In contrast, autochthonous tyrosine-like substances increased from 25% to 41% of total DOC, highlighting the role of bacterial metabolism in regulating DOM composition. The high correlation (R2 = 0.7) between the UVC-humic:UVA-humic ratio and salinity suggest that processes associated to freshwater input impinged on the DOC chemical characteristics and origins. Overall, our observations support the view that climate warming (freshwater input) and

⁎ Corresponding author at: Universidad Austral de Chile, Instituto de Ciencias Marinas y Limnológicas, Valdivia, Chile. E-mail address: [email protected] (H.E. González).

https://doi.org/10.1016/j.scitotenv.2018.11.445 0048-9697/© 2018 Elsevier B.V. All rights reserved.

1420

H.E. González et al. / Science of the Total Environment 657 (2019) 1419–1431

anthropogenic practices (aquaculture) boost the mobilization of terrestrial carbon pools and their intrusion into coastal ocean areas, a process that should be given more attention in climate prediction models. © 2018 Elsevier B.V. All rights reserved.

1. Introduction Estuarine systems are extremely complex, with high spatial and temporal variability. The physical and chemical characteristics of total organic matter (TOM), either dissolved (DOM) or particulate (POM), depend on the interplay among multi-factorial processes of production, utilization, reactivity, and export of organic matter along a continuous land-ocean gradient. For example, in aquatic systems DOM affects light attenuation, metal speciation and bioavailability (possible nutrient sources) for the microbial community (Cory and McKnight, 2005). Estuarine productivity is supported by total allochthonous organic matter (OMal) from various sources but is most likely mainly composed of plant litter, given that the northern Patagonian systems are surrounded by forests and agricultural land which most likely contribute a significant amount of refractory TOM (Raymond and Bauer, 2001). In estuarine systems, the upper freshwater plume seems to be mainly loaded with POMal (Vargas et al., 2011), while the underlying saline layer contains autochthonous organic matter (POMau) which most likely originates from microbial degradation and phytoplankton exudates (Canuel et al., 1995). Thus, in a terrestrial to marine gradient OMal and OMau might predominate at the heads (Silva and Prego, 2002) and mouths (Singh et al., 2010; Martineau et al., 2004) of the fjords, respectively. It is possible that the relative availability and utilization of OMal versus OMau by bacterioplankton would follow the same spatial gradient, enabling microbial communities to exploit energy that escapes from upstream ecosystems. Furthermore, a fraction of the OMal utilized by microorganisms can be mobilized to invertebrates and vertebrates through the pelagic and benthic food webs (Lusseau and Wing, 2006; McLeod and Wing, 2007; McLeod and Wing, 2009; Vargas et al., 2011). This organic matter flow from lower to higher trophic levels increases the overall energy utilization at the ecosystem level (Battin et al., 2008) and accounts for up to one-fifth of estuarine metabolism (Smith and Hollibaugh, 1993). Estuarine systems are usually in a heterotrophic state i.e. net primary production is lower than community respiration (PPn b CR) (Dale and Prego, 2005; Karlsson et al., 2007; Gupta et al., 2009), suggesting that they might also rely on OMal. The traditional view of the selfsustainability of aquatic systems has therefore been challenged and the input coupled to the important contribution of OMal to benthic and pelagic trophic webs has now been generally recognized (Pantoja et al., 2010), including implications for the metabolic balance of plankton communities. This terrestrial input of OMal may include a mixture of labile, semi-labile, and refractory matter, where components of the watershed, such as freshwater discharge, tidal currents, vegetation, river morphology, water residence time, and the chemical and physical characteristics of TOM itself play a major role in making DOM available (Kaiser et al., 2017). Chemical characteristics and concentration of OM can modify its flocculation properties and subsequent export to the bottom sediments. During diatom blooms the aggregation capability of phytodetritus increases as a result of the exudation of mucopolysaccharides that enhance the stickiness and flocculation of TOM. The input of total suspended sediments in Patagonian riverine systems that discharge into fjords and channels is very high, and large rivers such as the Baker River contribute ~0.10 g L−1 (González et al., 2013). As maximum floc size occurred at total suspended sediment concentrations of N0.10 g L−1 (Verney et al., 2009); a significant part of these sediments and its associated OM may be incorporated into local sediments. The vertical flux of POM presents a high seasonal variability in Reloncaví Fjord (RF) (41°38′S), increasing from 334 (winter) to 725 (spring) mgC m−2 d−1 (González et al., 2010). In Baker Fjord (47°56′

S) POC flux is slightly lower (234 mgC m−2 d−1), albeit representing 65% of net primary production (González et al., 2013). The study area includes a strong gradient in both freshwater input and its TOM load from the terrestrial forest watershed to oceanic waters, starting from RF into which two major rivers (Petrohué and Puelo) discharge freshwater at a combined rate of 500 to 1000 m3 s−1, with a minimum in February and a maximum in July. However, a decreasing trend in the terrestrial signal associated with reduced precipitation and river discharge has been reported over the last 400 years in northern Patagonia (Lara et al., 2008; Rebolledo et al., 2015). When and where rivers behave as active reactors versus passive pipes for DOM stands as a major knowledge gap in land-ocean continuum biogeochemistry (Casas-Ruiz et al., 2017). In a previous study (González et al., 2010) we demonstrated that freshwater discharge, nutrient availability, and solar radiation strongly affect the structure and functioning of the pelagic community of the RF and Inner Sea of Chiloé (ISCh). Said analysis lacked the DOC components (concentration, composition, and either autochthonous or allochthonous origin), and this contribution aimed to fill that knowledge gap regarding the carbon pools in the northern Chilean Patagonia fjords. We aimed to evaluate the spatial input of autochthonous and allochthonous OM and their biochemical composition, accounting for the connectivity across ecosystems (terrestrial-estuarine-marine) and the interactions within compartments (pelagic-benthic). The main objectives were to establish, in a transect from the land to the ocean, (i) the physical and chemical characteristics of the water (temperature, salinity, dissolved oxygen, and macronutrients), as well as UV and PAR light penetration into the water column, (ii) chlorophyll-a sizefractions, bacterioplankton abundance, and microplankton distribution and composition, and (iii) concentrations of DOM - POM and the origin of the OM (OMal versus OMau). The analysis was complemented by identifying a specific fluorescent dissolved components and a stable isotopic signal, both measured along a transect from the Puelo and Petrohué river mouths in RF to the coastal marine system of the ISCh. 1.1. Materials and methods 1.1.1. Study area and physico-chemical conditions Sampling took place from July 1 to 20, 2011, aboard the Chilean Navy oceanographic vessel AGS 61 “Cabo de Hornos”. The study area included RF (41°38′S), an east-west orientation with a total length and width of 55 and 2.8 km, respectively, and the ISCh (from 42 to 44°S). A transect with 19 sampling stations traversed from the head of RF (Stn 7) to the coastal oceanic area of the Guafo Entrance (Stn 50) (Fig. 1). Water samples were collected at these stations with a bottle-rosette system with the purpose of determining DOC and POC, 13C stable isotopes, sizefractionated chl a concentration, bacterioplankton abundance and biomass (at 1, 5, 10, and 25 m). For nutrient analyses (nitrate, orthophosphate, and silicic acid), 50 mL water samples were collected at selected depths (1, 5, 10, 25, 50, 75, 100, 150, 200, 250, 300, 350 and 400 m), stored at −20 °C in acid-cleaned high-density polyethylene bottles, and analyzed with a nutrient autoanalyzer (Technicon) according to the method proposed by Atlas et al. (1971). The physical structure of the water column (temperature, salinity, dissolved oxygen) was recorded with a Seabird 19 CTD\\O. Salinity and dissolved oxygen sensors were calibrated by measuring salinity (Autosal salinometer) and dissolved oxygen (Winkler method) in discrete water samples. Vertical profiles of chl a concentrations were recorded with a calibrated submersible fluorometer (FluoroProbe BBE, Moldeanke).

H.E. González et al. / Science of the Total Environment 657 (2019) 1419–1431

1421

Fig. 1. Map of the study area showing the oceanographic and processes stations.

