- Email: [email protected]
Distribution and origin of inorganic and organic carbon in the sediments of Kongsfjorden, Northwest Spitsbergen, European Arctic
Author’s Accepted Manuscript Distribution and origin of inorganic and organic carbon in the sediments of Kongsfjord, northwest Spitsbergen, European Arctic Katarzyna Koziorowska, Karol Kuliński, Janusz Pempkowiak www.elsevier.com/locate/csr
PII: DOI: Reference:
S0278-4343(17)30169-3 http://dx.doi.org/10.1016/j.csr.2017.08.023 CSR3669
To appear in: Continental Shelf Research Received date: 29 March 2017 Revised date: 29 August 2017 Accepted date: 30 August 2017 Cite this article as: Katarzyna Koziorowska, Karol Kuliński and Janusz Pempkowiak, Distribution and origin of inorganic and organic carbon in the sediments of Kongsfjord, northwest Spitsbergen, European Arctic, Continental Shelf Research, http://dx.doi.org/10.1016/j.csr.2017.08.023 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Distribution and origin of inorganic and organic carbon in the sediments of Kongsfjord, northwest Spitsbergen, European Arctic Katarzyna Koziorowska*, Karol Kuliński, Janusz Pempkowiak Institute of Oceanology Polish Academy of Sciences, ul. Powstańców Warszawy 55, Sopot, Poland *Corresponding author, e-mail: [email protected] Abstract Sedimentary organic carbon in the Arctic, including the continental shelf and fjords, has been relatively well investigated, whereas much less is known about sedimentary inorganic carbon (carbonates) in fjords. The distribution and provenience of both sedimentary organic and inorganic carbon in a high-Arctic fjord (Kongsfjord, 79ºN) was the subject of this study. Stratified bottom sediments (cores) and suspended particulate matter (SPM) were analyzed for total (Ctot), organic (Corg), and inorganic (Cinorg) carbon as well as calcium, magnesium, and strontium. The sediments were dated using the 210Pb method. Sedimentation rates ranged from 0.14 cm (fjord mouth, FM) to several cm (close to the glacier front, GF) year−1. Sedimentary Corg concentrations were higher at the FM (~20 mg g−1 dry sediment) than at the GF (~1 mg g−1), while concentrations of Cinorg were lower at the FM (16.8 mg g−1) than at the GF (45 mg g−1). SPM concentrations were highest, and Cinorg most abundant at the GF. The data suggest that Corg is mostly produced in situ, with glaciers serving as only a minor source. The Cinorg to Corg ratios, Ca, Mg, and Sr concentrations, and the molar ratios of Mg:Ca and Sr:Ca together indicated that carbonates close to the GF are of terrigenous origin and those at the FM almost exclusively biogenic. Carbonates originating from these two sources differ in their composition. The Mg:Ca and Sr:Ca molar ratios were 0.56 and 0.00015 for glacial carbonates and 0.94 and 0.00020 for biogenic carbonates. Key words: high-Arctic fjord; suspended matter; sedimentary organic matter; sedimentary carbonates; provenience; Mg:Ca ratio
1. Introduction
1
Accurate determinations of atmospheric carbon dioxide concentrations are of utmost importance for predicting the impacts of climate warming and ocean acidification. Marine and lacustrine sediments constitute a long-term natural carbon sink and thus indirectly condition carbon dioxide in the atmosphere. Accordingly, quantifying the distribution of carbon in sediments will contribute to a better understanding of carbon budgets. The two major species of carbon in the environment are organic carbon (Corg) and inorganic carbon (Cinorg). The material flux reaching marine bottom sediments comprises a mixture of the two forms. Both may originate from internal biological activity or/and be transported to the sea from the land. Terrestrial sources are also the result of biological activities. Because these activities occur under different environmental conditions, the properties of the resulting carbon-bearing matter that makes up sediments will differ as well (Bhushan et al., 2001). Until the 1990s, most information regarding carbon deposition and preservation in the surficial sediments was limited, derived from data on the carbon concentrations in samples collected with a grab sampler (Gorlich et al., 1987; Hulth et al., 1996); consequently, the age span represented by the sediments was unknown. With recent approaches to measuring dated sediment cores (Szczuciński et al., 2009), much has been learned about the distribution and provenience of Corg. In the Arctic, analyses of sediment cores have shown that carbon abundances are much greater in the sediments of Arctic fjords than in the Arctic Ocean. The sediments in fjords contain up to 40 mg g−1 Corg and even 70-100 % is of terrestrial provenience (Koziorowska et al., 2016). However, stable carbon isotopes may not be reliable markers of Corg provenience, because the typical pattern of a much lower
13
C abundance in
terrestrial vs. marine organic matter does not apply in the Spitsbergen fjords (Kumar et al., 2016). Moreover, sediments from fjords deliver much more carbon to carbon cycling than would be anticipated based on their areal contribution to the Arctic Ocean (Stein and Macdonald, 2004). This reflects the fact that fjords, with their high sedimentary Corg concentrations and sediment accumulation rates (SARs), have the highest area-normalized Corg burial rates (Smith et al., 2015). In addition, Corg concentrations have been often reported together with total nitrogen (Ntot) concentrations, Corg to Ntot molar ratios, and δ15Ntot and δ13Corg. By contrast, Cinorg (carbonates) has been almost totally neglected (Carroll et al., 2008; Kędra et al., 2012; Koziorowska et al., 2016; Kuliński et al., 2014; Zaborska et al., 2006), such that little is known about its distribution and provenience in bottom sediments of the Arctic Ocean and fjords.
2
Biogenic carbonates mostly consist of calcium carbonates and mixed calcium/strontium and calcium/magnesium carbonates (Andruleit et al., 1996; Carpenter and Lohmann, 1992; Ries, 2006). They can be delivered to marine sediments but also from land, the latter as a result of limestone weathering (Bhushan et al., 2001; Hauck et al., 2012). Moreover, carbonates can precipitate due to excess dissolution, most often during the mixing of fresh, carbonate-rich water with saline water (seawater) (Morse, 2003). The current biological activity of the fjords and Spitsbergen shelf as a source of carbonates to the bottom sediments has been described (Andruleit et al., 1996; Faust et al., 2014) whereas the Cinorg distribution in the sediments of the fjords has yet to be investigated. In the immediate vicinity of glaciers, primary production in the water column and the activities of sediment-dwelling organisms are very limited (Gorlich et al., 1987; Stein and Macdonald, 2004; Włodarska-Kowalczuk et al., 2007). Thus, neither marine Corg nor biogenic carbonates likely contribute to the bottom sediments of these regions. The large carbonate concentrations in the immediate vicinity of glaciers must therefore result from deposition of the weathering products of rocks, primarily limestone, (Trusel et al., 2010). However, there are no data on the contribution of weathered material in sediments. This study sought to fill this knowledge gap by measuring the organic and, especially, the inorganic carbon distribution in the sediments of a Spitsbergen fjord. In addition, given the different characteristics of autochthonous and allochthonous carbon pools, we traced the provenience of the identified carbon species. Specifically, we examined the distribution of Corg and Cinorg (carbonates) in surficial sediments in the Kongsfjord, a high Arctic (79ºN) fjord. Both vertical and horizontal Corg and Cinorg gradients were investigated by analyzing 210
Pb-dated sediment cores collected in a transect from the fronts of tidal glaciers (GFs) to the
mouth of the fjord (FM). Carbonates were characterized with respect to their concentrations and their molar ratios with calcium (Ca), magnesium (Mg), and strontium (Sr). The data were also used to distinguish between allochthonous (terrestrial) and autochthonous pools of inorganic and organic carbon.
2. Study area Kongsfjord is located on the northwestern coast of Spitsbergen (the main island of the Svalbard Archipelago). It is a relatively small fjord (26 km long and 4–10 km wide), with a wide opening to the ocean. A chain of islands (Lovénøyane) divides the Kongsfjord into two 3
parts. The outer basin, with the average depth of 200–300 m and only one large island (Blomstrandøya), is strongly affected by the relatively warm and saline North-Atlantic-type waters (temperature >2°C, salinity of 35 PSU) transported by the West Spitsbergen Current. The inner part (average depth of 50–60 m) is influenced by four tidewater glaciers: Kronebreen (the largest and most active), Kongsvegen, Conwaybreen, and Blomstrandbreen (Svendsen et al., 2002). Together, these glaciers annually supply ~0.33 km3 of fresh water to the fjord (Beszczyńska-Moller et al., 1997). Kongsbreen is retreating at a rate of up to 0.5 km per year and its meltwater annually supplies the fjord with ~2.6 × 105 m3 of mineral material. The geological cover of the catchment area of these four tidewater glaciers is mainly formed by moraines and marine shore deposits composed of quartz and polymict conglomerates, limestones, dolomites, marl, gypsum, anhydrite and carbonate breccia, siliceous shales, cherts, and sandstones (Streuff, 2013). The large supply of freshwater from ice calving and the melting of snow and ice in summer causes relatively high SARs. These vary from 20 000 g m−2 year−1 in the vicinity of the front of the glacier (inner part), to 1800–3800 g m−2 year−1 in the central part, and to ~200 g m−2 year−1 in the outer fjord (Zajaczkowski, 2008). Most surface sediments (below a water depth of 30–40 m) are composed of fairly uniform mud (Włodarska-Kowalczuk and Pearson, 2004). Despite numerous studies on the structures and functioning of ecosystems, little is known about annual primary production in Kongsfjord. According to the available literature data, it is lower than at other fjords in this region, with similar rates of 20 –40 g C m−2 year−1 (Piwosz et al., 2009) and 35–50 g C m−2 year−1 (Hop et al., 2002) reported in different studies.
3. Experimental 3.1. Sampling Sediment cores and suspended material from the water column were collected during the R/V Oceania cruise in July 2015. A Niemisto gravity corer was used to collect sediment cores at three stations located in the outer, central, and inner parts of the fjord (Fig. 1: Kb1, Kb2, and Kb3, respectively). Immediately after their collection, the cores were sliced into 10-mm thick layers and frozen (−20°C) until the analyses. SPM was sampled at four stations (Fig. 1: Kb0, Kb1, Kb2, and Kb3) by filtering the samples collected from four different depths through precombusted Whatman GF/F glass-fiber filters. The depths corresponded to the surface, the maximum fluorescence layer, the subsurface (40 m), and the bottom water (~2 m above the seafloor). Filters with the separated SPM were immediately frozen at −80°C. 4
All analyses were carried out in the Marine Biogeochemistry Laboratory of the Institute of Oceanology, Polish Academy of Sciences, Sopot, Poland.
