Iron and manganese speciation and cycling in glacially influenced high-latitude fjord sediments (West Spitsbergen, Svalbard): Evidence for a benthic recycling-transport mechanism

Abstract

Glacial environments may provide an important but poorly constrained source of potentially bioavailable iron and manganese phases to the coastal ocean in high-latitude regions. Little is known about the fate and biogeochemical cycling of glacially derived iron and manganese in the coastal marine realm. Sediment and porewater samples were collected along transects from the fjord mouths to the tidewater glaciers at the fjord heads in Smeerenburgfjorden, Kongsfjorden, and Van Keulenfjorden along Western Svalbard. Solid-phase iron and manganese speciation, determined by sequential chemical extraction, could be linked to the compositions of the local bedrock and hydrological/weathering conditions below the local glaciers. The concentration and sulfur isotope composition of chromium reducible sulfur (CRS) in Kongs- and Van Keulenfjorden sediments largely reflect the delivery rate and isotope composition of detrital pyrite originating from adjacent glaciers. The varying input of reducible iron and manganese oxide phases and the input of organic matter of varying reactivity control the pathways of organic carbon mineralization in the sediments of the three fjords. High reducible iron and manganese oxide concentrations and elevated metal accumulation rates coupled to low input of “fresh” organic matter lead to a strong expression of dissimilatory metal oxide reduction evidenced in very high porewater iron (up to 800 μM) and manganese (up to 210 μM) concentrations in Kongsfjorden and Van Keulenfjorden. Sediment reworking by the benthic macrofauna and physical sediment resuspension via iceberg calving may be additional factors that promote extensive benthic iron and manganese cycling in these fjords. On-going benthic recycling of glacially derived dissolved iron into overlying seawater, where partial re-oxidation and deposition occurs, facilitates the transport of iron across the fjords and potentially into adjacent continental shelf waters. Such iron-dominated fjord sediments are likely to provide significant fluxes of potentially bioavailable iron to coastal waters and beyond. By contrast, low delivery of reducible iron (oxyhydr)oxide phases and elevated organic carbon mineralization rates driven by elevated input of “fresh” marine organic matter allow organoclastic sulfate reduction to dominate carbon remineralization at the outer Smeerenburgfjorden sites, which may limit iron fluxes to the water column.

Introduction

Iron plays a vital role in controlling primary production, biological carbon export, and nitrogen fixation in many parts of the ocean (e.g., Martin and Fitzwater, 1988, Coale et al., 1996, Falkowski, 1997, Hutchins and Bruland, 1998, Falkowski et al., 1998, Boyd et al., 2000). Growing awareness of the essential role of iron for primary productivity in many ocean regions and the discovery of an apparent tight linkage of changes in iron supply to the ocean with glacial-interglacial variations in atmospheric CO₂ concentrations (Martin, 1990; Jickells et al., 2005) has steered a large body of research devoted to the biogeochemical iron cycle over the past decades. Many studies have focused on the role of aeolian dust as a source of bioavailable iron to the ocean (e.g., Bruland et al., 1994, de Baar and de Jong, 2001, Jickells and Spokes, 2001, Moore and Braucher, 2008) and, more recently, on benthic iron remobilization from shelf sediments (e.g., Johnson et al., 1999, Elrod et al., 2004, Moore and Braucher, 2008, Severmann et al., 2010, Noffke et al., 2012, Homoky et al., 2013). Iron deriving from glacial environments represents the least constrained source of iron to the ocean (Raiswell et al., 2006). There is growing awareness that glacial environments may provide this key nutrient to high latitude ocean regions, where this source likely dominates over other iron fluxes to coastal waters (Statham et al., 2008, Raiswell et al., 2008a, Lippiatt et al., 2010, de Jong et al., 2012, Wadham et al., 2010; Wadham et al., 2013, Gerringa et al., 2012, Bhatia et al., 2013, Hawkings et al., 2014).