1.1.2. Chlorophyll-a (chl a) For chl a determination, 200 mL of seawater were filtered (GF/F Whatman glass fiber filters, 0.7 μm nominal pore size) in triplicate and the loaded filters were immediately frozen (−20 °C) in darkness until later analysis by fluorometry (Turner Design TD-700), using acetone: water (90:10% v/v) for pigment extraction according to standard procedures (Parsons et al., 1984). Chl a size fractionation was carried out in 3

sequential steps: (1) for the nanoplankton fraction (2–20 μm), seawater (125 mL) was pre-screened (20 μm Nitex mesh sieve) and the particles were retained on a 2 μm Nuclepore filter with a low vacuum pressure (b10 mm Hg) to avoid rupturing phytoplankton cells (Kemp et al., 1993); (2) for the picoplankton fraction (0.7–2.0 μm), seawater (125 ml) was pre-filtered (2.0 μm Nuclepore filter) under vacuum and particles were retained on a 0.7 μm GF/F glass fiber filter; and (3) for

1422

H.E. González et al. / Science of the Total Environment 657 (2019) 1419–1431

the whole phytoplankton community, seawater (125 ml) was filtered under vacuum through a 0.7 μm Whatman glass fiber filter. The amount of chl a in the microphytoplankton fraction was obtained by subtracting the chl a values of the nano- and pico-plankton fractions from that of the whole plankton fraction. 1.1.3. Radiation spectrum Radiation quality and quantity, distributed in Photosynthetic Active Radiation (PAR; 400–700 nm) and UV-A (320–395 nm), were measured with a radiometer (Biospherical Mod. PUV 2500). This information was analyzed together with the phytoplankton distribution within the upper 35 m of the water column. The photic zone was defined as the water column stratum from 100% (surface) down to 1% of the surface PAR, UV-A. 1.1.4. Bacterioplankton abundance and biomass For bacterioplankton abundance (bacteria and archaea; in cells ml−1), 25 ml samples of seawater were preserved with glutaraldehyde (final concentration 2% v/v). From each of these samples, 2 ml of seawater were filtered under low vacuum pressure (b10 mmHg) to avoid rupturing the bacterial cells (García-Martín et al., 2018), on polycarbonate membrane filters (Nuclepore 0.2 μm), stained with the fluorochrome DAPI (Porter and Feig, 1980) and counted by epifluorescence microscopy (Zeiss Axiovert 200, 1000× magnification). Bacterial biomass was estimated using a conversion factor of 20 fg C cell−1 (Lee and Fuhrman, 1987). 1.1.5. POC, DOC and stable isotopes Organic matter, both dissolved and particulate (DOC and POC), was estimated at 19 oceanographic stations along a transect from the RF head (limnetic side) to the Guafo Entrance (oceanic side). At each station, water samples were collected from 4-four depths (0, 5, 10 and 25 m). For POC concentrations, including δ13C, the samples (0.5 to 1.0 L) were filtered under gentle vacuum through GF/F filters (pre combusted for 4 h at 450 °C) and stored in liquid nitrogen until later analysis following standard procedures (von Bodungen et al., 1991). All POC samples were run twice on an OI Analytical TIC-TOC Analyzer (model 1030). The first run was to determine the organic carbon concentration, and the second the δ13C isotope value. The TIC-TOC analyzer was interfaced with a Finnegan Mat Delta Plus isotope ratio mass spectrometer for continuous flow analysis. The analytical precision is 4% of the concentration (in ppm) for the quantitative, and ±0.4‰ for the isotopes. The instrument detection limit was 2 ppb C and data was normalized using two different internal organic standards (for further information see: http://thietbihiepphat.com/upload/files/1030_ manualW.pdf). Allochthonous POC (POCal) was estimated with a 2-source-mixing model (Bianchi, 2007): %POCal ¼ ðδ13 Csample −δ13 Cmarine Þ=ðδ13 Cterrestrial −δ13 Cmarine Þ 13

ð1Þ

where\ δ Csample\ is\ the\ isotopic\ composition\ of\ the\ sample,\ δ13 Cmarine\ is\ the\ marine\ end-member\ value\ from\ the\ most\ oceanic\ station,\ and\ δ13Cterrestrial\ is\ the\ riverine\ end-member\ value\ for\ POC. For total DOC concentration, each water sample was passed through a sterile Millex-GP syringe filter (0.22 μm) previously rinsed with ultrapure water and then sample water, and stored in a sterile 40 mL I-Chem 200 vial. The filtered sample was then acidified (pH ~2) with 400 μL of ultra-pure, concentrated HCl, to inhibit biological activity. The samples were refrigerated and analyzed by the G. G. Hatch stable isotope laboratory (Ottawa, Canada) (www.isotope.uottawa.ca/techniques/water. html) for both DOC concentration and stable isotope signal. The carbon concentration was measured in a TOC OI Analytical analyzer (Aurora Model 1030W) equipped with an automated sampler 1088 and a combustion system. In addition, the isotopic determination was estimated with a continuous-flow mass spectrometer. The isotopic data were

standardized with three organic standards, with a final precision of ± 0.2‰. To determine the original isotopic signal from the various DOC sources, water samples from the Petrohué, Cochamó, and Puelo rivers (DOCal end member with a δ13C of −29) and from the deep Guafo Entrance (DOCau end member with a δ13C of −19) were analyzed using the above protocol. 1.1.6. fDOM characterization fDOM was used as a proxy to determine the origin of the DOM, either allochthonous (humic and fulvic acids, Paczkowska et al., 2017) or autochthonous (protein-like compounds); for this, water samples were collected from four depths (0, 5, 10, and 25 m) at all stations. The samples were filtered (sterile PES 0.22 μm filters) and frozen for later DOM analysis. The optical properties of the filtered water samples were determined with a Pharo 300 Spectroquant photometer (Merck, Germany) and all fluorescence measurements were made with a Perkin-Elmer LS-50b fluorescence spectrometer. Measurements were taken at excitation wavelengths from 240 to 450 nm (5 nm steps) and emission wavelengths from 300 to 600 nm (2 nm steps) with a slit width of 5 nm to generate excitation-emission matrices (EEM, Fig. 1 Supplemented material) (Stedmon and Markeger, 2005a, 2005b). All samples were measured at room temperature and corrected to the absorbance spectra (800 to 190 nm) for the instrument baseline set (Green and Blough, 1994). For the inner-filter correction of the fluorescence measurements, absorbance in 1 cm cuvettes was recorded. Daily measurements of the area under the Raman peak for MilliQ water indicated instrument stability (Lawaetz and Stedmon, 2009). Excitation correction was achieved by using the correction provided by the manufacturer and normalized by the area under the Raman peak at 350 nm excitation wavelength (Lawaetz and Stedmon, 2009). These corrections were done with the FDOMcorr toolbox (Murphy et al., 2010) for Matlab (version R2012a, MathWorks, Ismaning, Germany) (Lawaetz and Stedmon, 2009; Murphy et al., 2010). In order to assess the composition of DOM by means of EEMs, a Parallel Factor Analysis Model (PARAFAC) was generated with the DOM Fluor Toolbox (version 1.7; Stedmon and Bro, 2008) for Matlab, using all measurements from the sampling track. Four PARAFAC components (i.e. fluorophores C1 to C4) were identified (Table 1; Supplemental material), split-half validated, and the best model fit was established by random initialization (Stedmon and Bro, 2008). DOM fluorescence of the PARAFAC components was expressed in percentage contributions to the peak fluorescence of each component to the sum of the peak fluorescence of all components. According to Cawley et al. (2012), many EEM fluorophores resolved by parallel factor analysis with different numbers of fluorescent components across water bodies of different biomes and different water sources have consistent spectral characteristics. This constancy permits a consistent and meaningful comparison of new DOM PARAFAC models with those previously developed. Therefore, the identity of the potential DOM fractions was assigned for each fluorescent component and compared with components reported from developed and published DOM-PARAFAC models. Since DOM matrices are complex mixtures of organic substances (e.g. lignins, lipids, amino acids), fluorescence components are good proxies and indicators of dynamic changes, allowing the tracking of microbial and photochemical transformations of DOM in situ (Cory and McKnight, 2005; Hall et al., 2005; Stedmon and Markager, 2005a, 2005b). Furthermore, fluorescent components provide qualitative and semiquantitative information on the source (i.e. DOMal, DOMau), biochemical composition (i.e. humic-, protein-like), and biogeochemical role of DOM in marine systems (Fellman et al., 2010). 2. Results 2.1. Study Area and physical and chemical conditions The freshwater coming from the continental adjacent rivers (Petrohué, Cochamó, and Puelo rivers) drains into the RF and from there into Reloncaví Sound (RS) and into the ISCh where it mixes with

H.E. González et al. / Science of the Total Environment 657 (2019) 1419–1431

1423

Table 1 Percentage of allochthonous particulate organic carbon (%POC alloch) along the transect sampled from the Reloncaví Fjord (RF) to the Reloncaví Sound (RS), Inner Sea of Chiloé (ISCh) down to the Corcovado Gulf (CG) and Guafo entrance (GE). Riverine/lake δ13C end-member (−27,35), marine end-member (δ13C −17,12), used as a reference for the Chilean Patagonia (Lafon et al., 2014; González et al., 2016), n = 2. Sector

Station

RF

7 6 5 4 Average (±st. dev.) 3 8 9 Average (±st. dev.) 14 16 20 21 32 33 36 38 Average (±st. dev.) 44 47 49 50 Average (±st. dev.)

