Rapid lateral particle transport in the Argentine Basin: Molecular 14C and 230Thxs evidence

https://doi.org/10.1016/j.dsr.2006.05.005Get rights and content

Abstract

Recent studies have revealed that lateral transport and focusing of particles strongly influences the depositional patterns of organic matter in marine sediments. Transport can occur in the water column prior to initial deposition or following sediment re-suspension. In both cases, fine-grained particles and organic-rich aggregates are more susceptible to lateral transport than coarse-grained particles (e.g., foraminiferal tests) because of the slower sinking velocities of the former. This may lead to spatial and, in the case of redistribution of re-suspended sediments, temporal decoupling of organic matter from coarser sediment constituents. Prior studies from the Argentine Basin have yielded evidence that suspended particles are displaced significant distances (100–1000 km) northward and downslope by strong surface and/or bottom currents. These transport processes result in anomalously cold alkenone-derived sea-surface temperature (SST) estimates (up to 6 °C colder than measured SST) and in the presence of frustules of Antarctic diatom species in surface sediments from this area. Here we examine advective transport processes through combined measurements of compound-specific radiocarbon ages of marine phytoplankton-derived biomarkers (alkenones) from core tops and excess 230Th (230Thxs)-derived focusing factors for late Holocene sediments from the Argentine Basin. On the continental slope, we observe 230Thxs-based focusing factors of 1.4–3.2 at sites where alkenone-based SST estimates were 4–6 °C colder than measured values. In contrast, alkenone radiocarbon data suggest coeval deposition of marine biomarkers and planktic foraminifera, as alkenones in core tops were younger than, or similar in age to, foraminifera. We therefore infer that the transport processes leading to the lateral displacement of these sediment components are rapid, and hence probably occur in the upper water column (<1500 m).

Introduction

The value of deep-sea sediments as records of climate history has long been recognized. Past environmental conditions of the oceans can be reconstructed from physical and chemical characteristics of sediment constituents, so-called proxy parameters. The carriers of the various proxies can reside in different grain-size fractions and are associated with different types of sedimentary constituents (e.g., biogenic carbonates, organic matter, and lithic grains such as clay particles).

It is well known that post-depositional processes (e.g., diagenesis or bioturbation) can affect the fidelity of sedimentary records and thereby reduce their potential for accurately recording past climate conditions (e.g., Guinasso and Schink, 1975; Bard et al., 1987; Thomson et al., 1995; Hedges et al., 1999; Bard, 2001). Pre-depositional lateral transport of particles within the water column is an additional process that potentially can affect the paleoceanographic record (e.g., Thomsen et al., 1998; Santschi et al., 1999; Freudenthal et al., 2001). Displacement of particles from their point of origin reduces our ability to accurately reconstruct past environmental conditions at a given location. Furthermore, lateral transport has the potential to sort sediment particles hydrodynamically according to size and sinking velocity and may thus result in decoupling of proxy records residing within different particle classes. These processes remain poorly understood but have recently been the subject of increased attention and debate (e.g., Marcantonio et al., 2001).

Two separate approaches have been taken to investigate sediment transport and its effect on paleoceanographic records. The first employs uranium (U)-series isotopes to understand and quantify lateral particle transport in the oceans (Suman and Bacon, 1989; Thomson et al., 1999; Marcantonio et al., 2001; François et al., 2004; Loubere et al., 2004). This method relies upon an essentially constant amount of dissolved U in seawater coupled with highly particle reactive daughter isotopes (primarily Thorium-230 (230Th) and Protactinium-231 (231Pa)). The estimated vertical component of sediment flux is based on the assumption of rapid scavenging of 230Th by sinking particles. It is assumed that at any given time and location the production of 230Th in the water column is known and equal to the rate of decay of 234U, and that the mode of 230Th scavenging into the sediment by adsorption of this isotope is also well known and constant with time (for details see François et al., 2004). Preserved vertical fluxes of sediments are calculated by normalizing to the decay-corrected concentration of 230Th in sediments as a function of water depth.

Estimates of sediment focusing can be based on the assumption that the scavenged 230Th inventory of the sediment between two given depths matches the production in the overlying water column integrated over the time of accumulation of this depth interval. If the inventory is bigger than expected, it is concluded that the sediments are focused. Focusing is quantified by a focusing factor (Ψ), which is calculated as the ratio of the inventory and the production over the time of accumulation. Thus, Ψ values greater than unity indicate focusing, while Ψ values less than 1 imply the occurrence of sediment winnowing. Using this approach, several authors have argued that observed sediment accumulation maxima (e.g., in the equatorial Pacific) formerly interpreted as paleoproductivity signals (Paytan et al., 1996) are instead a consequence of changes in lateral particle supply (Marcantonio et al., 2001; Loubere et al., 2004), although this interpretation is the subject of intense debate.