Glacial environments have been identified as important microbial habitats (Sharp et al., 1999, Skidmore et al., 2000, Gaidos et al., 2004, Mikucki et al., 2004, Lanoil et al., 2009) that function as crucial sites for the transformation of different carbon pools and for the microbially enhanced chemical weathering of glacial debris on the surface, below, and within ice sheets and glaciers (e.g., Tranter et al., 2003, Wadham et al., 2004, Skidmore et al., 2005, Wadham et al., 2010, Montross et al., 2012, Stibal et al., 2012). Biogeochemical weathering processes in subglacial and proglacial environments include (1) oxidation of sulfide minerals by oxygen, Fe(III), and nitrate; (2) dissolution of carbonates; and (3) silicate weathering (see Tranter et al., 2003, Skidmore et al., 2010, Wadham et al., 2013; for review). These weathering processes yield, for example, sulfate, bicarbonate, magnesium, calcium ions, and nutrients such as iron and phosphorus to glacial runoff waters (Wadham et al., 2001, Cooper et al., 2002, Tranter, 2005). Oxic glacial runoff may provide a comparably small amount of dissolved iron to coastal waters (Statham et al., 2008, Raiswell and Canfield, 2012, Wadham et al., 2013) that is accompanied by a large fraction of iron (oxyhydr)oxides and other, mechanically weathered, particulate iron phases (Poulton and Raiswell, 2002, Poulton and Raiswell, 2005, Raiswell et al., 2006). In contrast, anoxic runoff waters, entering coastal waters below tidewater glaciers, ice-sheets, and by groundwater discharge, may provide additional iron sources with higher aqueous and colloidal/nanoparticulate iron components (Skidmore et al., 2010, Raiswell and Canfield, 2012, Wadham et al., 2013). Icebergs and sea-ice contain nanoparticulate iron(oxyhydr)oxides preserved in the frozen matrix that can evade deposition in fjord sediments and thus reach open ocean waters (e.g., Raiswell et al., 2006, Raiswell et al., 2008a, Raiswell et al., 2008b, Raiswell et al., 2009, Lannuzel et al., 2010, van der Merwe et al., 2011, de Jong et al., 2013). Accordingly, these ice-borne inputs are now regarded as important sources of bioavailable iron to high-nutrient, low-chlorophyll areas of the Southern Ocean and Antarctic (Raiswell et al., 2006, Smith et al., 2007, Lannuzel et al., 2007, Lannuzel et al., 2008, Raiswell et al., 2008a, Lin et al., 2011, Shaw et al., 2011).

Glacially derived iron enters adjacent ocean waters in aqueous (here defined as <0.02 μm) plus colloidal/nanoparticulate (collectively termed ‘dissolved iron’) and particulate forms (>1 μm; Raiswell et al., 2006, Raiswell and Canfield, 2012). The composition and physico-chemical reactions of the iron pool determine its bioavailability and thus ultimately the potential to fuel primary production in surface waters (Moore and Braucher, 2008, Tagliabue et al., 2009, Breitbarth et al., 2010). Aqueous, colloidal, and nanoparticulate iron phases exhibit varying degrees of bioavailability (Hutchins et al., 1999, Yoshida et al., 2002, Kraemer, 2004, Borer et al., 2005, Shaked et al., 2005, Hunter and Boyd, 2007, Boyd and Ellwood, 2010, Breitbarth et al., 2010, Raiswell and Canfield, 2012, Shaked and Lis, 2012). Aqueous iron, which mainly occurs as Fe(III) bound to organic ligands in marine seawater (Van den Berg, 1995, Rue and Bruland, 1995, Hassler et al., 2011), is suggested to be bioavailable, although the ligand type may determine its availability for specific groups of organisms (see Hunter and Boyd, 2007, and Gledhill and Buck, 2012, for review). Colloidal/nanoparticulate species, which are suggested to be composed of small iron (oxyhydr)oxide and organic matter subunits (Boyé et al., 2010, Raiswell and Canfield, 2012), are believed to be only partially bioavailable to eukaryotic organisms (Chen and Wang, 2001, Chen et al., 2003, Cullen et al., 2006).

Manganese also functions as a micronutrient (Sunda et al., 1981, Coale, 1991, Morel and Price, 2003), and the scavenging of trace metals by manganese oxides can exert a strong control on the distribution of biologically essential trace elements, such as cobalt, in continental margin environments (Knauer et al., 1982, Shaw et al., 1990). Input from continental erosion represents a main source of manganese oxides to continental shelf waters. Here, the benthic recycling of manganese oxides and subsequent flux of manganese from the seabed represents an important pathway for the reintroduction of dissolved manganese into the water column (Heggie et al., 1987, Johnson et al., 1992, Aller, 1994, Pakhomova et al., 2007, McManus et al., 2012). The diagenetic recycling and transport of manganese from Arctic shelf sediments has been invoked as a main driver for the formation of distinct Mn-rich brown layers in adjacent slope and basin sedimentary sequences (März et al., 2011, März et al., 2012, Macdonald and Gobeil, 2012). These studies suggest that a large fraction of manganese reaching the inner Arctic shelves (<100 m water depth) must be recycled and transported to the Central Arctic Ocean in order to balance the modern manganese budget of the Arctic.