RS

ISCh

CG + GE

δ13C

POCalloch (%)

Surface

10 m

Surface

10 m

−24.30 ± 0.1 −24.24 ± 0.2 −22.53 ± 0.7 −22.88 ± 0.2

−22.39 ± 0.2 −24.30 ± 0.2 −23.13 ± 0.6 −23.30 ± 0.4

--22.59 ± 0.2 −21.04 ± 0.1 --23.64 ± 0.2

−23.14 ± 0.1 −23.90 ± 0.1 −23.28 ± 0.2

−21.54 ± 0.2 −21.26 ± 0.1 −21.66 ± 0.1 −23.56 ± 0.3 −23.52 ± 0.2 −20.81 ± 0.1 −23.68 ± 0.1 −22.06 ± 0.1

−20.30 ± 0.1 −21.37 ± 0.1 −22.42 ± 0.3 −22.97 ± 0.2 −23.64 ± 0.1 −22.42 ± 0.3 −23.62 ± 0.1 −21.85 ± 0.1

−21.56 ± 0.1 −20.39 ± 0.1 −19.23 ± 0.1 −20.06 ± 0.1

−21.92 ± 0.2 −20.06 ± 0.1 −19.38 ± 0.2 −20.45 ± 0.1

70,2 69,6 52,9 56,3 62,3 ± 8,9 53,5 38,3 63,7 51,8 ± 12,8 43,2 40,5 44,4 63,0 62,6 36,1 64,1 48,3 50,3 ± 11,3 46,3 32,0 20,6 28,7 31,9 ± 10,7

41,7 70,2 58,7 60,4 57,8 ± 11,8 58,8 66,3 60,2 61,8 ± 4,0 31,1 41,5 51,8 57,2 63,7 51,8 63,5 46,2 50,9 ± 11,1 46,9 28,7 22,1 32,6 32,6 ± 10,5

Sub-Antarctic Waters (SAAW). The main input of oceanic SAAW into the ISCh occurred through the Guafo Entrance (GE) in the Corcovado Gulf (CG) (Fig. 1). A well-defined salinity-dependent pycnocline was observed close to the surface (upper 10 m of the water column) in RF that deepens at Reloncaví Sound and at Ancud Gulf (N10 m depth). It nearly disappears in CG where the water column is quasihomogeneous (Fig. 2). The temperature in the upper brackish layer gradually decreased from 12 °C near the surface (Stns 7 to 16) to 10.5 °C at 25 m depth within the ISCh. This contrasted with a strong salinity front along the entire transect from salinity b 5 (fjord head) to 33 (GE). The dissolved oxygen concentration decreased from the freshwater plume in RF (8–9 mL L−1), to the deeper waters (down to 400 m) of the ISCh (4–5 mL L−1) (Fig. 2). The strong stratification in the upper estuary (RF, RS, and AG) also limits mixing and entrainment of nutrients from subsurface waters. In CG and the GE the stratification almost disappeared. Phosphate and nitrate surface concentrations at the head of RF (Stns 7 and 6) were below the autoanalyzer detection limit and increased to ca. 2 and 20 μM, respectively, below 50 m in the ISCh. Conversely, the silicic acid concentration was very high (100 μM) at the head of the fjord, decreasing to 12 μM in the upper 100 m of the ISCh (Fig. 2).

2.2. Chl a The highest chl a concentration (up to 7 μg L−1) were found along a subsurface plume (2–7 m depth) that extended from RF into Ancud Gulf. On entering the ISCh, this plume extended to 10–20 m depth and the chl a concentration decreased to 1.5 μg L−1 towards the Desertores Islands. Further south a slight increment in chl a (b4 μg L−1) was recorded within the upper 30 m layer, although concentrations did not exceed 2 μg L−1 close to the Guafo Entrance (Fig. 4). Overall, higher chl a concentrations (integrated through the water column) were found inside RF, with a predominance of the nano-phytoplankton size fraction. For example, at Stn 7 the integrated concentration within the upper 25 m was 62 mg m−2 chl a, distributed mainly between 5 and 10 m depth and with the pico-, nano-, and macro-phytoplankton size fractions contributing 1, 63 and 36% of total chl a respectively. A lower Chl a concentration (11 mg m−2) was found at the coastal oceanic

station (Stn 50), and the size-fraction was dominated by microphytoplankton (59%) (Fig. 3A). 2.3. POC, DOC, and stable isotopes POC values along the transect were slightly higher at the surface (mean ± SD 256 ± 132 mg m−3) than at 10 m depth (176 ± 73 mg m−3) (Fig. 3B). The δ13C for the particulate organic carbon (POC) fluctuated from −24.30‰ at station close to the Reloncaví fjord head and −19.23‰ close to the Guafo Entrance (Table 1). These values, in terms of allochthonous POC (%POC alloch), contributed from 62.3% (at surface waters near the RF head) to 31.9% at the Guafo Entrance (Table 1). Along the surveyed transect, significant differences in the spatial distribution of the different parameters (proxies) of allochthonous versus autochthonous DOC/POC were found between the Reloncaví Fjord and the coastal marine waters close to the Corcovado Gulf. These holds true for the C1 and C3 PARAFAC fluorophores (ANOVA F = 8.52, p b 0.002 and F = 6.33, p b 0.006, respectively); the δ13C isotopic signature (F = 5.82, p b 0.008), and the percentage of allochthonous POC in surface waters (F = 5.3, p b 0.011) along the land-ocean transect. The DOC concentration did not show significant differences between means for the surface and at 10 m depth (900 ± 100 mg m−3), nor between the upper 10 m in RF and the Corcovado Gulf (2000 ± 100 mg m−3) (Fig. 3C). The in situ continuous measurements through the water column along the entire transect (Fig. 5 upper panel) showed a mosaic of water parcels with DOC concentrations ranging between b0.7 and 1.27 mg L−1 (average 0.91 mg L−1). The Reloncaví Fjord and Ancud Sound showed the highest DOC concentrations (N1 mg L−1), mainly of terrestrial origin in the former and a mix of terrestrial and marine in the latter. The three rivers in the study area (Puelo, Cochamó, and Petrohué) discharge freshwater loaded with DOCal into RF, contributing ca. 50% of the total DOC at the head of the fjord and b 10% at the Guafo Entrance (Fig. 5 lower panel). Throughout the transect, except at depths above 10 m along RF, the DOC was mainly of marine origin. The carbon isotope fingerprint of the water indicated an average value of −29.14 ± 0.41‰ for δ13C in the Puelo, Petrohué, and Cochamó rivers and that this signal weakened rapidly in water entering the fjord-marine system: −22.41‰ at the head of RF (Stns 7–5), −21.64 ± 0.28‰ at Reloncaví Sound, −21.00 ± 0.53‰ at the Corcovado Gulf and − 20.47 ± 0.81‰

1424

H.E. González et al. / Science of the Total Environment 657 (2019) 1419–1431 Guafo entrance

Desertores I. Ancud Gulf

Corcovado Gulf

Reloncaví S. Reloncaví F.

Guafo entrance

( ) (m) Depth

0

50

49

47

44

38

36 33

32

21

Stations 20

16 14 9

8 3 4

5 6 C7 B7

50 0

50

50

100

100

200

200

49

400 400

( ) (m) Depth

350 49

300 44

250

200

38

36 33

32

150

100

20

16 14 9

21

50 8 3 4

0 9

5 6C7 B7

400 400 50 0

50

100

100

200

200

300

36 33

32

21

20

16 14 9

0.6

8 3 4

5 6 C7 B7

350 49

47

300 44

250 38

36 33

200 32

350 49

47

300 44

250 38

200 36 33

32

150 21

20

100 16 14 9

50 8 3 4

0 5 6 C7 B7

Phosphate (µM) 150 21

20

100 16 14 9

0

400 400

5 6 C7 B7

50 0

50 8 3 4

50

50

100

100

200

200

350 49

47

300 44

250 38

200 36

33

32

21

150

100

20

16 14 9

50 8 3 4

0 5 6 C7 B7

300

300

Silicic acid (µM)

Nitrate (µM) 400

38

300

Dissolved Oxygen (mL L-1)

400 400

Depth (m)

47

50

400

44

Salinity

Temperature (ºC)

50 0

47

300

300

50 0

Reloncaví S. Reloncaví F.