Fig. 1. Locations of sampling stations in the Kongsfjord
3.2. Corg and Cinorg analyses Ctot and Corg concentrations of the samples were analyzed in an elemental analyzer (Flash EA 1112 series) combined with an isotopic ratio mass spectrometer (Delta V Advantage (Thermo Electron Corp., Germany), according to the procedure described by Kuliński et al. (2014). Thirty mg (0.001 mg accuracy) of freeze-dried and homogenized material was weighed into silver capsules. Samples used to determine the Corg concentration were additionally soaked in 2M HCl to remove carbonates and then dried at 60°C for 24 h (the procedure was repeated until a constant weight was reached) prior to the analysis. Quantitative measurements were calibrated against analyses of certified reference materials (Fluβsediment) provided by HEKAtech GmbH (Germany). The precision (expressed as the relative standard deviation) was better than 1.6% (n=5) for Ctot measurements and 1.4% for Corg measurements. Cinorg concentrations were obtained by subtracting the Corg concentration from that of Ctot. The carbon concentrations are expressed as mg g−1 and mmol g−1 (12 mg = 1 mmol). 3.3. 210Pb analyses
5
The
Pb method was used to calculate annual SARs (in cm year−1). The rates were
210
determined from profiles of excess
210
Pb activity (210Pbex = total
210
Pb−supported
210
Pb) vs.
the porosity-corrected sediment depth and calculated by assuming a secular equilibrium between
210
Po and
210
Pb. A linear relation between log Pbex and depth in the analyzed cores
was assumed to calculate SAR using a least squares fit (Pempkowiak, 1991; Zaborska et al., 2007). Details of the radiochemical
210
Po activity measurements were described previously
(Pempkowiak, 1991; Zaborska et al., 2007). Briefly, ~200 mg of sediment were spiked with the 209Po chemical yield tracer and digested using 2 ml of concentrated perchloric acid and 3 ml of 12 M hydrofluoric acid.
210
Po was deposited spontaneously on a silver disk and then
counted for at least 24 h in a multi-channel analyzer (Canberra) equipped with a Si/Li detector.
210
Po activity concentrations were calculated by assuming equal chemical recovery
of the spiked (209Po) and sedimentary (210Po) radio-polonium activity. Blanks and reference materials (IAEA 300 and 326) were measured for quality control. Both the recovery and the precision of the determinations were highly satisfactory (Koziorowska et al., 2016).
3.4 Metal concentrations analyses The concentrations of Ca, Mg, and Sr in the solutions from the wet digestion of the sediment samples were measured according to the procedure described by Walkusz et al. (1992). After wet digesting the material and dissolving the dry residue in 0.1 mol HNO3 l−1, the samples were analyzed using an atomic absorption spectrometer (AAS; type 6800, Shimadzu, Japan). Calibration curves in the range of 0–8.0 µg ml−1 were used to quantify the concentrations. Quality control was based on the addition of standard solutions to the actual samples. Recovery was in the range of 99.1–100.3 %.
4. Results and Discussion 4.1. Sediment dating The SAR at station Kb1 was 0.14 cm year–1. The deepest layer of this sediment core (21–22 cm below the sediment surface) was dated to 1881–1886. The SAR at station Kb2 was higher, 0.21 cm year−1. At a depth of 8 cm, the activity concentration of 210Pbex reached 18.1 Bq kg−1, assigned to the supported radiolead (Fig. 2). The deepest layer (27–28 cm) was dated to 19131916. At the innermost station (Kb3), the SAR was too high to be determined by the radio-Pb method because a constant
Pb activity concentration of 14.3 Bq kg−1 was measured along
210
the entire profile. The high SAR at station Kb3 is likely due to the small distance to the GF, 6
which is an enormous source of mineral material. Streuff et al. (2015) recently showed that the minimum sedimentation rate after the Kongsvegen surge, in 1948, was 1.8 cm year−1. In our study we used the SAR value of 7 cm year−1, estimated from acoustic measurements performed in this region by Elverhoi et al. (1983).
Fig. 2. Profiles of 210Pb total activity concentrations at stations Kb1 and Kb2
4.2. Concentrations of total, organic, and inorganic carbon in surface sediments There were no significant spatial differences in the Ctot concentrations in the studied cores, as the results varied between 32.2 and 35.2 mg g−1 at station Kb1, 31.1 and 35.3 mg g−1 at Kb2, and 31.3 and 45.5 mg g−1 at station Kb3 (Fig. 3a). However, very clear patterns were observed when Corg and Cinorg concentrations were separated from the Ctot pool. Corg concentrations were lowest at station Kb3 (1.0 mg g−1) but increased along the fjord axis towards the FM (18.0 mg g−1 at station Kb1) (Fig. 3b); they were therefore close to the concentrations previously measured in the Spitsbergen fjords (Kuliński et al., 2014; Winkelmann and Knies, 2005; Włodarska-Kowalczuk et al., 2007; Zaborska et al., 2006). In contrast to the distribution of Corg, the highest Cinorg concentration (45.5 mg g−1) was at station Kb3, situated close to the GF, and the lowest (17.05 mg g−1) at Kb1, the outermost station (Fig. 3c). This indicates that glaciers are a source of carbonate inputs into the fjord, as previously reported for the nearby fjords Hornsund (Gorlich, 1986) and Tempelfjord (Forwick et al., 2010). The Corg and Cinorg concentrations in the vertical profiles differed significantly, especially at stations Kb1 and Kb2 (Fig. 3b, c, e, f). The most evident changes occurred in 7
sediment layers deposited during the last 40 years. Corg concentrations increased towards the sediment surface by 20–25 % (4.1 mg g−1 at Kb1 and 4.5 mg g−1 at Kb2, Fig. 3e). This was attributed to (i) increased Corg deposition in the bottom sediments of Kongsfiord in recent years and/or (ii) the gradual removal of Corg due to organic matter mineralization in the sediments. The latter implies that sediment surface layers contain labile organic matter that is mineralized over time. However, unlike Corg, Cinorg concentrations decreased in the sediment surface layers of the core segments. Moreover, the magnitude of these changes was comparable to those in Corg: 3.1 mg g−1 at station Kb1 and 5.7 mg g−1 at Kb2 (Fig. 3f). This decrease in Cinorg towards the sediment surface can be explained by a smaller carbonate load deposition in the sediments and/or the formation of soluble bicarbonates using CO2 originated in the course of organic matter mineralization.
8
Fig. 3. Vertical profiles of total (a, d), organic (b, e), and inorganic (c, f) carbon concentrations in the investigated sediment cores plotted against depth (a–c) and deposition year (d–f)
4.3. Calcium, magnesium, and strontium concentrations in surface sediments Sedimentary Ca concentrations differed significantly between sampling stations and generally decreased with increasing distance from the GF (Fig. 4a). The values at stations Kb3, Kb2, 9
and Kb1 were in the range of 51.5–63.9, 40.1–51.7, and 32.3–41.1 mg g−1, respectively. The Mg concentrations were quite similar at all stations: 18.4–22.0 mg g−1 at Kb1, 18.4–21.5 mg g−1 at Kb2, and 19.7–24.4 mg g−1 at Kb3 (Fig. 4b). The measured Sr concentrations were much lower, with ranges of 10.9–18.6, 17.5– 22.6, and 15.3–23.6 µg g−1 at Kb1, Kb2, and Kb3, respectively (Fig. 4c). Due to the relatively constant Mg concentrations along the fjord axis and the decreasing Ca concentrations towards the FM, the Mg:Ca molar ratios differed significantly: 0.6 ± 0.02 at station Kb3, 0.7 ± 0.04 at Kb2, and 0.9 ± 0.05 at Kb1 (Fig. 4d). There were no significant spatial differences in the Sr:Ca (Sr × 1000) molar ratios, as in the surface sediment layers at all stations the ratios were similar (0.18 ± 0.0003), while in the deeper layers they differed slightly and were lower (0.14 ± 0.02) at the station close to the GF (Kb3) than at the stations at the central (Kb2) and open (Kb1) parts of the fjord (0.19 ± 0.04) (Fig. 4e).
Fig. 4. Vertical profiles of calcium (a), magnesium (b), and strontium (c) concentrations and the molar ratios Mg:Ca (d) and Sr:Ca (Sr × 1000) (e), in the investigated sediment cores
4.4. Suspended particulate matter (SPM) as a source of sedimentary material. The extent and composition of material supplied to the bottom sediments of the fjord by glaciers and biota were determined in analyses of SPM samples. The results provide only a snapshot of the SPM distribution in mid-summer, a period when glacial melting and biological activity are highest. Our results confirmed that glaciers are an important source of SPM. The concentration at station Kb3 (10.2 ± 3.3 mg dm−3) was significantly higher than at stations Kb1 and Kb2 (Fig. 5a). A similar spatial distribution was determined for particulate 10
inorganic carbon (PIC) concentrations, which were very high at station Kb3 (0.19–0.75 mg dm-3) and significantly lower at stations Kb1 and Kb2 (0.01–0.07 mg dm-3) (Fig. 5c). Particulate organic carbon (POC) concentrations were comparable in samples collected across the entire fjord: 0.18 ± 0.05 mg dm−3 at station Kb3, 0.09 ± 0.06 mg dm−3 at Kb2, 0.12 ± 0.03 mg dm−3 at Kb1, and 0.12 ± 0.02 mg dm−3 at Kb0 (Fig. 5b). This suggested a similar magnitudes of primary production at stations Kb0–Kb2 whereas POC contributing to the SPM at station Kb3 reflected the specific characteristics of the SPM supplied by the glacier. Support for this conclusion comes from the percent contribution of POC to SPM, which increased from ~0.5% at the GF to nearly 28% at the FM. Primary productivity in the study area changes from 20 to 50 g m−2 year−1 in most of the fjord area to very low at the GF (Hop et al., 2002; Piwosz et al., 2009) are also consistent with our assignment of the POC in SPM.
Fig. 5. Concentrations [mg dm−3] of suspended particulate matter (a), particulate organic carbon (b), and particulate inorganic carbon (c) in water samples collected from four sampling stations: Kb3 (glacier front), Kb2, Kb1 (fjord mouth), and Kb0 (shelf) at four water layers: surface, subsurface (maximum fluorescence), subsurface (40 m), and bottom (5 m above the sea floor)
4.5. Provenience of Corg and Cinorg in sediments. The distribution of Corg and Cinorg presented in Fig. 3 and Fig. 5 suggests that Corg is mostly produced in fjord waters and Cinorg mostly in glaciers. Thus, the high concentrations of Cinorg at station Kb3 can be attributed to the abundant supply of carbonates from the Kronebreen glacier. A substantial proportion of the rocks in the region of Spitsbergen where the 11
Kongsbreen and the Kronebreen glaciers are located are limestone (Streuff, 2013). Moving glaciers scratch the bedrock and transport crushed rocks, including limestone, to the fjord (Bhushan et al., 2001; Hauck et al., 2012). However, somewhat lower but still high concentrations of carbonates were also determined in the sediments of station Kb1, located at the FM, and in the SPM collected from stations Kb0 and Kb1, both located far from the Kongsbreen and Kronebreen glaciers. Thus, particulate carbonates are either transported over long distances by the GF or/and are produced directly in the fjord. The latter possibility is best explained by the activity of calciferous organisms, whereas it is unlikely that the former is the sole source of carbonates, because both the SPM and the PIC concentrations were already fairly low at the Kb2 sampling station, with no further decrease towards the FM (Fig. 5). Differing distributions of both sedimentary Corg and Cinorg can be used to evaluate the origin of organic vs. inorganic carbon species. Allochthonous (glacier-supplied) and autochthonous (produced in situ) carbonates can be distinguished by assuming that carbonates are either glacial or biogenic in their origin and that organic matter is biogenically derived (Corg in glacier-supplied material is negligible). In the first step, Cinorg concentrations were normalized to Corg and plotted against Corg (Fig. 6), used here as an indicator of biological production. Cinorg:Corg ratios (10.3–35.1) were highest at station Kb3 (Fig. 6), which suggested a strong predominance of carbonates over Corg in the sediments and that they consisted of glacially supplied material. Cinorg:Corg ratios decreased gradually with increasing distance from the glacier. Both the lowest Cinorg:Corg ratios (0.93–1.48) and the least amount of variability therein were measured at the outermost station, Kb1. The dependence of the Cinorg:Corg ratios on Corg decreased exponentially with increasing Corg concentrations in the samples (Fig. 6). This pattern is typical of the mixing of two carbonate sources: a large, local source independent of Corg (referred to in the following as glacial carbonates) and a second source linked to biological activity. The presence of organic matter originating from biological activity can be used as a proxy of the biogenic carbonates produced in the fjord. Of note is that the plots of all the measured sediment samples deposited in different years and in the different regions of Kongsfjord represented by the cores yielded a single exponential curve (Fig. 6). Thus, the mixing of carbonates from these two sources can explain the distribution of carbonates in the fjord sediments. Interestingly, there was no variability in Cinorg to Corg ratios related to the age of the samples, which suggests that the mineralization of organic matter in the course of early diagenesis does not significantly alter Cinorg:Corg ratios, which instead depend only on the relative contributions of autochthonous and allochthonous sources in the 12
mixing process. While this does not exclude the mineralization of organic matter during the early stages of diagenesis in sediments, it does indicate that the mineralization of 1 mol of Corg must be accompanied by the dissolution of 1 mol of carbonates. This conclusion is in line with that of Pfeifer et al. (2002), who, based on empirical data, demonstrated that the molar ratio of carbonate dissolution and carbon dioxide release from Corg mineralization is roughly 1:1.