Given the large volume of glacial runoff in many high-latitude regions, adjacent fjords represent important depocenters for glacially derived iron and manganese. However, little is known about the cycling and transfer of these metals in the fjords, which ultimately represent the transition between the glacial and marine environments. A large fraction of aqueous and nanoparticulate iron phases originating from glacial sources is suggested to precipitate during their passage across the salinity gradient from glaciers to adjacent coastal waters (e.g., Lippiatt et al., 2010), for example, as iron (oxyhydr)oxide coatings on particles. Glacial runoff waters, however, generally also contain a large fraction of particulate iron phases, with concentrations exceeding those of dissolved iron by orders of magnitude (Poulton and Raiswell, 2005, Raiswell et al., 2006, Lippiatt et al., 2010). The fate of the large amounts of iron (oxyhydr)oxides delivered to fjords directly by glacial runoff, or indirectly through precipitation of dissolved glacial iron, remains poorly understood. It is assumed that further coagulation in saline fjord waters results in trapping of 90% of these iron phases in fjord sediments (Raiswell, 2006, Raiswell et al., 2006). Biogeochemical processes in these fjords, particularly in the fjord sediments, have not been considered previously, but these processes may strongly affect the distribution and speciation of iron phases and ultimately alter the bioavailability of the glacially derived iron phases.

A better understanding of the input of glacially derived iron and manganese, and the processes and feedbacks that tie the iron cycle to recent climate change, are needed for the Arctic environment. The Arctic Ocean is very susceptible to global climate change through various feedback mechanisms (Anisimov et al., 2007; Meehl et al., 2007; Spielhagen et al., 2011). For instance, sea-ice cover of the Arctic Ocean is retreating unexpectedly fast (Shimada et al., 2006, Comiso et al., 2008, Giles et al., 2008, Perovich et al., 2008), thus exposing increasingly large areas of this region to sunlight. This change may allow for enhanced phytoplankton growth in summer (Arrigo et al., 2008, Slagstad et al., 2011), depending on the availability of nutrients to sustain summer primary production (Carmack et al., 2006, Popova et al., 2010). The input of iron may become one limiting factor for primary productivity, especially if ice-borne transport of iron to the open ocean is reduced in a warming climate (Carmack et al., 2006). Runoff from local glaciers may provide additional input of dissolved and colloidal iron to coastal waters and potentially beyond (Statham et al., 2008, Lippiatt et al., 2010, Gerringa et al., 2012). Shelf areas of the Arctic are at present responsible for a disproportionately high fraction of primary production of up to 24% (Pabi et al., 2008, Popova et al., 2010), and the southeast Greenland Sea and the Barents Sea around Svalbard are amongst the most productive areas of the Arctic Ocean (Sakshaug, 2004, Carmack et al., 2006, Popova et al., 2010). The glaciers of the Svalbard Archipelago have retreated significantly in the past 50 years (Dowdeswell, 1995, Nordli et al., 1996, Ziaja, 2001, Kohler et al., 2007), and there is evidence for unprecedented increases in summer temperatures compared to the past 1800 years in this region (Spielhagen et al., 2011, D’Andrea et al., 2012). The Svalbard fjords thus represent excellent model locations to evaluate the coupling between climatic changes, glacial iron delivery mechanisms, and biogeochemical feedback processes.

In this study, we take a comprehensive approach in our examination of the biogeochemical processes in fjord sediments of three Western Svalbard fjords: Smeerenburgfjorden, Kongsfjorden, and Van Keulenfjorden. First, we determine how the local bedrock geology and the glacial coverage may affect the speciation and amount of delivered iron and manganese, water column productivity, and quality of deposited organic matter. The three chosen fjords are very different with respect to their bedrock composition and the size of the catchment area glacial coverage. We hypothesize that glaciated areas characterized by relatively Fe-poor lithologies deliver glacial flour low in reducible iron to the adjacent fjords. In contrast, iron- and pyrite-rich rock types, such as sandstones, may allow for the mechanical and biogeochemical production of higher amounts of dissolved iron and iron (oxyhydr)oxide phases compared to igneous and metamorphic bedrock lithologies. Second, the extent of glacial coverage in the fjord-surrounding areas may affect the volume and suspended load of glacial runoff and thus ultimately the amount of weathered iron and manganese phases that are delivered to the adjacent fjords. Third, we tie these findings to the prevailing sedimentary biogeochemical processes and elucidate the resulting differences amongst the different fjords. We specifically investigate how the biogeochemical processes control the fixation, remobilization, and ultimately the down-fjord transport of glacially derived iron and manganese phases. We shed light on the diagenetic cycling of these phases as an additional, previously unconsidered, mechanism that may aid the transfer of glacially derived metals across fjords to the outer shelf. Finally, we discuss the biogeochemical processes in the glacially influenced fjords within the context of the sensitivity of the Arctic region to recent climate change.