Desertores I. Ancud Gulf

Corcovado Gulf

Stations

350

300

250

200

150

100

50

0

400 400

350

300

250

200

150

100

50

0

Distance (km)

Distance (km)

Fig. 2. Vertical distribution of temperature (°C), salinity, dissolved oxygen (ml l−1), nitrate, phosphate and silicic acid (μM) along the Reloncaví Fjord - Guafo Entrance transect. Note that the y-axis scale changes at 100 m.

at the Guafo Entrance. We gave a broad description of the terrestrial signal along the land-ocean continuum because the application of carbon stable isotope analysis of DOC from natural seawater has been limited owing to the inherent difficulty of said analysis (Barber et al., 2017). The increasing trend observed in the DOC:POC ratio from RF (2.61) towards the Corcovado Gulf (6.02) was due mainly to a decrease in the POC concentration from the inner fjord (terrestrial side) to the coastal ocean (marine side) (Fig. 3B and C). Bacterioplankton abundance and biomass did not show a significant difference between the surface (800,000 ± 400,000 cells mL−1 and 16 ± 8 mgC m−3) and 10 m depth (700,000 ± 300,000 cells mL−1 and 14 ± 6 mgC m−3) along the entire transect (Fig. 3D). Only the southern Reloncaví Sound and Ancud Gulf areas showed a slight increase in abundance and biomass (1,187,648 cells mL −1 and 24 mgC m−3) (Fig. 3D). Along the transect, a continuous southward deepening of the photic zone (PAR penetration) from 12 m depth at RF (Stn 7) to a maximum of 29 m at the GE (Stns 49–50) was recorded. Overall, a direct relationship between chl a and the depth of the photic zone was observed, with a sharper reduction in PAR penetration along the Reloncaví Fjord and Sound than in the ISCh. In line with this, a high chl a concentration (4 to 8 μg L−1) in the RF was observed, which deepened towards the ISCh. The penetration of the different radiation wavelengths showed an expected pattern with a sharp attenuation of the most energetic

radiation (320 nm) within the upper 10 m of the water column, followed by UVA (340, 380–395 nm), and finally PAR (400–700 nm) (Fig. 4). 2.4. Plankton abundance and biomass The mesozooplankton within the upper 50 m of the water column was represented by 17 functional groups, 5 meroplankton (ophiopluteus, zoea, megalopa, actinotrocha, and cephalopod larvae) and 12 holoplankton. The most abundant mesozooplanktonic group was copepods (89.8%, predominantly Paracalanus and Metridia), followed by chaetognaths (3.1%), and ophiopluteus larvae (2.7%). Along the transect, meroplankton were more abundant in the Reloncaví Fjord and Sound and in the Ancud Gulf (inner area). Some carnivores and omnivores, such as chaetognaths, amphipods, and euphausiids were mainly found in the Desertores Islands and Corcovado Gulf (middle area), whereas copepods were encountered mainly between the Corcovado Gulf and the Guafo Entrance (outer area). 2.5. fDOM characterization The fDOM components of the PARAFAC model were summarized and compared with data from the scientific literature (Table 1 supplementary information). Although fDOM only represents a smaller

H.E. González et al. / Science of the Total Environment 657 (2019) 1419–1431

mg m-3

Guafo Entrance 70 60 50 40 30 20 10 0

Corcovado Gulf

A

Desertores Islands

1425

Reloncaví Sound

Ancud Gulf

Reloncaví Fjord

Chl >20 um Chl 2-20 um Chl <2 um

n.d.

n.d.

n.d.

n.d.

600 500

B

400

COP 0 m (mg m-3) COP 10 m

300 200 100 0

mg C m- 3

1500 1200

C

DOC 0 m DOC 10 m

900 600 300

n.d.

0 40 30

D

Bact 0 m Bact 10 m

20 10 0

50 49 47 44 38 36 33 32 21 20 16 14

9

8

3

4

5

6

7

Stations Fig. 3. (A–D): integrated (upper 25 m of the water column) concentrations of (A) fractionated chlorophyll a (chl a), (B) particulate organic carbon, (C) dissolved organic carbon, and (D) bacteria biomasses at surface and 10 m depth, along the Reloncaví Fjord – Guafo Entrance transect in the North Chilean Patagonia. n.d.: not determined.

fraction of the DOM pool (Fellman et al., 2010), we consider that fluorescent components provide a semi-quantitative evidence on the source in marine systems (Fellman et al., 2010), which in our case (RF and ISCh area) is composed mainly by humic like fDOM from terrestrial plant derived OMal (Vargas et al., 2011) and protein like fDOM originated by enhanced biological production (Yamashita et al., 2008) and possibly also due to the presence of salmon aquaculture (Nimptsch et al., 2015). Regarding the more refractory DOM material, values for the terrestrial humic-like component (C1 – humic like allochthonous type UVC) were highest at PR station (Puelo River) in RF, which is primarily influenced by freshwater input, concomitant terrestrial DOM (C4 – marine humic-like UVA type) tended to increase in the opposite direction, with highest values in the “Guafo Entrance sector” and slightly lower values in more freshwater influenced stations (RF sector), where the DOM pool is dominated partly by allochthonous humic-like DOM from terrestrial inputs and partly by humic matter described as of marine origin (Fig. 6). In contrast, labile DOM components resulting from autochthonous biological processes (e.g. C2 tyrosine-like and C3 tryptophanlike) showed an increase from the RF sector (mainly freshwater influenced) towards the GE sector. Thus, values for the C2 and C3 components were greater at stations 21 and 33 (at the end of the sector “Desertores Islands”) than at stations 44, 47, and 49 in the GE sector, where the DOM pool is dominated by autochthonous protein-like DOM (Fig. 6). A combined analysis of the two PARAFAC components which are more representative of terrestrial and marine compounds (i.e. fluorophores C1 and C3), together with the average percentage of allochthonous and δ13C signature in POC, showed a conspicuous gradient from the head of the RF down to the coastal oceanic area in the GE.

While the terrestrial fDOM (C1) decreased from 18.6 ± 1.2 to 10.7 ± 1.9, the marine fDOM (C3) increased from 20.9 ± 4.4 to 34.4 ± 4.9 in the land-ocean transect (Fig. 7A). The percentage of allochthonous POC decreased from 62.3 ± 8.9 to 31.9 ± 10.7 (Fig.7B), while the δ13C stable isotope increased from −23.5 ± 0.8‰ to −20.3 ± 1.0‰ at the surface along the transect (Fig. 7C). A positive linear relationship was found between the UVC-humic: UVA-humic ratio and salinity (R2 = 0.7) in the DOM along the transect, showing an increasing contribution of marine-derived humic-like DOM towards the GE (Fig. 8). 3. Discussion A freshwater plume extends from RF to the ISCh, where it deepens and widens after mixing (Figs. 2 & 4) to form the Modified Estuarine Water (MEW; Sievers and Silva, 2008). A permanent upper pycnocline, from RF to Ancud Gulf, is established all year round, being stronger during winter when the Puelo River freshwater discharge is at its maximum. However, a slight seasonal change has been reported in its mean depth, despite significant variation in winds and freshwater discharge (Castillo et al., 2015). The relationship between chl a concentration and photic zone penetration in the water column seems to be related to the phytoplankton self-shading along the entire transect, especially within RF and the Ancud Gulf, where higher chl a concentrations occurred (Fig. 4). Furthermore, a freshwater plume, loaded with organic and inorganic material, both particulate and dissolved, extends from the head of RF to the GE, i.e. flowing from a terrestrial to a marine system. This plume grows thicker as it descends moving towards the ocean, and therefore

1426

H.E. González et al. / Science of the Total Environment 657 (2019) 1419–1431

Fig. 4. Distribution of dissolved organic carbon (DOC, mg L−1) along the Reloncaví Fjord - Guafo Entrance transect (upper panel) and percentage of terrestrial DOC along the same transect (lower panel).

the particulate load is distributed through a thicker water column. This decreases the light attenuation rate, allowing solar radiation (both UV and PAR) to penetrate deeper into the water column (Fig. 4) and most likely enhancing processes linked with photomineralization (Fichot and Benner, 2014). Interestingly, chl a concentration is higher in RF, with a predominance of nanophytoplankton (2–20 μm), than in the ISCh dominated by microphytoplankton (N20 μm). Within RF, plankton self-shading, a high load of suspended particles and very low nitrate and phosphate concentrations most likely favor the growth of mediumsized to small phytoplankton. Small cell size is often beneficial in an environment limited by light and/or nutrients, as absorption efficiency is higher than in larger cells (Finkel et al., 2010). Stratification of the estuary also restricts mixing and entrainment of nutrients from subsurface waters, resulting in a limited concentration of N available for photosynthesis (González et al., 2010), a condition prevailing in most Chilean Patagonian fjords due to a low N/P ratio (b10) in relation to the theoretical (16:1) Redfield ratio (Iriarte et al., 2007). In contrast, large cells are more efficient for growth in clearer waters with deeper UV penetration and high nitrate and phosphate concentrations, such as the ISCh (Figs. 2 and 4). Higher efficiency in larger cells may be attributable to the accumulation of sufficient UV-absorbing compounds to shield UV-sensitive cellular components (Raven, 1991), and to more efficient growth when nutrients are more readily available (Finkel et al., 2010). The isotopic composition in POC (δ13C) in both extremes of the landocean transect ranged between −24.30‰ and −19.23‰, respectively (Table 1, Fig. 7C), nevertheless, when terrestrial primary producers such as herbs, shrubs, and tree leaves are included, the reported isotopic

signature ranges between −30.5‰ and −26,5‰ (Lafon et al., 2014). These authors reported that along the Serrano River Basin and the fjord system (51–52°S), the percentage of allochthonous terrestrial particulate carbon varied between 89% and 70%. These values are relatively higher than those reported in RF (41°S) that ranged by 53–70% at the surface and by 21–47% near the GE, close to the coastal ocean (Table 1, Fig. 7B). Concentrations of DOC in RF and the ISCh (range 0.4–1.3 mg L−1, average ± st. dev. 0.9 ± 0.1 mg L−1) (Fig. 3C) were within the lower range of values recorded in Martínez and Baker Fjords (1.4 mg L−1) (González et al., 2013), most likely due to the total freshwater discharge of the Baker and Pascua Rivers into Baker Fjord (800–3000 m3 s−1) which is up to five times higher than that of the Puelo River (650 m3 s−1) into RF (Niemeyer and Cereceda, 1984; Lara et al., 2008; Castillo et al., 2015), resulting in a higher POC and DOC input into Baker Fjord. Nevertheless, our values were within the lower, albeit most frequent, DOC concentrations in coastal waters of the world's oceans (see Fig. 2 from Barrón and Duarte, 2015). The input of allochthonous dissolved organic carbon (DOCal) into the head of RF accounted for 60% of the total DOC present. This is consistent with published data for the Southern Chilean Patagonia suggesting that up to 70% of the total organic carbon input at river-fjord boundaries is of terrestrial origin (Lafon et al., 2014). Most of the OMal is removed quickly from the upper water column as it moves from RF towards the ISCh, either recycled by microbial communities in the water column or exported to the sediments, while the remaining DOMal continues to be advected towards coastal marine areas (such as the GE), where it comprises only 10% of total DOC (Fig. 5).