Fig. 6. Cinorg normalized to Corg is plotted against the Corg concentrations [mmol g−1] in the investigated Kongsfjord sediments
In samples with high Corg concentrations, the Cinorg:Corg ratios stabilized and did not decrease significantly with increasing Corg, as was the case with samples from Kb1 (Fig. 6). To maintain a constant Cinorg:Corg ratio in samples with different Corg concentration implies that any change of Corg must be accompanied by a change in Cinorg. This direct link between Cinorg and Corg suggests the biogenic character of the carbonates deposited in these samples. Thus, the lowest value of Cinorg:Corg from our dataset, recorded at the highest Corg concentration in sediments (0.93), is presumably characteristic of biogenic carbonates produced in the fjord. Multiplying this value by the Corg concentration in each sediment sample provides an estimate of the fraction of Cinorg that originated from in situ production in the fjord (biogenic Cinorg). Glacial Cinorg can then be calculated as the difference between total Cinorg and biogenic Cinorg.
13
However, it must be taken into account that a certain fraction of Corg in the Kongsfiord sediments may be allochthonous, which would influence the accuracy of the results presented above. Nonetheless, assuming that the spatial distribution of glacial Corg follows that of glacial Cinorg, then glaciers are not an important source of Corg in the surficial sediments and the organic matter buried in Kongsfjord’s bottom sediments, especially in the outer part of the fjord, is therefore mostly autochthonous. While some fraction of terrigenous Corg may be supplied by rivers as relatively fresh organic matter, this load would be a minor one compared to the autochthonous organic matter load originating from primary production (Kuliński et al., 2014). Previous reports (Forwick et al., 2010; Hop et al., 2002; Kuliński et al., 2014; Zajaczkowski, 2008) indicated that the activity of tidal glaciers cause not only gradients in salinity but also in the quality and quantity of sedimentary organic and inorganic matter. This accounts for the steep gradients in the concentrations of sedimentary organic carbon, which decrease sharply towards the GF due to the depletion of organic matter in the weathered mineral material delivered by the glaciers (Bourgeois et al., 2016; Kuliński et al., 2014; Kumar et al., 2016). Bourgeois et al. (2016) reported a low POC content in the vicinity of the glacier, accompanied by surprisingly high values of the δ13C signatures of organic carbon (ranging from −20 to – 22‰), suggestive of in situ production as a source of organic matter. A low supply of terrigenous organic matter was also determined by Kuliński et al. (2014), who, based on POC concentrations and the freshwater volume supplied to the fjord, estimated that the annual terrestrial (including glaciers and rivers) POC load constitutes only 5–10 % of the total carbon load delivered from primary production in the fjord. Kumar et al. (2016) also measured high δ13Corg values close to the GF; however, by assuming that terrestrial OM in this region is relatively isotopically heavy, they suggested, that organic carbon originating from allochthonous sources predominates at the GF over carbon from autochthonous sources. While the assumption that terrestrial organic carbon is isotopically heavy needs to be validated, the approach used in this study is valid irrespective of the isotopic signatures of Corg in the investigated system.
4.6. Vertical distributions of glacial and biogenic Cinorg. Having dated the investigated sediment cores we were able to assess the time dependence of the glacial Cinorg concentrations at the outer and central stations (Fig. 7a). The systematic shifts in the profiles were indicative of sedimentary processes related to, e.g., glacial surging. 14
Most of the glaciers in the Kongsfjord catchment are of the surge-type (Dowdeswell et al., 1991; Hagen, 1993; Ottesen and Dowdeswell, 2006; Ottesen et al., 2008), such that their activity is characterized by a succession of active and passive phases. During active phases, the ice front advances rapidly or, at least, stagnates, while passive phases are characterized by slow glacial retreat, which can take from 50 to 500 years. Three glacier surges have been documented at our study site: Kronebreen in 1869, Kongsvegen in 1948, and Blomstrandbreen in 1960 (Dowdeswell et al., 1991; Hagen, 1993; Ottesen and Dowdeswell, 2006; Ottesen et al., 2008; Plassen et al., 2004). Thus, the investigated sediment cores, whose deepest sediment layer at Kb1 and Kb2 dated to 1881–1886 and 1913–1916, respectively, included two surges. The absence of significant changes in the glacial Cinorg concentrations profile coinciding with the Kongsvegen surge in 1948 was probably due to the substantial distance between this glacier and the outer (Kb1) and central (Kb2) stations. At station Kb2, there was a visible change in glacial Cinorg concentrations during the period 1960–1970, which was probably related either to the changes in the position of the Kronebreen GF or, albeit less likely, to the Blomstrandbreen surge in 1960. There were significant changes in both glacial and biogenic Cinorg during the last 25–35 years (Fig. 7 a, b) when the increase in biogenic Cinorg was accompanied by a substantial decrease in glacial Cinorg. The higher concentrations of biogenic Cinorg in the uppermost part of the sediment cores were attributable to increased biological activity in the water column, (Pfeifer et al., 2002), whereas the drop in glacial Cinorg was most likely due to changes in glacial activity, including the occurrence of a retreat.
15
Fig. 7. Concentrations of glacial (a) and biogenic (b) Cinorg plotted against the deposition year for the investigated cores
4.7. Characteristics of glacial vs. biogenic carbonates. The charge balance performed in all samples, in which Ca, Mg, and Sr were measured in parallel with Cinorg, suggested that carbonates are the sole carrier of these metals in sediments. Thus, for all samples, the molar concentration of Cinorg was almost entirely balanced by the sum of the Ca and Mg molar concentrations. By contrast, the contribution of Sr was less important, as its concentration was roughly three orders of magnitude lower than that of Ca or Mg (Fig. 4). The maximum deviation from the equality of the molar concentrations of the sum of the concentrations of Ca and Mg cations and carbonates was <5%, a difference not exceeding analytical uncertainty. Since the distribution of Mg and Sr concentrations was relatively constant along the fjord axis and the concentrations of Ca decreased significantly towards the FM (Fig. 4), the molar ratios of Mg:Ca and Sr:Ca could be used to characterize the chemical composition of carbonates originating from different sources and to perhaps distinguish between biogenic and glacial 16
carbonates. A plot of these ratios against the concentrations of biogenic and glacial Cinorg concentrations revealed the dependences shown in Fig. 8. The intercepts of the linear regressions at the null concentrations of the respective carbonates indicated the ratios that characterized the glacial and biogenic end-members. The Mg:Ca and Sr:Ca ratios for glacial carbonates were 0.56 (Fig. 8a) and 0.00015 (Fig. 8b) whereas for biogenic carbonates they were 0.94 (Fig. 8c) and 0.00020 (Fig. 8d), respectively. Thus, both the glacial and the biogenic carbonates deposited in the Kongsfiord sediments are relatively rich in Mg (Long et al., 2014). These characteristic ratios could be used to identify the species responsible for the biogenic carbonate supply in the Kongfiord ecosystem. Although beyond the scope of our study, a literature screening indicated that Ostracoda, Echinodermata, red corals, red coralline algae, and calcareous serpulid worms (Farmer et al., 2012; Long et al., 2014; Veis, 2011), among others, produce large amounts of magnesium calcites.
Fig. 8. Glacial (a, b) and biogenic (c, d) end-members of the Mg:Ca (a, c) and Sr:Ca (b, d) ratios as indicated by the intercepts of the linear functions presented in the plots
5. Conclusions
17
This study tested the hypothesis that sediments of the Svalbard fjords contain important amounts of biogenic and terrestrial sources of Cinorg. Total sedimentary carbon concentrations along the axis of the Kongsfjord, from the FM (station Kb1), to an intermediate area (station Kb2), and finally to the GF (station Kb3) were in the range of 30–45 mg (g dry sediment)−1. The contribution of sedimentary Cinorg (carbonates) was determined to be substantial whereas that of Corg was much smaller, as evidenced by the composition of total carbon: 1–18 mg Corg (g dry sediment)−1 and 17–43 mg Cinorg (g dry sediment)−1. The horizontal gradient of Corg concentrations, which ranged from 1–2 mg g−1 at Kb3 to 13– 18 mg g−1 at Kb1 could be attributed to the supply of organic-matter-depleted, glacial mineral material and to the limited contribution from the biota, with both factors having the largest effect close to the GF. The horizontal gradient of Cinorg (Kb3: 29-43 mg g−1 and Kb1: 17–20 mg g-1) was most likely caused by the large supply of terrigenous carbonates from the GF, and the lower but still substantial supply of biogenic carbonates produced in the central and outer parts of the fjord.
`
Using the dependences of Cinorg:Corg vs Corg and the concentrations of sedimentary Ca, Mg, and Sr, we developed a method, that allows quantification of the contributions of glacial and biogenic carbonates to total carbonates. Glaciers are an important source of carbonates in the Kongsfiord, but our results for both sediments and SPM suggested that glacial material is mostly deposited in the vicinity of the GF, such that the carbonates at Kb1 are almost exclusively biogenic in origin and those at Kb2 a mixture of comparable amounts from the two sources. The vertical zonation of glacial Cinorg concentrations at station Kb2 may be the product of either glacial surges or changes in glacial activity. The less pronounced vertical zonation at station Kb1 was most likely a function of the large distance to the GFs of Kronebreen and Kongsvegen. In addition, we determined Mg:Ca and Sr:Ca ratios for glacial carbonates of 0.56 and 0.00015, and for biogenic carbonates of 0.94 and 0.00020, respectively. Our study demonstrates that in quantifications of the carbon sinks in the Kongsfjord sediments and in those of the other Svalbard fjords, carbonates need to be considered and their source (glacial vs. biogenic) identified.