H.E. González et al. / Science of the Total Environment 657 (2019) 1419–1431

On average, the DOC:POC ratio increased from 2.6 to 6.0 along the land-ocean transect, owing mainly to a decrease in POC. The OM entering RF is mainly of terrestrial origin (up to 60%), most likely the remains of terrestrial plants (δ13C between −29.1‰ and − 22.4‰). The more recalcitrant fraction likely cannot be efficiently degraded by the bacterial community, implying that it is not mobilized through the upper trophic layers but rather stored in the sediments (Middelburg and Nieuwehuize, 1989; Middelburg, 2018; Silva et al., 2011) or advected to the ocean. Bacteria are able to decompose semi-labile and labile DOCal and introduce it into the aquatic food web (Pomeroy, 1974; Azam et al., 1983), although the process remains unclear. For example, the role of bacterioplankton in DOM cycling depends on the various bacterial functional groups and DOM characteristics (either labile, semirecalcitrant or recalcitrant); the manner in which this occurs is not yet fully understood (Comte and del Giorgio, 2010). Labile autochthonous TOM (TOMau) is more likely (than allochthonous TOMal) to be re-mineralized by the bacterial community and partially utilized by higher trophic levels such as invertebrates and higher vertebrates (Lusseau and Wing, 2006; McLeod and Wing, 2009; Vargas et al., 2011). For example, terrestrial-derived OM may support gelatinous zooplankton during periods of low metazooplankton abundance by means of a detritus-based microbial food web (Morais et al., 2017), and/or enhance alternative carbon pathways such as mixotrophy (Paczkowska et al., 2017). It is well known that aquatic bacterial communities differ in their capacity to degrade terrestrially-derived DOM (both in quality and quantity) (Logue et al., 2016), challenging the classical view that terrestrial organic carbon is recalcitrant and contributes little to the support of aquatic metabolisms (Battin et al., 2008). In RF and the ISCh, the high seasonal variability in the Puelo River discharge (which also affect suspended organic matter subsidy) and the solar radiation regime (González et al., 2010) are important determinants for the structure and function of the pelagic community. The DOM pool within coastal areas (especially RF) showed a marked terrestrial influence dominated by humic-like components. A major fraction of the DOMal is most likely composed of cellulose, lignin-derived phenols and other biologically-recalcitrant debris of plant origin (Yamashita et al., 2015). Oxidation of the remaining labile or semilabile organic matter (especially proteins) may release nutrients able to stimulate microbial growth and facilitate development of algal blooms (Cole et al., 2006). In RF, local continental river inputs terrestrial DOM. Their fresh waters are oligotrophic in nitrate and phosphate, but rich in silicic acid (Silva and Vargas, 2014), which may increase diatom primary production. Nitrate and phosphate are mainly provided by subsurface Sub Antarctic Waters (Fig. 2, Silva, 2008), whose exchange may

Fig. 5. Distribution of chlorophyll a (μg L−1) along the Reloncaví Fjord-Guafo Entrance transect. The UV (320–395 nm) and PAR (400–700 nm) light penetration wavelengths are also depicted from the surface to 35 m depth.

1427

also be affected by tidal pumping (Liu et al., 2017). Furthermore, biological activity may boost the production of more labile protein-like DOCau compounds, especially in the ISCh, similar to reports for other terrestrial-coastal systems (Kowalczuk et al., 2003). Humic-like substances (C1) are ubiquitous in aquatic systems (Coble, 1996), especially in fjords such as RF, where the input of terrestrial organic matter is enhanced due to the intense forestry and agriculture activities in and around the Puelo River watershed, which intensify the surface and underground runoff/lateral seepage drainage into the fjord and adjacent waters. Thus, the C1 fraction of the fDOM impinges along the studied transect, contributing from 20% (local rivers and RF) to 10% (Guafo Entrance) of total fDOM (Figs. 6, 7A). The C2 and C4 signals that characterize tryptophan-like material (proteins and lessdegraded peptides) and marine humic-like UVA compounds respectively, did not show much variability, ranging from 22 to 26% of total fDOM along the transect. The C3 fraction (tyrosine-like degraded peptide material) of the OMau derived from plankton activity (mainly bacterial decomposition) contributed 20% of total fDOM in the local rivers and RF, ~30% in CG and ~40% at the GE (Figs. 6, 7A). These figures are consistent with published values for the input of bacterial-derived organic matter in Arctic watersheds, ranging from 21 to 42% of DOC and associated with extensive heterotrophic processing of DOM derived from plants and soil (Kaiser et al., 2017) and from watershed areas of the Iberian Peninsula affected by forest and agricultural areas, where protein-like DOM (both terrestrial and aquatic origin) seemed to drive bulk DOM patterns (Casas-Ruiz et al., 2017). This information highlights the role of bacterial metabolism in regulating DOM composition, DOM reactivity, and carbon flux in marine and freshwater systems. The high correlation (R2 = 0.7) between the UVC-humic like:UVAhumic like ratio and salinity (Fig. 8) indicates that different carbon sources (marine versus terrestrial) have different chemical characteristics. The strong relationship between DOC concentration and fDOM component C3 suggests that aromatic proteins associated with tyrosine-like compounds contribute to autochthonous (microbially derived) DOM (Zhang et al., 2015). The data indicate that along the surveyed transect, the high content of protein like fDOM (C2-C3) seems to be related to both a higher chl a concentration and increased riverine input (Figs. 5, 6, 7A). As suggested by Loureiro et al. (2009) labile (protein like) DOM is a potential trigger of harmful algal blooms (HAB). This may be a potential problem for aquaculture practices which are the main economic activity in the studied area as bivalves and salmon are farmed in RF and the ISCh, respectively, these areas might be under high risk of HAB's as shown by León-Muñoz et al. (2018). In addition, it also may contribute to the labile protein like DOM pool (Kamjunke et al., 2017), which in the case of salmon farms accounts for 21% of the carbon applied as feed and 76% of the annual fish production (Nimptsch et al., 2015). All of the above discussed factors likely affect the DOM composition as freshwater moves from RF to the GE (Figs. 5 & 6). During this process, terrestrial and marine humic-like DOM might be produced in situ after being biologically oxidized by microbial (Kamjunke et al., 2017) and metazoan communities. In CG, an increase in the percentage of allochthonous DOM (25%; Fig. 4) and chl a (5–6 μg L−1; Fig. 5) appeared near Stns. 44–47. A possible increase in %DOM may be due to the freshwater input from the Tantauco Natural Park (http://www.parquetantauco.cl/el-parque/). Here, a lake (Chaiguaco Lake) and small rivers (i.e. Chaigua River) discharge freshwater into CG and likely contribute with allochthonous DOC. The increase in chl a might be due to the “fertilizing effect” of the SAAW. The main exchange with the oceanic SAAW takes place in the GE, which connects with the CG. This water mass is nutrient-rich which might have local positive effects on phytoplankton growth and biomass accumulation (expressed as chl a). The bulk of the terrestrial DOC showed a sharp decrease along the land-ocean transect from RF (60%) to CG (15%) (Fig. 4 lower panel). Several factors may explain this distribution, however, as total DOM is based on the isotopic fractionation model, and PARAFAC is based on

1428

H.E. González et al. / Science of the Total Environment 657 (2019) 1419–1431

Fig. 6. Dissolved organic matter component means (0–25 m depth) along the sampling transect starting from Puelo River (PR - freshwater) to the Guafo Entrance (GE - marine water). DOM is decomposed into the four components of autochthonous (mainly C3 and less C2, C4) and allochthonous (mainly C1, and less C3) origins.

optical properties, the direct comparison of both DOC fractions (isotopic- versus chromophorical-based measurements) is somewhat difficult. In addition, it seems that a high proportion of the bulk of the terrestrial DOC, contrary to what we previously though, is quickly recycled/removed. For example, agricultural practices that promote the decomposition of soil organic matter and subsequent microbial inputs in terrestrial sources show a shift towards a more microbial/algal

and less plant/soil-derived character as human disturbance increases (Lambert et al., 2017). On the contrary, the terrestrial fraction of the chromophorically active DOC (C1 component of PARAFAC) was removed at a much more reduced rate (Fig. 7A). Thus, optical analyses suggested that although utilization of chromophoric DOC was similar, freshwater and estuarine bacterial communities differed in their preference to the humic fractions (Søndergaard et al., 2003).

H.E. González et al. / Science of the Total Environment 657 (2019) 1419–1431

A)

Percentage (%)

40 30 20 10 0 RF

RS

ISCh

CG

Sector along the transect

Allochthonous POC (%)

80

B)

60

40

The orientation of the terrestrial-marine transition pathway and its geomorphology, together with its riparian and marine vegetation and associated physical, chemical and biological processes (naturally- and anthropogenically-induced), can influence the fate terrestrial origin carbon, including its partitioning among the atmosphere, geosphere, and ocean (Wohl et al., 2017), and its distribution, transport, and utilization. A low latitude ecosystem (28–30°N, off the Louisiana shelf) studied by Fichot and Benner (2014) revealed that a significant portion of DOCal is mineralized within the mixed layer, mainly in summer (71% of the amount mineralized annually), and the coastal margin is a major sink for DOCal. In high latitude marine ecosystems, such as the Southern Chilean Patagonia or Antarctica, areas newly opened by retreating glaciers may enhance carbon utilization and become organic carbon oxidation “hot spots” (Hood et al., 2009; Cui et al., 2017). Thus, climate warming (freshwater input) and anthropogenic practices (aquaculture) boost the mobilization of terrestrial carbon pools (DOCal and POCal) and their intrusion into coastal ocean waters. These processes are intensified along terrestrial-marine transition systems and merit more attention in future climate prediction models. 4. Conclusions

20

0 0

RF

RS

ISCh

CG

RF

RS

ISCh

CG

C) POC Stable Isotope (δ13C)

1429

-5 -10 -15

-20 -25

Fig. 7. Spatial distribution of the most conspicuous fDOM signal of terrestrial (C1) and marine (C3) components (A), the percentage of allochthonous POC (B), and the δ13C stable isotope signature (C). The average data (and standard deviation) was plotted for the four major areas along the land-ocean transect: The Reloncaví Fjord (RF), Reloncaví Sound (RS), Inner Sea of Chiloé (ISCh), and Corcovado Gulf (CG).

A conspicuous gradient in physical, chemical and biological parameters along a terrestrial - marine transition system characterized the northern Patagonia fjords. The head of the RF was characterized by high values of terrigenous-derived DOC (−24.3‰) and silicic acid, and a high concentration of nano-phytoplankton (62 mg m−2) in a shallow photic zone (12 m). This contrast with a coastal marine area characterized by high values of autochthonous DOC (−19.2‰) and nitrate/phosphate, and relatively low values of micro-phytoplankton (11 mg m−2) in a deep photic zone (29 m). The system moved from a predominance of humic-C1 terrigenousderived and protein-like C2 DOC at the fjord head towards a tyrosinelike C3 autochthonous DOC at the coastal ocean area. Isotopic fractionation-based terrestrial DOM shows a quicker recycling/removal from the source (fjord head) than the optically-based DOM estimations. Overall, direct comparison of isotopic and chromophorical (optical) estimations should be analyzed with care given that fDOM is only a subset (usually highly variable) of total DOM, although simultaneous measurements can be used as “sentinel” of terrestrial influence in marine systems. The high correlation (R2 = 0.7) between the UVC-humic: UVA-humic ratio and salinity points to the fact that processes associated with freshwater input impinged on the DOC chemical characteristics and origins. Overall, climate warming (freshwater input) and anthropogenic practices (aquaculture) boost the mobilization of terrestrial carbon pools and their intrusion into coastal ocean areas, a process that should be given more attention in climate prediction models. Supplementary data to this article can be found online at https://doi. org/10.1016/j.scitotenv.2018.11.445. Acknowledgements This work was supported by the Projects CONA-C17F 11-03 (Granted to H.E. González), and from FONDAP-IDEAL 15150003. The authors would like to thank L. Lizarraga, P. Reinoso, E. Menschel, for their valuable help and support during sampling and laboratory analysis. References

Fig. 8. Relationship between UVC humic: UVA humic ratio and salinity for all stations sampled along the terrestrial (Reloncaví Fjord) – marine (Guafo Entrance) transect.

Atlas, E.S., Hager, S., Gordon, L., Park, P., 1971. A practical manual for use of the Technicon Autoanalyser in sea water nutrient analyses. Department of Oceanography. Oregon State University, Corvallis, OR. Azam, F., Fenchel, T., Field, J.G., Gray, J.S., Meyer-Reil, L.A., Thingstad, F., 1983. The ecological role of water-column microbes in the sea. Mar. Ecol. Prog. Ser. 10, 257–263. www.int-res.com/articles/meps/10/m010p257. Barber, A., Sirois, M., Chaillou, G., Gélinas, Y., 2017. Stable isotope analysis of dissolved organic carbon in Canada's eastern coastal waters. Limnol. Oceanogr. 62, 85–94. https:// doi.org/10.1002/lno.10666.