18
Acknowledgement Financial support from the Polish National Science Centre (grant no. 2015/19/N/ST10/01652) and the statutory activities of the Institute of Oceanology (Sopot, Poland) are acknowledged. Katarzyna Koziorowska’s research for this paper was supported by the Centre for Polar Studies, KNOW – Leading National Research Centre, Sosnowiec, Poland. We thank A. Iglikowska for her valuable comments on the CO2 system and J. Walkusz-Miotk for her assistance in the AAS analyses. References Andruleit, H., Freiwald, A., Schafer, P., 1996. Bioclastic carbonate sediments on the southwestern Svalbard shelf. Marine Geology 134, 163-182. Beszczyńska-Moller, A., Węsławski, J.M., Walczowski, W., Zajączkowski, M., 1997. Estimation of glacial meltwater discharge into Svalbard coastal waters. Oceanology 39, 289-298. Bhushan, R., Dutta, K., Somayajulu, B.L.K., 2001. Concentrations and burial fluxes of organic and inorganic carbon on the eastern margins of the Arabian Sea. Marine Geology 178, 95-113. Bourgeois, S., Kerherve, P., Calleja, M.L., Many, G., Morata, N., 2016. Glacier inputs influence organic matter composition and prokaryotic distribution in a high Arctic fjord (Kongsfjorden, Svalbard). Journal of Marine Systems 164, 112-127. Carpenter, S.J., Lohmann, K.C., 1992. Sr/Mg ratios of modern narine calcite - empirical indicators of ocean chemistry and precipitation rate. Geochimica Et Cosmochimica Acta 56, 1837-1849. Carroll, J., Zaborska, A., Papucci, C., Schirone, A., Carroll, M.L., Pempkowiak, J., 2008. Accumulation of organic carbon in western Barents Sea sediments. Deep-Sea Research Part Ii-Topical Studies in Oceanography 55, 2361-2371. Dowdeswell, J.A., Hamilton, G.S., Hagen, J.O., 1991. The duration of the active phase on surge-type-glaciers-contrasts between Svalbard and other regions. Journal of Glaciology 37, 388-400. Elverhoi, A., Lonne, O., Seland, R., 1983. Glacimarine sedimentation in a modern fjord environment. Polar Research 1, 127-149.
19
Farmer, J.R., Cronin, T.M., Dwyer, G.S., 2012. Ostracode Mg/Ca paleothermometry in the North Atlantic and Arctic oceans: Evaluation of a carbonate ion effect. Paleoceanography 27. Faust, J.C., Knies, J., Slagstad, T., Vogt, C., Milzer, G., Giraudeau, J., 2014. Geochemical composition of Trondheimsfjord surface sediments: Sources and spatial variability of marine and terrigenous components. Continental Shelf Research 88, 61-71. Forwick, M., Vorren, T.O., Hald, M., Korsun, S., Roh, Y., Vogt, C., Yoo, K.C., 2010. Spatial and temporal influence of glaciers and rivers on the sedimentary environment in Sassenfjorden and Tempelfjorden, Spitsbergen. Fjord Systems and Archives 344, 163193. Gorlich, K., 1986. Glacimarine sedimentation of muds in Hornsund Fjord, Spitsbergen. Annale Societatis Geologorum Poloniae 56, 433-477. Gorlich, K., Weslawski, J.M., Zajaczkowski, M., 1987. Suspension settling effect on macrobenthos biomass distribution in the Hornsund Fjord, Spitsbergen. Polar Research 5, 175-192. Hagen, J., 1993. Glacier atlas of Svalbard and Jan Mayen. Norsk Polarinstitutt Middelelser. Hauck, J., Gerdes, D., Hillenbrand, C., Hoppema, M., Kuhn, G., Nehrke, G., Voelker, C., Wolf-Gladrow, D.A., 2012. Distribution and mineralogy of carbonate sediments on Antarctic shelves. Journal of Marine Systems 90, 77-87. Hop, H., Pearson, T., Hegseth, E.N., Kovacs, K.M., Wiencke, C., Kwaśniewski, S., Eiane, K., Mehlum, F., Gulliksen, B., Włodarska-Kowalczuk, M., Lydersen, C., Węsławski, J.M., Cochrane, S., Gabrielsen, G.W., Leakey, R.J.G., Lonne, O.J., Zajączkowski, M., FalkPetersen, S., Kendall, M., Wangberg, S.A., Bischof, K., Voronkov, A.Y., Kovaltchouk, N.A., Wiktor, J., Poltermann, M., di Prisco, G., Papucci, C., Gerland, S., 2002. The marine ecosystem of Kongsfjorden, Svalbard. Polar Research 21, 167-208. Hulth, S., Hall, P.O.J., Blackburn, T.H., Landen, A., 1996. Arctic sediments (Svalbard): Pore water and solid phase distributions of C, N, P and Si. Polar Biology 16, 447-462. Kędra, M., Kuliński, K., Walkusz, W., Legeżyńska, J., 2012. The shallow benthic food web structure in the high Arctic does not follow seasonal changes in the surrounding environment. Estuarine Coastal and Shelf Science 114, 183-191. Koziorowska, K., Kuliński, K., Pempkowiak, J., 2016. Sedimentary organic matter in two Spitsbergen fjords: terrestrial and marine contribution based on carbon and nitrogen contents and stable isotopes composition. Continental Shelf Research 113, 38-46.
20
Kuliński, K., Kędra, M., Legeżyńska, J., Głuchowska, M., Zaborska, A., 2014. Particulate organic matter sinks and sources in high Arctic fjord. Journal of Marine Systems 139, 27-37. Kumar, V., Tiwari, M., Nagoji, S., Tripathi, S., 2016. Evidence of Anomalously Low delta C13 of Marine Organic Matter in an Arctic Fjord. Scientific Reports 6. Long, X., Ma, Y.R., Qi, L.M., 2014. Biogenic and synthetic high magnesium calcite - A review. Journal of Structural Biology 185, 1-14. Morse, J.W., 2003. Formation and diagenesis of carbonate sediments, In Treatise on Geochemistry. Elsevier, Amsterdam, pp. 67-85. Ottesen, D., Dowdeswell, J.A., 2006. Assemblages of submarine landforms produced by tidewater glaciers in Svalbard. Journal of Geophysical Research-Earth Surface 111. Ottesen, D., Dowdeswell, J.A., Benn, D.I., Kristensen, L., Christiansen, H.H., Christensen, O., Hansen, L., Lebesbye, E., Forwick, M., Vorren, T.O., 2008. Submarine landforms characteristic of glacier surges in two Spitsbergen fjords. Quaternary Science Reviews 27, 1583-1599. Pempkowiak, J., 1991. Enrichment factors of heavy-metals in the Southern Baltic surface sediments dated with Pb210 and Cs137. Environment International 17, 421-428. Pfeifer, K., Hensen, C., Adler, M., Wenzhofer, F., Weber, B., Schulz, H.D., 2002. Modeling of subsurface calcite dissolution, including the respiration and reoxidation processes of marine sediments in the region of equatorial upwelling off Gabon. Geochimica Et Cosmochimica Acta 66, 4247-4259. Piwosz, K., Walkusz, W., Hapter, R., Wieczorek, P., Hop, H., Wiktor, J., 2009. Comparison of productivity and phytoplankton in a warm (Kongsfjorden) and a cold (Hornsund) Spitsbergen fjord in mid-summer 2002. Polar Biology 32, 549-559. Plassen, L., Vorren, T.O., Forwick, M., 2004. Integrated acoustic and coring investigation of glacigenic deposits in Spitsbergen fjords. Polar Research 23, 89-110. Ries, J.B., 2006. Mg fractionation in crustose coralline algae: Geochemical, biological, and sedimentological implications of secular variation in the Mg/Ca ratio of seawater. Geochimica Et Cosmochimica Acta 70, 891-900. Smith, R.W., Bianchi, T.S., Allison, M., Savage, C., Galy, V., 2015. High rates of organic carbon burial in fjord sediments globally. Nature Geoscience 8, 450-U446. Stein, R., Macdonald, R.W., 2004. The organic carbon cycle in Arctic Ocean. Springer. Streuff, K., 2013. Landform assemblages in inner Kongsfjorden, Svalbard: evidence of recent glacial (surge) activity. University of Tromso. 21
Streuff, K., Forwick, M., Szczuciński, W., Andreassen, K., Cofaigh, C.O., 2015. Submarine landform assemblages and sedimentary processed related to glacier surging in Kongsfjorden, Svalbard. Arktos 1: 14. Svendsen, H., Beszczynska-Moller, A., Hagen, J.O., Lefauconnier, B., Tverberg, V., Gerland, S., Orbaek, J.B., Bischof, K., Papucci, C., Zajaczkowski, M., Azzolini, R., Bruland, O., Wiencke, C., Winther, J.G., Dallmann, W., 2002. The physical environment of Kongsfjorden-Krossfjorden, an Arctic fjord system in Svalbard. Polar Research 21, 133166. Szczuciński, W., Zajączkowski, M., Scholten, J., 2009. Sediment accumulation rates in subpolar fjords - Impact of post-Little Ice Age glaciers retreat, Billefjorden, Svalbard. Estuarine Coastal and Shelf Science 85, 345-356. Trusel, L.D., Powell, R.D., Cumpston, R.M., Brigham-Grette, J., 2010. Modern glacimarine processes and potential future behaviour of Kronebreen and Kongsvegen polythermal tidewater glaciers, Kongsfjorden, Svalbard. Fjord Systems and Archives 344, 89-102. Veis, A., 2011. Organic Matrix-related mineralization of sea urchin spicules, spines, test and teeth. Frontiers in Bioscience-Landmark 16, 2540-2560. Walkusz, J., Roman, S., Pempkowiak, J., 1992. Contamination of the Southern Baltic surface sediments with heavy metals. Bull. Sea Fis. Inst. 1 (125), 33-37. Winkelmann, D., Knies, J., 2005. Recent distribution and accumulation of organic carbon on the continental margin west off Spitsbergen. Geochemistry Geophysics Geosystems 6. Włodarska-Kowalczuk, M., Pearson, T.H., 2004. Soft-bottom macrobenthic faunal associations and factors affecting species distributions in an Arctic glacial fjord (Kongsfjord, Spitsbergen). Polar Biology 27, 155-167. Włodarska-Kowalczuk, M., Szymelfenig, M., Zajączkowski, M., 2007. Dynamic sedimentary environments of an Arctic glacier-fed river estuary (Adventfjorden, Svalbard). II: Meioand macrobenthic fauna. Estuarine Coastal and Shelf Science 74, 274-284. Zaborska, A., Carroll, J., Papucci, C., Pempkowiak, J., 2007. Intercomparison of alpha and gamma spectrometry techniques used in Pb-210 geochronology. Journal of Environmental Radioactivity 93, 38-50. Zaborska, A., Pempkowiak, J., Papucci, C., 2006. Some sediment characteristic and sedimentation rates in an Arctic fjord (Kongsfjorden, Svalbard). Annales Environment Protection 8, 79-96. Zajaczkowski, M., 2008. Sediment supply and fluxes in glacial and outwash fjords, Kongsfjorden and Adventfjorden, Svalbard. Polish Polar Research 29, 59-72. 22
Highlights
Stratified bottom sediments and SPM were collected in Kongsfjord, Svalbard Corg, Cinorg, Ca, Mg and Sr were measured in the collected material Terrestrial vs marine pools of both carbon species are distinguished Sedimentary Corg is mostly marine, while glaciers are only its minor source Cinorg close to glacier is terrigenous while at the fjord mouth- is mostly biogenic
23
PII: DOI: Reference:
S0278-4343(17)30169-3 http://dx.doi.org/10.1016/j.csr.2017.08.023 CSR3669