1430

H.E. González et al. / Science of the Total Environment 657 (2019) 1419–1431

Barrón, C., Duarte, C.M., 2015. Dissolved organic carbon pools and export from the coastal ocean. Glob. Biogeochem. Cy. 29, 1725–1738. https://doi.org/10.1002/ 2014GB005056. Battin, T.J., Kaplan, L.A., Findlay, S., Hopkinson, C.S., Marti, E., Packman, A.I., Newbold, J.D., Sabater, F., 2008. Biophysical controls on organic carbon fluxes in fluvial networks. Nat. Geosci. 1, 95–100. https://doi.org/10.1038/ngeo101. Bianchi, T.S., 2007. Biogeochemistry of Estuaries. Oxford Univ. Press (706 pp). Canuel, E.A., Cloern, J.E., Ringelberg, D.B., Guckert, J.B., Rau, G.H., 1995. Molecular and isotopic tracers used to examine sources of organic matter and its incorporation into the food webs of San Francisco Bay. Limnol. Oceanogr. 40, 67–81. https://doi.org/ 10.4319/lo.1995.40.1.0067. Casas-Ruiz, J.P., Catalán, N., Gómez-Gener, L., von Schiller, D., Obrador, B., Kothawala, D.N., López, P., Sabater, S., Marcé, R., 2017. A tale of pipes and reactors: control son the instream dynamics of dissolved organic matter in rivers. Limnol. Oceanogr. 62, 85–94. https://doi.org/10.1002/lno.10471. Castillo, M.I., Cifuentes, U., Pizarro, O., Djurfeldt, L., Cáceres, M., 2015. Seasonal hydrography and surface outflow in a fjord with a deep sill: the Reloncaví fjord, Chile. Ocean Sci. Discuss. 12, 2535–2564. https://doi.org/10.5194/os-12-533-2016. Cawley, K.M., Butler, K.D., Aiken, G.R., Larsen, L.G., Huntington, T.G., McKnight, D.M., 2012. Identifying fluorescent pulp mill effluent in the Gulf of Maine and its watershed. Mar. Pol. Bull. 64, 1678–1687. https://doi.org/10.1016/j.marpolbul.2012.05.040. Coble, P.G., 1996. Characterization of marine and terrestrial DOM in seawater using excitation-emission matrix spectroscopy. Mar. Chem. 51, 325–346. https://doi.org/ 10.1016/0304-4203(95)00062-3. Cole, J.J., Carpenter, S.R., Pace, M.L., de Bogert, M.C., Kitchell, J.L., Hodson, J.R., 2006. Differential support of lake food webs by three types of terrestrial organic carbon. Ecol. Lett. 9, 558–568. https://doi.org/10.1111/j.1461-0248.2006.00898.x. Comte, J., del Giorgio, P.A., 2010. Linking the patterns of change in composition and function in bacterioplankton successions along environmental gradients. Ecology 91, 1466–1476. https://doi.org/10.1890/09-0848.1. Cory, R.M., McKnight, D.M., 2005. Fluorescence spectroscopy reveals ubiquitous presence of oxidized and reduced quinones in dissolved organic matter. Environ. Sci. Technol. 39, 8142–8149. https://doi.org/10.1021/es0506962. Cui, X., Bianchi, T.S., Kenney, W.F., Wang, J., Curtis, J.H., Xu, K., Savage, C., 2017. Carbon dynamics along a temperate fjord-head delta: Linkages with carbon burial in fjords. J. Geophys. Res. Biogeosci. 122, 3419–3430. https://doi.org/10.1002/2017JG003891. Dale, A.W., Prego, R., 2005. Net autotrophy and heterotrophy in the Pontevera Ria upwelling system (NW Iberian margin). Cienc. Mar. 31, 213–220. http://www.scielo.org.mx/ pdf/ciemar/v31n1b/v31n1ba8.pdf. Fellman, J.B., Hood, E., Spencer, R.G.M., 2010. Fluorescence spectroscopy opens new windows into dissolved organic matter dynamics in freshwater ecosystems: a review. Limnol. Oceanogr. 55 (6), 2452–2462. https://doi.org/10.4319/ lo.2010.55.6.2452. Fichot, C.G., Benner, R., 2014. The fate of terrigenous dissolved organic carbon in a riverinfluenced ocean margin. Glob. Biogeochem. Cy. 28, 300–318. https://doi.org/ 10.1002/2013GB004670. Finkel, Z.V., Beardall, J., Flynn, K.J., Quigg, A., Rees, T.A.V., Raven, J.A., 2010. Phytoplankton in a changing world: cell size and elemental stoichiometry. J. Plankton Res. 32, 119–137. https://doi.org/10.1093/plankt/fbp098. García-Martín, E.E., Daniels, C.J., Davidson, K., Davis, C.E., Mahaffey, C., Mayers, K.M.J., McNeill, S., Poulton, A.J., Purdie, D.A., Tarran, G.A., Robinson, C., 2018. Seasonal changes in plankton respiration and bacterial metabolism in a temperate shelf sea. Prog. Oceanogr. https://doi.org/10.1016/j.pocean.2017.12.002 in press. González, H.E., Calderón, M.J., Castro, L., Clément, A., Cuevas, L.A., Daneri, G., Iriarte, J.L., Lizárraga, L., Martínez, R., Menschel, E., Silva, N., Carrasco, C., Valenzuela, C., Vargas, C.A., Molinet, C., 2010. Primary production and plankton dynamics in the Reloncaví Fjord and the interior Sea of Chiloé, Northern Patagonia. Mar. Ecol. Prog. Ser. 402, 13–30. https://doi.org/10.3354/meps08360. González, H.E., Castro, L.R., Daneri, G., Iriarte, J.L., Silva, N., Tapia, F., Teca, E., Vargas, C.A., 2013. Land-ocean gradient in haline stratification and its effects on plankton dynamics and trophic carbon fluxes in Chilean Patagonian fjords (47°–50°S). Prog. Oceanogr. 119, 32–47. https://doi.org/10.1016/j.pocean.2013.06.003. González, H.E., Graeve, M., Kattner, G., Silva, N., Castro, L., Iriarte, J.L., Osmán, L., Daneri, G., Vargas, C., 2016. Carbon flow through the pelagic food web in southern Chilean Patagonia: relevance of Euphausia vallentini as key species. Mar. Ecol. Progr. Ser. 557, 91–110. https://doi.org/10.3354/meps11826. Green, S., Blough, N., 1994. Optical absorption and fluorescence properties of chromophoric dissolved organic matter in natural waters. Limnol. Oceanogr. 39, 1903–1916. https://doi.org/10.4319/lo.1994.39.8.1903. Gupta, G.V.M., Thottathil, S.D., Balachandran, K.K., Madhu, N.V., Madeswaran, P., Nair, S., 2009. CO2 supersaturation and net heterotrophy in a tropical estuary (Cochin, India): influence of anthropogenic effect. Ecosystems 12, 1145–1157. https://doi. org/10.1007/s10021-009-9280-2. Hall, G.J., Clow, K.E., Kenny, J.E., 2005. Estuarial fingerprinting through multidimensional fluorescence and multivariate analysis. Environ. Sci. Technol. 39, 7560–7567. https://doi.org/10.1021/es0503074. Hood, E., Fellman, J., Spencer, R.G.M., Hernes, P.J., Edwards, R., D'Àmore, D.D., Scott, D., 2009. Glaciers as a source of ancient and labile organic matter to the marine environment. Nature 462, 1044–1048. https://doi.org/10.1038/nature08580. Iriarte, J.L., González, H.E., Liu, K.K., Rivas, C., Valenzuela, C., 2007. Spatial and temporal variability of chlorophyll and primary productivity in surface waters of southern Chile (41.5–43°S). Estuar. Coast. Shelf Sci. 74, 471–480. https://doi.org/10.1016/j. ecss.2007.05.015. Kaiser, K., Canedo-Oropeza, M., McMahon, R., Amon, R.M.W., 2017. Origins and transformations of dissolved organic matter in large Arctic rivers. Sci. Rep. 7, 13064. https:// doi.org/10.1038/s41598-017-12729-1.

Kamjunke, N., Nimptsch, J., Harir, M., Herzsprung, P., Schmitt-Kopplin, P., Neu, T.R., Graeber, D., Osorio, S., Valenzuela, J., Reyes, J.C., Woelfl, S., Hertkorn, N., 2017. Landbased salmon aquacultures change the quality and bacterial degradation of riverine dissolved organic matter. Sci. Rep. 123, 513–523. https://doi.org/10.1038/srep43739. Karlsson, J., Jansson, M., Jonsson, A., 2007. Respiration of allochthonous organic carbon in unproductive forest lakes determined by the Keeling plot method. Limnol. Oceanogr. 52, 603–608. Kemp, P.F., Sherr, B.F., Sherr, E.B., Cole, J.J., 1993. Handbook of Methods in Aquatic Microbial Ecology. CRC Press, Boca Raton (800 pp. ISBN 0-87371-564-0). Kowalczuk, P., Cooper, W.J., Whitehead, R.F., Durako, M.J., Sheldon, W., 2003. Characterization of CDOM in an organic-rich river and surrounding coastal ocean in the South Atlantic Bight. Aquat. Sci. 65, 384–401. https://doi.org/10.1007/s00027-003-0678-1. Lafon, A., Silva, N., Vargas, C.A., 2014. Contribution of allochthonous organic carbon across the Serrano River Basin and the adjacent fjord system in southern Chilean Patagonia: insights from the combined use of stable isotopes and fatty acid biomarkers. Prog. Oceanogr. 129, 98–113. https://doi.org/10.1016/j.pocean.2014.03.004. Lambert, T., Bouillon, S., Darchambeau, F., Morana, C., Roland, F.A.E., Descy, J.-P., Borges, A.V., 2017. Effects of human land use on the terrestrial and aquatic sources of fluvial organic matter in a temperate river basin (the Meuse River, Belgium). Biogeochemistry 136, 191–211. https://doi.org/10.1007/s10533-017-0387-9. Lara, A., Villalba, R., Urrutia, R., 2008. A 400-year tree-ring record of the Puelo River summer– fall streamflow in the Valdivian Rainforest eco-region. Clim. Chang. 86, 331–356. https://doi.org/10.1007/s10584-007-9287-7. Lawaetz, A.J., Stedmon, C., 2009. Fluorescence intensity calibration using Raman scatter peak. Appl. Spectrosc. 63, 936–940. https://doi.org/10.1366/000370209788964548. Lee, S., Fuhrman, J., 1987. Relationship between biovolume and biomass of naturally derived marine bacterioplankton. Appl. Environ. Microbiol. 53, 1298–1303. León-Muñoz, J., Urbina, M.A., Garreaud, R., Iriarte, J.L., 2018. Hydroclimatic conditions trigger record harmful algal bloom in western Patagonia (summer 2016). Sci. Rep. 8, 1330. https://doi.org/10.1038/s41598-018-19461-4. Liu, Y., Jiao, J.J., Liang, W., Luo, X., 2017. Tidal pumping-induced nutrient dynamics and biogeochemical implications in an intertidal aquifer. J. Geophys. Res. Biogeosci. 122, 3322–3342. https://doi.org/10.1002/2017JG004017. Logue, J.B., Stedmon, C.A., Kellerman, A.M., Nielsen, N.J., Anderson, A.F., Laudon, H., Lindström, E.S., Kritzberg, E.S., 2016. Experimental insights into the importance of aquatic bacterial community composition to the degradation of dissolved organic matter. ISME J. 10, 533–545. https://doi.org/10.1038/ismej.2015.131. Loureiro, S.A., Garcés, E., Collos, Y., Vaqué, D., Camp, J., 2009. Effect of marine autotrophic dissolved organic matter (DOM) on Alexandrium catenella in semi-continuous cultures. J. Plankton Res. 31, 1363–1372. https://doi.org/10.1093/plankt/fbp080. Lusseau, S.M., Wing, S.R., 2006. Importance of local production versus pelagic subsidies in the diet of an isolated population of bottlenose dolphins Tursiops sp. Mar. Ecol. Prog. Ser. 321, 283–293. https://doi.org/10.3354/meps321283. Martineau, C., Vincent, W.F., Frenette, J.J., Dodson, J.J., 2004. Primary consumers and particulate organic matter: isotopic evidence of strong selectivity in the estuarine transition zone. Limnol. Oceanogr. 49, 1679–1686. https://doi.org/10.4319/ lo.2004.49.5.1679. McLeod, R., Wing, S., 2007. Hagfish in the New Zealand fjords are supported by chemoautotrophy of forest carbon. Ecology 88, 809–816. McLeod, R., Wing, S., 2009. Strong pathways for incorporation of terrestrially derived organic matter into benthic communities. Estuar. Coast. Shelf Sci. 82, 645–653. Middelburg, J.J., 2018. Reviews and synthesis: to the bottom of carbón processing at the seafloor. Biogeosciences 15, 413–427. https://doi.org/10.5194/bg-15-413-2018. Middelburg, J.J., Nieuwehuize, J., 1989. Carbon and nitrogen stable isotope in suspended matter and sediments from the Schelde Estuary. Mar. Chem. 60, 217–225. Morais, P., Dias, E., Cruz, J., Chainho, P., Angélico, M.M., Costa, J.L., Barbosa, A.B., Teodósio, M.A., 2017. Allochthonous-derived organic matter subsidizes the food sources of estuarine jellyfish. J. Plankton Res. 39, 870–877. https://doi.org/10.1093/plankt/fbx049. Murphy, K.R., Butler, K.D., Spencer, R.G.M., Stedmon, C.A., Boehme, J.R., Aiken, G.R., 2010. Measurement of dissolved organic matter fluorescence in aquatic environments: an Interlaboratory comparison. Environ. Sci. Technol. 44, 9405–9412. https://doi.org/ 10.1021/es102362t. Niemeyer, H., Cereceda, P., 1984. Hidrografía. Colección Geográfica de Chile. Tomo VIII. Instituto Geográfico Militar, Santiago, Chile, p. 313. Nimptsch, J., Woelfl, S., Osorio, S., Valenzuela, J., Ebersbach, P., von Tuempling, W., Palma, R., Encina, F., Figueroa, D., Kamjunke, N., Graeber, D., 2015. Tracing dissolved organic matter (DOM) from land-based aquaculture systems in North Patagonia streams. Sci. Total Environ. 537, 129–138. https://doi.org/10.1016/j.scitotenv.2015.07.160. Paczkowska, J., Rowe, O., Schluter, L., Legrand, C., Karlson, B., Andersson, A., 2017. Allochthonous matter: an important factor shaping the phytoplankton community in the Baltic Sea. J. Plankton Res. 30, 23–34. https://doi.org/10.1093/plankt/fbw081. Pantoja, S., Iriarte, J.L., Gutiérrez, M., Calvete, C., 2010. The Southern Chile continental margin. In: Liu, K.K., Atkinson, L., Quiñones, R., Talaue-McManus, L. (Eds.), Carbon and Nutrient Fluxes in Continental Margins: A Global Synthesis. Global Change. The IGBP Series. Springer Verlag, Berlin, pp. 265–272. Parsons, T.R., Maita, R., Lalli, C.M., 1984. Counting, media and preservation. A Manual of Chemical and Biological Methods for Seawater Analysis. Pergamon Press, Toronto. Pomeroy, L.R., 1974. The ocean's food web, a changing paradigm. Bioscience 24, 499–504. https://doi.org/10.2307/1296885. Porter, K.G., Feig, Y.S., 1980. The use of DAPI for identifying and counting aquatic microflora. Limnol. Oceanogr. 25, 943–994. Raven, J.A., 1991. Responses of aquatic photosynthetic organisms to increased solar UVB. J. Photochem. Photobiol. B Biol. 9, 239–244. https://doi.org/10.1016/1011-1344(91) 80158-E. Raymond, P.A., Bauer, J.E., 2001. Use of 14C and 13C natural abundances for evaluating riverine, estuarine, and coastal DOC and POC sources and cycling: a review and