To appear in: Continental Shelf Research Received date: 29 March 2017 Revised date: 29 August 2017 Accepted date: 30 August 2017 Cite this article as: Katarzyna Koziorowska, Karol Kuliński and Janusz Pempkowiak, Distribution and origin of inorganic and organic carbon in the sediments of Kongsfjord, northwest Spitsbergen, European Arctic, Continental Shelf Research, http://dx.doi.org/10.1016/j.csr.2017.08.023 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Distribution and origin of inorganic and organic carbon in the sediments of Kongsfjord, northwest Spitsbergen, European Arctic Katarzyna Koziorowska*, Karol Kuliński, Janusz Pempkowiak Institute of Oceanology Polish Academy of Sciences, ul. Powstańców Warszawy 55, Sopot, Poland *Corresponding author, e-mail: [email protected] Abstract Sedimentary organic carbon in the Arctic, including the continental shelf and fjords, has been relatively well investigated, whereas much less is known about sedimentary inorganic carbon (carbonates) in fjords. The distribution and provenience of both sedimentary organic and inorganic carbon in a high-Arctic fjord (Kongsfjord, 79ºN) was the subject of this study. Stratified bottom sediments (cores) and suspended particulate matter (SPM) were analyzed for total (Ctot), organic (Corg), and inorganic (Cinorg) carbon as well as calcium, magnesium, and strontium. The sediments were dated using the 210Pb method. Sedimentation rates ranged from 0.14 cm (fjord mouth, FM) to several cm (close to the glacier front, GF) year−1. Sedimentary Corg concentrations were higher at the FM (~20 mg g−1 dry sediment) than at the GF (~1 mg g−1), while concentrations of Cinorg were lower at the FM (16.8 mg g−1) than at the GF (45 mg g−1). SPM concentrations were highest, and Cinorg most abundant at the GF. The data suggest that Corg is mostly produced in situ, with glaciers serving as only a minor source. The Cinorg to Corg ratios, Ca, Mg, and Sr concentrations, and the molar ratios of Mg:Ca and Sr:Ca together indicated that carbonates close to the GF are of terrigenous origin and those at the FM almost exclusively biogenic. Carbonates originating from these two sources differ in their composition. The Mg:Ca and Sr:Ca molar ratios were 0.56 and 0.00015 for glacial carbonates and 0.94 and 0.00020 for biogenic carbonates. Key words: high-Arctic fjord; suspended matter; sedimentary organic matter; sedimentary carbonates; provenience; Mg:Ca ratio
1. Introduction
1
Accurate determinations of atmospheric carbon dioxide concentrations are of utmost importance for predicting the impacts of climate warming and ocean acidification. Marine and lacustrine sediments constitute a long-term natural carbon sink and thus indirectly condition carbon dioxide in the atmosphere. Accordingly, quantifying the distribution of carbon in sediments will contribute to a better understanding of carbon budgets. The two major species of carbon in the environment are organic carbon (Corg) and inorganic carbon (Cinorg). The material flux reaching marine bottom sediments comprises a mixture of the two forms. Both may originate from internal biological activity or/and be transported to the sea from the land. Terrestrial sources are also the result of biological activities. Because these activities occur under different environmental conditions, the properties of the resulting carbon-bearing matter that makes up sediments will differ as well (Bhushan et al., 2001). Until the 1990s, most information regarding carbon deposition and preservation in the surficial sediments was limited, derived from data on the carbon concentrations in samples collected with a grab sampler (Gorlich et al., 1987; Hulth et al., 1996); consequently, the age span represented by the sediments was unknown. With recent approaches to measuring dated sediment cores (Szczuciński et al., 2009), much has been learned about the distribution and provenience of Corg. In the Arctic, analyses of sediment cores have shown that carbon abundances are much greater in the sediments of Arctic fjords than in the Arctic Ocean. The sediments in fjords contain up to 40 mg g−1 Corg and even 70-100 % is of terrestrial provenience (Koziorowska et al., 2016). However, stable carbon isotopes may not be reliable markers of Corg provenience, because the typical pattern of a much lower
13
C abundance in
terrestrial vs. marine organic matter does not apply in the Spitsbergen fjords (Kumar et al., 2016). Moreover, sediments from fjords deliver much more carbon to carbon cycling than would be anticipated based on their areal contribution to the Arctic Ocean (Stein and Macdonald, 2004). This reflects the fact that fjords, with their high sedimentary Corg concentrations and sediment accumulation rates (SARs), have the highest area-normalized Corg burial rates (Smith et al., 2015). In addition, Corg concentrations have been often reported together with total nitrogen (Ntot) concentrations, Corg to Ntot molar ratios, and δ15Ntot and δ13Corg. By contrast, Cinorg (carbonates) has been almost totally neglected (Carroll et al., 2008; Kędra et al., 2012; Koziorowska et al., 2016; Kuliński et al., 2014; Zaborska et al., 2006), such that little is known about its distribution and provenience in bottom sediments of the Arctic Ocean and fjords.
2
Biogenic carbonates mostly consist of calcium carbonates and mixed calcium/strontium and calcium/magnesium carbonates (Andruleit et al., 1996; Carpenter and Lohmann, 1992; Ries, 2006). They can be delivered to marine sediments but also from land, the latter as a result of limestone weathering (Bhushan et al., 2001; Hauck et al., 2012). Moreover, carbonates can precipitate due to excess dissolution, most often during the mixing of fresh, carbonate-rich water with saline water (seawater) (Morse, 2003). The current biological activity of the fjords and Spitsbergen shelf as a source of carbonates to the bottom sediments has been described (Andruleit et al., 1996; Faust et al., 2014) whereas the Cinorg distribution in the sediments of the fjords has yet to be investigated. In the immediate vicinity of glaciers, primary production in the water column and the activities of sediment-dwelling organisms are very limited (Gorlich et al., 1987; Stein and Macdonald, 2004; Włodarska-Kowalczuk et al., 2007). Thus, neither marine Corg nor biogenic carbonates likely contribute to the bottom sediments of these regions. The large carbonate concentrations in the immediate vicinity of glaciers must therefore result from deposition of the weathering products of rocks, primarily limestone, (Trusel et al., 2010). However, there are no data on the contribution of weathered material in sediments. This study sought to fill this knowledge gap by measuring the organic and, especially, the inorganic carbon distribution in the sediments of a Spitsbergen fjord. In addition, given the different characteristics of autochthonous and allochthonous carbon pools, we traced the provenience of the identified carbon species. Specifically, we examined the distribution of Corg and Cinorg (carbonates) in surficial sediments in the Kongsfjord, a high Arctic (79ºN) fjord. Both vertical and horizontal Corg and Cinorg gradients were investigated by analyzing 210
Pb-dated sediment cores collected in a transect from the fronts of tidal glaciers (GFs) to the
mouth of the fjord (FM). Carbonates were characterized with respect to their concentrations and their molar ratios with calcium (Ca), magnesium (Mg), and strontium (Sr). The data were also used to distinguish between allochthonous (terrestrial) and autochthonous pools of inorganic and organic carbon.
2. Study area Kongsfjord is located on the northwestern coast of Spitsbergen (the main island of the Svalbard Archipelago). It is a relatively small fjord (26 km long and 4–10 km wide), with a wide opening to the ocean. A chain of islands (Lovénøyane) divides the Kongsfjord into two 3
parts. The outer basin, with the average depth of 200–300 m and only one large island (Blomstrandøya), is strongly affected by the relatively warm and saline North-Atlantic-type waters (temperature >2°C, salinity of 35 PSU) transported by the West Spitsbergen Current. The inner part (average depth of 50–60 m) is influenced by four tidewater glaciers: Kronebreen (the largest and most active), Kongsvegen, Conwaybreen, and Blomstrandbreen (Svendsen et al., 2002). Together, these glaciers annually supply ~0.33 km3 of fresh water to the fjord (Beszczyńska-Moller et al., 1997). Kongsbreen is retreating at a rate of up to 0.5 km per year and its meltwater annually supplies the fjord with ~2.6 × 105 m3 of mineral material. The geological cover of the catchment area of these four tidewater glaciers is mainly formed by moraines and marine shore deposits composed of quartz and polymict conglomerates, limestones, dolomites, marl, gypsum, anhydrite and carbonate breccia, siliceous shales, cherts, and sandstones (Streuff, 2013). The large supply of freshwater from ice calving and the melting of snow and ice in summer causes relatively high SARs. These vary from 20 000 g m−2 year−1 in the vicinity of the front of the glacier (inner part), to 1800–3800 g m−2 year−1 in the central part, and to ~200 g m−2 year−1 in the outer fjord (Zajaczkowski, 2008). Most surface sediments (below a water depth of 30–40 m) are composed of fairly uniform mud (Włodarska-Kowalczuk and Pearson, 2004). Despite numerous studies on the structures and functioning of ecosystems, little is known about annual primary production in Kongsfjord. According to the available literature data, it is lower than at other fjords in this region, with similar rates of 20 –40 g C m−2 year−1 (Piwosz et al., 2009) and 35–50 g C m−2 year−1 (Hop et al., 2002) reported in different studies.
3. Experimental 3.1. Sampling Sediment cores and suspended material from the water column were collected during the R/V Oceania cruise in July 2015. A Niemisto gravity corer was used to collect sediment cores at three stations located in the outer, central, and inner parts of the fjord (Fig. 1: Kb1, Kb2, and Kb3, respectively). Immediately after their collection, the cores were sliced into 10-mm thick layers and frozen (−20°C) until the analyses. SPM was sampled at four stations (Fig. 1: Kb0, Kb1, Kb2, and Kb3) by filtering the samples collected from four different depths through precombusted Whatman GF/F glass-fiber filters. The depths corresponded to the surface, the maximum fluorescence layer, the subsurface (40 m), and the bottom water (~2 m above the seafloor). Filters with the separated SPM were immediately frozen at −80°C. 4
All analyses were carried out in the Marine Biogeochemistry Laboratory of the Institute of Oceanology, Polish Academy of Sciences, Sopot, Poland.
Fig. 1. Locations of sampling stations in the Kongsfjord
3.2. Corg and Cinorg analyses Ctot and Corg concentrations of the samples were analyzed in an elemental analyzer (Flash EA 1112 series) combined with an isotopic ratio mass spectrometer (Delta V Advantage (Thermo Electron Corp., Germany), according to the procedure described by Kuliński et al. (2014). Thirty mg (0.001 mg accuracy) of freeze-dried and homogenized material was weighed into silver capsules. Samples used to determine the Corg concentration were additionally soaked in 2M HCl to remove carbonates and then dried at 60°C for 24 h (the procedure was repeated until a constant weight was reached) prior to the analysis. Quantitative measurements were calibrated against analyses of certified reference materials (Fluβsediment) provided by HEKAtech GmbH (Germany). The precision (expressed as the relative standard deviation) was better than 1.6% (n=5) for Ctot measurements and 1.4% for Corg measurements. Cinorg concentrations were obtained by subtracting the Corg concentration from that of Ctot. The carbon concentrations are expressed as mg g−1 and mmol g−1 (12 mg = 1 mmol). 3.3. 210Pb analyses
5
The
Pb method was used to calculate annual SARs (in cm year−1). The rates were
210
determined from profiles of excess
210
Pb activity (210Pbex = total
210
Pb−supported
210
Pb) vs.
the porosity-corrected sediment depth and calculated by assuming a secular equilibrium between
210
Po and
210
Pb. A linear relation between log Pbex and depth in the analyzed cores
was assumed to calculate SAR using a least squares fit (Pempkowiak, 1991; Zaborska et al., 2007). Details of the radiochemical
210
Po activity measurements were described previously
(Pempkowiak, 1991; Zaborska et al., 2007). Briefly, ~200 mg of sediment were spiked with the 209Po chemical yield tracer and digested using 2 ml of concentrated perchloric acid and 3 ml of 12 M hydrofluoric acid.
210
Po was deposited spontaneously on a silver disk and then
counted for at least 24 h in a multi-channel analyzer (Canberra) equipped with a Si/Li detector.
210
Po activity concentrations were calculated by assuming equal chemical recovery
of the spiked (209Po) and sedimentary (210Po) radio-polonium activity. Blanks and reference materials (IAEA 300 and 326) were measured for quality control. Both the recovery and the precision of the determinations were highly satisfactory (Koziorowska et al., 2016).
3.4 Metal concentrations analyses The concentrations of Ca, Mg, and Sr in the solutions from the wet digestion of the sediment samples were measured according to the procedure described by Walkusz et al. (1992). After wet digesting the material and dissolving the dry residue in 0.1 mol HNO3 l−1, the samples were analyzed using an atomic absorption spectrometer (AAS; type 6800, Shimadzu, Japan). Calibration curves in the range of 0–8.0 µg ml−1 were used to quantify the concentrations. Quality control was based on the addition of standard solutions to the actual samples. Recovery was in the range of 99.1–100.3 %.
4. Results and Discussion 4.1. Sediment dating The SAR at station Kb1 was 0.14 cm year–1. The deepest layer of this sediment core (21–22 cm below the sediment surface) was dated to 1881–1886. The SAR at station Kb2 was higher, 0.21 cm year−1. At a depth of 8 cm, the activity concentration of 210Pbex reached 18.1 Bq kg−1, assigned to the supported radiolead (Fig. 2). The deepest layer (27–28 cm) was dated to 19131916. At the innermost station (Kb3), the SAR was too high to be determined by the radio-Pb method because a constant
Pb activity concentration of 14.3 Bq kg−1 was measured along
210
the entire profile. The high SAR at station Kb3 is likely due to the small distance to the GF, 6
which is an enormous source of mineral material. Streuff et al. (2015) recently showed that the minimum sedimentation rate after the Kongsvegen surge, in 1948, was 1.8 cm year−1. In our study we used the SAR value of 7 cm year−1, estimated from acoustic measurements performed in this region by Elverhoi et al. (1983).
Fig. 2. Profiles of 210Pb total activity concentrations at stations Kb1 and Kb2
4.2. Concentrations of total, organic, and inorganic carbon in surface sediments There were no significant spatial differences in the Ctot concentrations in the studied cores, as the results varied between 32.2 and 35.2 mg g−1 at station Kb1, 31.1 and 35.3 mg g−1 at Kb2, and 31.3 and 45.5 mg g−1 at station Kb3 (Fig. 3a). However, very clear patterns were observed when Corg and Cinorg concentrations were separated from the Ctot pool. Corg concentrations were lowest at station Kb3 (1.0 mg g−1) but increased along the fjord axis towards the FM (18.0 mg g−1 at station Kb1) (Fig. 3b); they were therefore close to the concentrations previously measured in the Spitsbergen fjords (Kuliński et al., 2014; Winkelmann and Knies, 2005; Włodarska-Kowalczuk et al., 2007; Zaborska et al., 2006). In contrast to the distribution of Corg, the highest Cinorg concentration (45.5 mg g−1) was at station Kb3, situated close to the GF, and the lowest (17.05 mg g−1) at Kb1, the outermost station (Fig. 3c). This indicates that glaciers are a source of carbonate inputs into the fjord, as previously reported for the nearby fjords Hornsund (Gorlich, 1986) and Tempelfjord (Forwick et al., 2010). The Corg and Cinorg concentrations in the vertical profiles differed significantly, especially at stations Kb1 and Kb2 (Fig. 3b, c, e, f). The most evident changes occurred in 7
sediment layers deposited during the last 40 years. Corg concentrations increased towards the sediment surface by 20–25 % (4.1 mg g−1 at Kb1 and 4.5 mg g−1 at Kb2, Fig. 3e). This was attributed to (i) increased Corg deposition in the bottom sediments of Kongsfiord in recent years and/or (ii) the gradual removal of Corg due to organic matter mineralization in the sediments. The latter implies that sediment surface layers contain labile organic matter that is mineralized over time. However, unlike Corg, Cinorg concentrations decreased in the sediment surface layers of the core segments. Moreover, the magnitude of these changes was comparable to those in Corg: 3.1 mg g−1 at station Kb1 and 5.7 mg g−1 at Kb2 (Fig. 3f). This decrease in Cinorg towards the sediment surface can be explained by a smaller carbonate load deposition in the sediments and/or the formation of soluble bicarbonates using CO2 originated in the course of organic matter mineralization.
8
Fig. 3. Vertical profiles of total (a, d), organic (b, e), and inorganic (c, f) carbon concentrations in the investigated sediment cores plotted against depth (a–c) and deposition year (d–f)
4.3. Calcium, magnesium, and strontium concentrations in surface sediments Sedimentary Ca concentrations differed significantly between sampling stations and generally decreased with increasing distance from the GF (Fig. 4a). The values at stations Kb3, Kb2, 9
and Kb1 were in the range of 51.5–63.9, 40.1–51.7, and 32.3–41.1 mg g−1, respectively. The Mg concentrations were quite similar at all stations: 18.4–22.0 mg g−1 at Kb1, 18.4–21.5 mg g−1 at Kb2, and 19.7–24.4 mg g−1 at Kb3 (Fig. 4b). The measured Sr concentrations were much lower, with ranges of 10.9–18.6, 17.5– 22.6, and 15.3–23.6 µg g−1 at Kb1, Kb2, and Kb3, respectively (Fig. 4c). Due to the relatively constant Mg concentrations along the fjord axis and the decreasing Ca concentrations towards the FM, the Mg:Ca molar ratios differed significantly: 0.6 ± 0.02 at station Kb3, 0.7 ± 0.04 at Kb2, and 0.9 ± 0.05 at Kb1 (Fig. 4d). There were no significant spatial differences in the Sr:Ca (Sr × 1000) molar ratios, as in the surface sediment layers at all stations the ratios were similar (0.18 ± 0.0003), while in the deeper layers they differed slightly and were lower (0.14 ± 0.02) at the station close to the GF (Kb3) than at the stations at the central (Kb2) and open (Kb1) parts of the fjord (0.19 ± 0.04) (Fig. 4e).
Fig. 4. Vertical profiles of calcium (a), magnesium (b), and strontium (c) concentrations and the molar ratios Mg:Ca (d) and Sr:Ca (Sr × 1000) (e), in the investigated sediment cores
4.4. Suspended particulate matter (SPM) as a source of sedimentary material. The extent and composition of material supplied to the bottom sediments of the fjord by glaciers and biota were determined in analyses of SPM samples. The results provide only a snapshot of the SPM distribution in mid-summer, a period when glacial melting and biological activity are highest. Our results confirmed that glaciers are an important source of SPM. The concentration at station Kb3 (10.2 ± 3.3 mg dm−3) was significantly higher than at stations Kb1 and Kb2 (Fig. 5a). A similar spatial distribution was determined for particulate 10
inorganic carbon (PIC) concentrations, which were very high at station Kb3 (0.19–0.75 mg dm-3) and significantly lower at stations Kb1 and Kb2 (0.01–0.07 mg dm-3) (Fig. 5c). Particulate organic carbon (POC) concentrations were comparable in samples collected across the entire fjord: 0.18 ± 0.05 mg dm−3 at station Kb3, 0.09 ± 0.06 mg dm−3 at Kb2, 0.12 ± 0.03 mg dm−3 at Kb1, and 0.12 ± 0.02 mg dm−3 at Kb0 (Fig. 5b). This suggested a similar magnitudes of primary production at stations Kb0–Kb2 whereas POC contributing to the SPM at station Kb3 reflected the specific characteristics of the SPM supplied by the glacier. Support for this conclusion comes from the percent contribution of POC to SPM, which increased from ~0.5% at the GF to nearly 28% at the FM. Primary productivity in the study area changes from 20 to 50 g m−2 year−1 in most of the fjord area to very low at the GF (Hop et al., 2002; Piwosz et al., 2009) are also consistent with our assignment of the POC in SPM.
Fig. 5. Concentrations [mg dm−3] of suspended particulate matter (a), particulate organic carbon (b), and particulate inorganic carbon (c) in water samples collected from four sampling stations: Kb3 (glacier front), Kb2, Kb1 (fjord mouth), and Kb0 (shelf) at four water layers: surface, subsurface (maximum fluorescence), subsurface (40 m), and bottom (5 m above the sea floor)
4.5. Provenience of Corg and Cinorg in sediments. The distribution of Corg and Cinorg presented in Fig. 3 and Fig. 5 suggests that Corg is mostly produced in fjord waters and Cinorg mostly in glaciers. Thus, the high concentrations of Cinorg at station Kb3 can be attributed to the abundant supply of carbonates from the Kronebreen glacier. A substantial proportion of the rocks in the region of Spitsbergen where the 11
Kongsbreen and the Kronebreen glaciers are located are limestone (Streuff, 2013). Moving glaciers scratch the bedrock and transport crushed rocks, including limestone, to the fjord (Bhushan et al., 2001; Hauck et al., 2012). However, somewhat lower but still high concentrations of carbonates were also determined in the sediments of station Kb1, located at the FM, and in the SPM collected from stations Kb0 and Kb1, both located far from the Kongsbreen and Kronebreen glaciers. Thus, particulate carbonates are either transported over long distances by the GF or/and are produced directly in the fjord. The latter possibility is best explained by the activity of calciferous organisms, whereas it is unlikely that the former is the sole source of carbonates, because both the SPM and the PIC concentrations were already fairly low at the Kb2 sampling station, with no further decrease towards the FM (Fig. 5). Differing distributions of both sedimentary Corg and Cinorg can be used to evaluate the origin of organic vs. inorganic carbon species. Allochthonous (glacier-supplied) and autochthonous (produced in situ) carbonates can be distinguished by assuming that carbonates are either glacial or biogenic in their origin and that organic matter is biogenically derived (Corg in glacier-supplied material is negligible). In the first step, Cinorg concentrations were normalized to Corg and plotted against Corg (Fig. 6), used here as an indicator of biological production. Cinorg:Corg ratios (10.3–35.1) were highest at station Kb3 (Fig. 6), which suggested a strong predominance of carbonates over Corg in the sediments and that they consisted of glacially supplied material. Cinorg:Corg ratios decreased gradually with increasing distance from the glacier. Both the lowest Cinorg:Corg ratios (0.93–1.48) and the least amount of variability therein were measured at the outermost station, Kb1. The dependence of the Cinorg:Corg ratios on Corg decreased exponentially with increasing Corg concentrations in the samples (Fig. 6). This pattern is typical of the mixing of two carbonate sources: a large, local source independent of Corg (referred to in the following as glacial carbonates) and a second source linked to biological activity. The presence of organic matter originating from biological activity can be used as a proxy of the biogenic carbonates produced in the fjord. Of note is that the plots of all the measured sediment samples deposited in different years and in the different regions of Kongsfjord represented by the cores yielded a single exponential curve (Fig. 6). Thus, the mixing of carbonates from these two sources can explain the distribution of carbonates in the fjord sediments. Interestingly, there was no variability in Cinorg to Corg ratios related to the age of the samples, which suggests that the mineralization of organic matter in the course of early diagenesis does not significantly alter Cinorg:Corg ratios, which instead depend only on the relative contributions of autochthonous and allochthonous sources in the 12
mixing process. While this does not exclude the mineralization of organic matter during the early stages of diagenesis in sediments, it does indicate that the mineralization of 1 mol of Corg must be accompanied by the dissolution of 1 mol of carbonates. This conclusion is in line with that of Pfeifer et al. (2002), who, based on empirical data, demonstrated that the molar ratio of carbonate dissolution and carbon dioxide release from Corg mineralization is roughly 1:1.