H.E. González et al. / Science of the Total Environment 657 (2019) 1419–1431 synthesis. Org. Geochem. 32, 469–485. https://doi.org/10.1016/S0146-6380(00) 00190-X. Rebolledo, L., Lange, C., Bertrand, S., Muñoz, P., 2015. Late Holocene precipitation variability recorded in the sediments of Reloncaví Fjord (41°S, 72°W), Chile. Quat. Res. 84, 21–36. https://doi.org/10.1016/j.yqres.2015.05.006. Sievers, H., Silva, N., 2008. Water masses and circulation in austral Chilean channels and fjords. In: Silva, N., Palma, S. (Eds.), Progress in the oceanographic knowledge of Chilean interior waters, from Puerto Montt to Cape Horn. Comité Oceanográfico Nacional - Pontificia Universidad Católica de Valparaíso, Valparaíso, pp. 53–58. Silva, N., 2008. Dissolved oxygen, pH, and nutrients in the austral Chilean channels and fjords. In: Silva, N., Palma, S. (Eds.), Progress in the Oceanographic Knowledge of Chilean Interior Waters, From Puerto Montt to Cape Horn. Comité Oceanográfico Nacional - Pontificia Universidad Católica de Valparaíso, Valparaíso, pp. 37–43. Silva, N., Prego, R., 2002. Carbon and nitrogen spatial segregation and stoichiometry in the surface sediments of southern Chilean inlets (41°–56°S). Estuar. Coast. Shelf Sci. 55, 763–775. Silva, N., Vargas, C.A., 2014. Hypoxia in Chilean Patagonian Fjords. Prog. Oceanogr. 129, 62–74. Silva, N., Vargas, C.A., Prego, R., 2011. Land–ocean distribution of allochthonous organic matter in surface sediments of the Chiloé and Aysén interior seas (Chilean Northern Patagonia). Cont. Shelf Res. 31, 330–339. Singh, S., D'Sa, E.J., Swenson, E.M., 2010. Chromophoric dissolved organic matter (CDOM) variability in Barataria Basin using excitation–emission matrix (EEM) fluorescence and parallel factor analysis (PARAFAC). Sci. Total Environ. 408, 3211–3222. https:// doi.org/10.1016/j.scitotenv.2010.03.044. Smith, S.V., Hollibaugh, J.T., 1993. Coastal metabolism and the ocean organic carbon balance. Rev. Geophys. 31, 75–89. https://doi.org/10.1029/92RG02584. Søndergaard, M., Stedmon, C.A., Borch, N.H., 2003. Fate of terrigenous dissolved organic matter (DOM) in estuaries: aggregation and bioavailability. Ophelia 57, 161–176. https://doi.org/10.1080/00785236.2003.1040951. Stedmon, C.A., Bro, R., 2008. Characterizing dissolved organic matter fluorescence with parallel factor analysis: a tutorial. Limnol. Oceanogr. Methods 6, 572–579. https:// doi.org/10.4319/lom.2008.6.572.

1431

Stedmon, C.A., Markeger, S.S., 2005a. Resolving the variability in dissolved organic matter fluorescence in a temperate estuary and its catchment using PARAFAC analysis. Limnol. Oceanogr. 50 (2), 686–697. https://doi.org/10.4319/lo.2005.50.2.0686. Stedmon, C.A., Markeger, S.S., 2005b. Tracing the production and degradation of autochthonous fractions of dissolved organic matter by fluorescence analysis. Limnol. Oceanogr. 50 (5), 1415–1426. https://doi.org/10.4319/lo.2005.50.5.1415. Vargas, C.A., Martínez, A., San Martin, V., Aguayo, M., Silva, N., Torres, R., 2011. Allochthonous subsides of organic matter across a lake-river-fjord landscape in the Chilean Patagonia: implications for marine zooplankton in inner fjord areas. Cont. Shelf Res. 31 (3–4), 187–201. https://doi.org/10.1016/j.csr.2010.06.016. Verney, R., Lafite, R., Brun-Cottan, J.-C., 2009. Flocculation potential of estuarine particles: the importance of environmental factors and of the spatial and seasonal variability of suspended particulate matter. 2009. Est. Coast. 32, 678–693. https://doi.org/10.1007/ s12237-009-9160-1. von Bodungen, B., Wunsch, M., Fürderer, H., 1991. Sampling and analysis of suspended and sinking particles in the northern North Atlantic. In: Hurd, D.C., Spencer, D.W. (Eds.), Marine Particles: Analysis and Characterization. AGU Geophys. Monogr 63, pp. 47–56. Wohl, E., Hall Jr., R.O., Lininger, K.B., Sutfin, N.A., Walters, D.M., 2017. Carbon dynamics of river corridors and the effects of human alterations. Ecol. Monogr. 87, 379–409. https://doi.org/10.1002/ecm.1261. Yamashita, Y., Rudolf, J., Nagamitsu, M., Eiichiro, T., 2008. Assessing the dynamics of dissolved organic matter (DOM) in coastal environments by excitation emission matrix fluorescence and parallel factor analysis (EEM-PARAFAC). Limnol. Oceanogr. 53, 1900–1908. https://doi.org/10.4319/lo.2008.53.5.1900. Yamashita, Y., Fichot, C.G., Shen, Y., Jaffé, R., Benner, R., 2015. Linkages among fluorescent dissolved organic matter, dissolved amino acids and lignin-derived phenols in a river-influenced ocean margin. Front. Mar. Sci. 2, 92. https://doi.org/10.3389/ fmars.2015.00092. Zhang, Y., Liang, X., Wang, Z., Lixian, X., 2015. A novel approach combining self-organizing map and parallel factor analysis for monitoring water quality of watersheds under non-point source pollution. Sci. Rep. 5, 16079. https://doi.org/10.1038/srep16079.