Fig. 6. Cinorg normalized to Corg is plotted against the Corg concentrations [mmol g−1] in the investigated Kongsfjord sediments
In samples with high Corg concentrations, the Cinorg:Corg ratios stabilized and did not decrease significantly with increasing Corg, as was the case with samples from Kb1 (Fig. 6). To maintain a constant Cinorg:Corg ratio in samples with different Corg concentration implies that any change of Corg must be accompanied by a change in Cinorg. This direct link between Cinorg and Corg suggests the biogenic character of the carbonates deposited in these samples. Thus, the lowest value of Cinorg:Corg from our dataset, recorded at the highest Corg concentration in sediments (0.93), is presumably characteristic of biogenic carbonates produced in the fjord. Multiplying this value by the Corg concentration in each sediment sample provides an estimate of the fraction of Cinorg that originated from in situ production in the fjord (biogenic Cinorg). Glacial Cinorg can then be calculated as the difference between total Cinorg and biogenic Cinorg.
13
However, it must be taken into account that a certain fraction of Corg in the Kongsfiord sediments may be allochthonous, which would influence the accuracy of the results presented above. Nonetheless, assuming that the spatial distribution of glacial Corg follows that of glacial Cinorg, then glaciers are not an important source of Corg in the surficial sediments and the organic matter buried in Kongsfjord’s bottom sediments, especially in the outer part of the fjord, is therefore mostly autochthonous. While some fraction of terrigenous Corg may be supplied by rivers as relatively fresh organic matter, this load would be a minor one compared to the autochthonous organic matter load originating from primary production (Kuliński et al., 2014). Previous reports (Forwick et al., 2010; Hop et al., 2002; Kuliński et al., 2014; Zajaczkowski, 2008) indicated that the activity of tidal glaciers cause not only gradients in salinity but also in the quality and quantity of sedimentary organic and inorganic matter. This accounts for the steep gradients in the concentrations of sedimentary organic carbon, which decrease sharply towards the GF due to the depletion of organic matter in the weathered mineral material delivered by the glaciers (Bourgeois et al., 2016; Kuliński et al., 2014; Kumar et al., 2016). Bourgeois et al. (2016) reported a low POC content in the vicinity of the glacier, accompanied by surprisingly high values of the δ13C signatures of organic carbon (ranging from −20 to – 22‰), suggestive of in situ production as a source of organic matter. A low supply of terrigenous organic matter was also determined by Kuliński et al. (2014), who, based on POC concentrations and the freshwater volume supplied to the fjord, estimated that the annual terrestrial (including glaciers and rivers) POC load constitutes only 5–10 % of the total carbon load delivered from primary production in the fjord. Kumar et al. (2016) also measured high δ13Corg values close to the GF; however, by assuming that terrestrial OM in this region is relatively isotopically heavy, they suggested, that organic carbon originating from allochthonous sources predominates at the GF over carbon from autochthonous sources. While the assumption that terrestrial organic carbon is isotopically heavy needs to be validated, the approach used in this study is valid irrespective of the isotopic signatures of Corg in the investigated system.
4.6. Vertical distributions of glacial and biogenic Cinorg. Having dated the investigated sediment cores we were able to assess the time dependence of the glacial Cinorg concentrations at the outer and central stations (Fig. 7a). The systematic shifts in the profiles were indicative of sedimentary processes related to, e.g., glacial surging. 14
Most of the glaciers in the Kongsfjord catchment are of the surge-type (Dowdeswell et al., 1991; Hagen, 1993; Ottesen and Dowdeswell, 2006; Ottesen et al., 2008), such that their activity is characterized by a succession of active and passive phases. During active phases, the ice front advances rapidly or, at least, stagnates, while passive phases are characterized by slow glacial retreat, which can take from 50 to 500 years. Three glacier surges have been documented at our study site: Kronebreen in 1869, Kongsvegen in 1948, and Blomstrandbreen in 1960 (Dowdeswell et al., 1991; Hagen, 1993; Ottesen and Dowdeswell, 2006; Ottesen et al., 2008; Plassen et al., 2004). Thus, the investigated sediment cores, whose deepest sediment layer at Kb1 and Kb2 dated to 1881–1886 and 1913–1916, respectively, included two surges. The absence of significant changes in the glacial Cinorg concentrations profile coinciding with the Kongsvegen surge in 1948 was probably due to the substantial distance between this glacier and the outer (Kb1) and central (Kb2) stations. At station Kb2, there was a visible change in glacial Cinorg concentrations during the period 1960–1970, which was probably related either to the changes in the position of the Kronebreen GF or, albeit less likely, to the Blomstrandbreen surge in 1960. There were significant changes in both glacial and biogenic Cinorg during the last 25–35 years (Fig. 7 a, b) when the increase in biogenic Cinorg was accompanied by a substantial decrease in glacial Cinorg. The higher concentrations of biogenic Cinorg in the uppermost part of the sediment cores were attributable to increased biological activity in the water column, (Pfeifer et al., 2002), whereas the drop in glacial Cinorg was most likely due to changes in glacial activity, including the occurrence of a retreat.
15
Fig. 7. Concentrations of glacial (a) and biogenic (b) Cinorg plotted against the deposition year for the investigated cores
4.7. Characteristics of glacial vs. biogenic carbonates. The charge balance performed in all samples, in which Ca, Mg, and Sr were measured in parallel with Cinorg, suggested that carbonates are the sole carrier of these metals in sediments. Thus, for all samples, the molar concentration of Cinorg was almost entirely balanced by the sum of the Ca and Mg molar concentrations. By contrast, the contribution of Sr was less important, as its concentration was roughly three orders of magnitude lower than that of Ca or Mg (Fig. 4). The maximum deviation from the equality of the molar concentrations of the sum of the concentrations of Ca and Mg cations and carbonates was <5%, a difference not exceeding analytical uncertainty. Since the distribution of Mg and Sr concentrations was relatively constant along the fjord axis and the concentrations of Ca decreased significantly towards the FM (Fig. 4), the molar ratios of Mg:Ca and Sr:Ca could be used to characterize the chemical composition of carbonates originating from different sources and to perhaps distinguish between biogenic and glacial 16
carbonates. A plot of these ratios against the concentrations of biogenic and glacial Cinorg concentrations revealed the dependences shown in Fig. 8. The intercepts of the linear regressions at the null concentrations of the respective carbonates indicated the ratios that characterized the glacial and biogenic end-members. The Mg:Ca and Sr:Ca ratios for glacial carbonates were 0.56 (Fig. 8a) and 0.00015 (Fig. 8b) whereas for biogenic carbonates they were 0.94 (Fig. 8c) and 0.00020 (Fig. 8d), respectively. Thus, both the glacial and the biogenic carbonates deposited in the Kongsfiord sediments are relatively rich in Mg (Long et al., 2014). These characteristic ratios could be used to identify the species responsible for the biogenic carbonate supply in the Kongfiord ecosystem. Although beyond the scope of our study, a literature screening indicated that Ostracoda, Echinodermata, red corals, red coralline algae, and calcareous serpulid worms (Farmer et al., 2012; Long et al., 2014; Veis, 2011), among others, produce large amounts of magnesium calcites.
Fig. 8. Glacial (a, b) and biogenic (c, d) end-members of the Mg:Ca (a, c) and Sr:Ca (b, d) ratios as indicated by the intercepts of the linear functions presented in the plots
5. Conclusions
17
This study tested the hypothesis that sediments of the Svalbard fjords contain important amounts of biogenic and terrestrial sources of Cinorg. Total sedimentary carbon concentrations along the axis of the Kongsfjord, from the FM (station Kb1), to an intermediate area (station Kb2), and finally to the GF (station Kb3) were in the range of 30–45 mg (g dry sediment)−1. The contribution of sedimentary Cinorg (carbonates) was determined to be substantial whereas that of Corg was much smaller, as evidenced by the composition of total carbon: 1–18 mg Corg (g dry sediment)−1 and 17–43 mg Cinorg (g dry sediment)−1. The horizontal gradient of Corg concentrations, which ranged from 1–2 mg g−1 at Kb3 to 13– 18 mg g−1 at Kb1 could be attributed to the supply of organic-matter-depleted, glacial mineral material and to the limited contribution from the biota, with both factors having the largest effect close to the GF. The horizontal gradient of Cinorg (Kb3: 29-43 mg g−1 and Kb1: 17–20 mg g-1) was most likely caused by the large supply of terrigenous carbonates from the GF, and the lower but still substantial supply of biogenic carbonates produced in the central and outer parts of the fjord.
`
Using the dependences of Cinorg:Corg vs Corg and the concentrations of sedimentary Ca, Mg, and Sr, we developed a method, that allows quantification of the contributions of glacial and biogenic carbonates to total carbonates. Glaciers are an important source of carbonates in the Kongsfiord, but our results for both sediments and SPM suggested that glacial material is mostly deposited in the vicinity of the GF, such that the carbonates at Kb1 are almost exclusively biogenic in origin and those at Kb2 a mixture of comparable amounts from the two sources. The vertical zonation of glacial Cinorg concentrations at station Kb2 may be the product of either glacial surges or changes in glacial activity. The less pronounced vertical zonation at station Kb1 was most likely a function of the large distance to the GFs of Kronebreen and Kongsvegen. In addition, we determined Mg:Ca and Sr:Ca ratios for glacial carbonates of 0.56 and 0.00015, and for biogenic carbonates of 0.94 and 0.00020, respectively. Our study demonstrates that in quantifications of the carbon sinks in the Kongsfjord sediments and in those of the other Svalbard fjords, carbonates need to be considered and their source (glacial vs. biogenic) identified.
18
Acknowledgement Financial support from the Polish National Science Centre (grant no. 2015/19/N/ST10/01652) and the statutory activities of the Institute of Oceanology (Sopot, Poland) are acknowledged. Katarzyna Koziorowska’s research for this paper was supported by the Centre for Polar Studies, KNOW – Leading National Research Centre, Sosnowiec, Poland. We thank A. Iglikowska for her valuable comments on the CO2 system and J. Walkusz-Miotk for her assistance in the AAS analyses. References Andruleit, H., Freiwald, A., Schafer, P., 1996. Bioclastic carbonate sediments on the southwestern Svalbard shelf. Marine Geology 134, 163-182. Beszczyńska-Moller, A., Węsławski, J.M., Walczowski, W., Zajączkowski, M., 1997. Estimation of glacial meltwater discharge into Svalbard coastal waters. Oceanology 39, 289-298. Bhushan, R., Dutta, K., Somayajulu, B.L.K., 2001. Concentrations and burial fluxes of organic and inorganic carbon on the eastern margins of the Arabian Sea. Marine Geology 178, 95-113. Bourgeois, S., Kerherve, P., Calleja, M.L., Many, G., Morata, N., 2016. Glacier inputs influence organic matter composition and prokaryotic distribution in a high Arctic fjord (Kongsfjorden, Svalbard). Journal of Marine Systems 164, 112-127. Carpenter, S.J., Lohmann, K.C., 1992. Sr/Mg ratios of modern narine calcite - empirical indicators of ocean chemistry and precipitation rate. Geochimica Et Cosmochimica Acta 56, 1837-1849. Carroll, J., Zaborska, A., Papucci, C., Schirone, A., Carroll, M.L., Pempkowiak, J., 2008. Accumulation of organic carbon in western Barents Sea sediments. Deep-Sea Research Part Ii-Topical Studies in Oceanography 55, 2361-2371. Dowdeswell, J.A., Hamilton, G.S., Hagen, J.O., 1991. The duration of the active phase on surge-type-glaciers-contrasts between Svalbard and other regions. Journal of Glaciology 37, 388-400. Elverhoi, A., Lonne, O., Seland, R., 1983. Glacimarine sedimentation in a modern fjord environment. Polar Research 1, 127-149.
19
Farmer, J.R., Cronin, T.M., Dwyer, G.S., 2012. Ostracode Mg/Ca paleothermometry in the North Atlantic and Arctic oceans: Evaluation of a carbonate ion effect. Paleoceanography 27. Faust, J.C., Knies, J., Slagstad, T., Vogt, C., Milzer, G., Giraudeau, J., 2014. Geochemical composition of Trondheimsfjord surface sediments: Sources and spatial variability of marine and terrigenous components. Continental Shelf Research 88, 61-71. Forwick, M., Vorren, T.O., Hald, M., Korsun, S., Roh, Y., Vogt, C., Yoo, K.C., 2010. Spatial and temporal influence of glaciers and rivers on the sedimentary environment in Sassenfjorden and Tempelfjorden, Spitsbergen. Fjord Systems and Archives 344, 163193. Gorlich, K., 1986. Glacimarine sedimentation of muds in Hornsund Fjord, Spitsbergen. Annale Societatis Geologorum Poloniae 56, 433-477. Gorlich, K., Weslawski, J.M., Zajaczkowski, M., 1987. Suspension settling effect on macrobenthos biomass distribution in the Hornsund Fjord, Spitsbergen. Polar Research 5, 175-192. Hagen, J., 1993. Glacier atlas of Svalbard and Jan Mayen. Norsk Polarinstitutt Middelelser. Hauck, J., Gerdes, D., Hillenbrand, C., Hoppema, M., Kuhn, G., Nehrke, G., Voelker, C., Wolf-Gladrow, D.A., 2012. Distribution and mineralogy of carbonate sediments on Antarctic shelves. Journal of Marine Systems 90, 77-87. Hop, H., Pearson, T., Hegseth, E.N., Kovacs, K.M., Wiencke, C., Kwaśniewski, S., Eiane, K., Mehlum, F., Gulliksen, B., Włodarska-Kowalczuk, M., Lydersen, C., Węsławski, J.M., Cochrane, S., Gabrielsen, G.W., Leakey, R.J.G., Lonne, O.J., Zajączkowski, M., FalkPetersen, S., Kendall, M., Wangberg, S.A., Bischof, K., Voronkov, A.Y., Kovaltchouk, N.A., Wiktor, J., Poltermann, M., di Prisco, G., Papucci, C., Gerland, S., 2002. The marine ecosystem of Kongsfjorden, Svalbard. Polar Research 21, 167-208. Hulth, S., Hall, P.O.J., Blackburn, T.H., Landen, A., 1996. Arctic sediments (Svalbard): Pore water and solid phase distributions of C, N, P and Si. Polar Biology 16, 447-462. Kędra, M., Kuliński, K., Walkusz, W., Legeżyńska, J., 2012. The shallow benthic food web structure in the high Arctic does not follow seasonal changes in the surrounding environment. Estuarine Coastal and Shelf Science 114, 183-191. Koziorowska, K., Kuliński, K., Pempkowiak, J., 2016. Sedimentary organic matter in two Spitsbergen fjords: terrestrial and marine contribution based on carbon and nitrogen contents and stable isotopes composition. Continental Shelf Research 113, 38-46.
20
Kuliński, K., Kędra, M., Legeżyńska, J., Głuchowska, M., Zaborska, A., 2014. Particulate organic matter sinks and sources in high Arctic fjord. Journal of Marine Systems 139, 27-37. Kumar, V., Tiwari, M., Nagoji, S., Tripathi, S., 2016. Evidence of Anomalously Low delta C13 of Marine Organic Matter in an Arctic Fjord. Scientific Reports 6. Long, X., Ma, Y.R., Qi, L.M., 2014. Biogenic and synthetic high magnesium calcite - A review. Journal of Structural Biology 185, 1-14. Morse, J.W., 2003. Formation and diagenesis of carbonate sediments, In Treatise on Geochemistry. Elsevier, Amsterdam, pp. 67-85. Ottesen, D., Dowdeswell, J.A., 2006. Assemblages of submarine landforms produced by tidewater glaciers in Svalbard. Journal of Geophysical Research-Earth Surface 111. Ottesen, D., Dowdeswell, J.A., Benn, D.I., Kristensen, L., Christiansen, H.H., Christensen, O., Hansen, L., Lebesbye, E., Forwick, M., Vorren, T.O., 2008. Submarine landforms characteristic of glacier surges in two Spitsbergen fjords. Quaternary Science Reviews 27, 1583-1599. Pempkowiak, J., 1991. Enrichment factors of heavy-metals in the Southern Baltic surface sediments dated with Pb210 and Cs137. Environment International 17, 421-428. Pfeifer, K., Hensen, C., Adler, M., Wenzhofer, F., Weber, B., Schulz, H.D., 2002. Modeling of subsurface calcite dissolution, including the respiration and reoxidation processes of marine sediments in the region of equatorial upwelling off Gabon. Geochimica Et Cosmochimica Acta 66, 4247-4259. Piwosz, K., Walkusz, W., Hapter, R., Wieczorek, P., Hop, H., Wiktor, J., 2009. Comparison of productivity and phytoplankton in a warm (Kongsfjorden) and a cold (Hornsund) Spitsbergen fjord in mid-summer 2002. Polar Biology 32, 549-559. Plassen, L., Vorren, T.O., Forwick, M., 2004. Integrated acoustic and coring investigation of glacigenic deposits in Spitsbergen fjords. Polar Research 23, 89-110. Ries, J.B., 2006. Mg fractionation in crustose coralline algae: Geochemical, biological, and sedimentological implications of secular variation in the Mg/Ca ratio of seawater. Geochimica Et Cosmochimica Acta 70, 891-900. Smith, R.W., Bianchi, T.S., Allison, M., Savage, C., Galy, V., 2015. High rates of organic carbon burial in fjord sediments globally. Nature Geoscience 8, 450-U446. Stein, R., Macdonald, R.W., 2004. The organic carbon cycle in Arctic Ocean. Springer. Streuff, K., 2013. Landform assemblages in inner Kongsfjorden, Svalbard: evidence of recent glacial (surge) activity. University of Tromso. 21
Streuff, K., Forwick, M., Szczuciński, W., Andreassen, K., Cofaigh, C.O., 2015. Submarine landform assemblages and sedimentary processed related to glacier surging in Kongsfjorden, Svalbard. Arktos 1: 14. Svendsen, H., Beszczynska-Moller, A., Hagen, J.O., Lefauconnier, B., Tverberg, V., Gerland, S., Orbaek, J.B., Bischof, K., Papucci, C., Zajaczkowski, M., Azzolini, R., Bruland, O., Wiencke, C., Winther, J.G., Dallmann, W., 2002. The physical environment of Kongsfjorden-Krossfjorden, an Arctic fjord system in Svalbard. Polar Research 21, 133166. Szczuciński, W., Zajączkowski, M., Scholten, J., 2009. Sediment accumulation rates in subpolar fjords - Impact of post-Little Ice Age glaciers retreat, Billefjorden, Svalbard. Estuarine Coastal and Shelf Science 85, 345-356. Trusel, L.D., Powell, R.D., Cumpston, R.M., Brigham-Grette, J., 2010. Modern glacimarine processes and potential future behaviour of Kronebreen and Kongsvegen polythermal tidewater glaciers, Kongsfjorden, Svalbard. Fjord Systems and Archives 344, 89-102. Veis, A., 2011. Organic Matrix-related mineralization of sea urchin spicules, spines, test and teeth. Frontiers in Bioscience-Landmark 16, 2540-2560. Walkusz, J., Roman, S., Pempkowiak, J., 1992. Contamination of the Southern Baltic surface sediments with heavy metals. Bull. Sea Fis. Inst. 1 (125), 33-37. Winkelmann, D., Knies, J., 2005. Recent distribution and accumulation of organic carbon on the continental margin west off Spitsbergen. Geochemistry Geophysics Geosystems 6. Włodarska-Kowalczuk, M., Pearson, T.H., 2004. Soft-bottom macrobenthic faunal associations and factors affecting species distributions in an Arctic glacial fjord (Kongsfjord, Spitsbergen). Polar Biology 27, 155-167. Włodarska-Kowalczuk, M., Szymelfenig, M., Zajączkowski, M., 2007. Dynamic sedimentary environments of an Arctic glacier-fed river estuary (Adventfjorden, Svalbard). II: Meioand macrobenthic fauna. Estuarine Coastal and Shelf Science 74, 274-284. Zaborska, A., Carroll, J., Papucci, C., Pempkowiak, J., 2007. Intercomparison of alpha and gamma spectrometry techniques used in Pb-210 geochronology. Journal of Environmental Radioactivity 93, 38-50. Zaborska, A., Pempkowiak, J., Papucci, C., 2006. Some sediment characteristic and sedimentation rates in an Arctic fjord (Kongsfjorden, Svalbard). Annales Environment Protection 8, 79-96. Zajaczkowski, M., 2008. Sediment supply and fluxes in glacial and outwash fjords, Kongsfjorden and Adventfjorden, Svalbard. Polish Polar Research 29, 59-72. 22
Highlights
Stratified bottom sediments and SPM were collected in Kongsfjord, Svalbard Corg, Cinorg, Ca, Mg and Sr were measured in the collected material Terrestrial vs marine pools of both carbon species are distinguished Sedimentary Corg is mostly marine, while glaciers are only its minor source Cinorg close to glacier is terrigenous while at the fjord mouth- is mostly biogenic
23