Potential temporal decoupling of organic and inorganic proxies, associated with the fine and coarse sediment fractions, respectively, resulting from lateral sediment transport has also been investigated by molecular-level radiocarbon dating (Ohkouchi et al., 2002; Mollenhauer et al., 2003, Mollenhauer et al., 2005). Significant age offsets between a suite of marine organic biomarkers (alkenones) associated with the fine fraction and coarse-grained calcareous foraminiferal tests have been found at sites of high local sedimentation rates. Age offsets of up to 8 kyr occur at the Bermuda Rise, a well-described sediment drift, and have been attributed to the advection of fine-grained sediment from the Canadian continental margin (cf. Keigwin and Jones, 1989; Ohkouchi et al., 2002). Significant age differences between organic matter and foraminifera have also been reported from continental margins where high sedimentation rates are observed in local depo-centers (Mollenhauer et al., 2003, Mollenhauer et al., 2005). Other studies have found independent evidence of lateral displacement of fine-grained particles based on sedimentological considerations (Thomsen et al., 1998; Benthien and Müller, 2000; Freudenthal et al., 2001; Englebrecht and Sachs, 2005).

The Argentine Basin represents a good example of an oceanic area affected by large-scale sediment advection. Several studies have suggested that extensive northward displacement of fine-grained particles and phytoplankton detritus occurs in this region. For example, core-top sediment characteristics such as diatom floral assemblages (Romero and Hensen, 2002) and reconstructed sea-surface temperatures (SSTs) based on the alkenone-derived U37K′ index (Brassell et al., 1986; Prahl et al., 1988; Müller et al., 1998) are inconsistent with overlying surface water conditions (Benthien and Müller, 2000; Fig. 1C). These observations are interpreted as being the result of particle advection via surface currents containing colder, Antarctic waters, or from cooler coastal areas via bottom current transport. Conte et al. (2006) report cold anomalies in SST estimates derived from alkenones in surface waters of the Argentine Basin, arguing for long-range transport via surface currents.

The study area (Fig. 1A) is situated in the southwestern Atlantic Ocean ranging from approximately 35–60° W and from 20–45°S. Bathymetrically, this area is divided into the Brazil Basin in the north and the Argentine Basin in the south, both with maximum depths over 5000 m. The two basins are separated by the Rio Grande Rise and the Santos Plateau. The Vema Channel, a narrow (20 km wide) incision at 4620 m depth allowing for deep water to propagate from the south into the Brazil Basin, separates the Rio Grande Rise from the Santos Plateau and the continental slope.

The upper-level circulation pattern in the study area (Fig. 1B) is dominated by the southward-flowing Brazil Current and the northward-directed Malvinas Current, both flowing along the continental margin (Peterson and Stramma, 1991). The warm-water Brazil Current originates near 10°S, where the South Equatorial Current bifurcates. It is a weak western boundary current carrying warm and salty subtropical water masses with surface velocities between 20 and 60 cm/s. The flow strengthens at more southern latitudes under the influence of a recirculation cell (Peterson and Stramma, 1991). The Malvinas Current originates east of the Drake Passage as a branch of the Antarctic Circumpolar Current (Peterson, 1992). It transports cold and less saline subantarctic waters and encounters the Brazil Current near 38°S, an area referred to as the Brazil–Malvinas Confluence. There both currents separate from the continental margin and flow seaward in a south-eastward direction. In the confluence zone, strong thermal gradients in surface waters are observed, often reaching 1 °C/km (Olson et al., 1988). Direct measurements of current velocities in the Malvinas Current are scarce. Drifter experiments revealed a northward core flow velocity of 60–80 cm/s (Peterson et al., 1996). Highest velocities of up to 100 cm/s in the Malvinas Current region were observed in the southward Brazil–Malvinas Return flow, whose flow path is offshore from the boundary currents (Peterson et al., 1996; Piola et al., 2001, http://www.po.gso.uri.edu/wbc/Piola/bmc.htm).

At the Brazil–Malvinas Confluence, the lower-level circulation is also characterized by the interplay between several water masses (e.g., Reid et al., 1977). The subtropical water masses of the Brazil Current and the subantarctic waters of the Malvinas Current meet in the upper 800 m of the water column. Antarctic Intermediate Water is detected between 500 and 1500 m. Bottom currents underlying the Malvinas Current between 400 and 1600 m depth range between 20 and 30 cm/s (Peterson et al., 1996). North Atlantic Deep Water (NADW) is found at around 2500 m water depth and flows southward along the continental rise until it turns eastward at around 38°S (Maamaatuaiahutapu et al., 1992). Flow velocities in the NADW depth range are less than 5 cm/s (Hogg et al., 1996). Northward flowing Circumpolar Deep Water is separated by NADW into an upper and a lower branch. At depth, Weddell Sea Deep Water, as part of Antarctic Bottom Water (AABW), flows northward near the bottom of the Argentine Basin (Peterson, 1992). While most currents below 800 m depth are relatively sluggish (<5 cm/s), flow of AABW through the Vema Channel can reach velocities up to 25 cm/s (Hogg et al., 1996). The Vema Channel thus is the major conduit for AABW to propagate northward into the Brazil Basin, with transport of up to 4 Sv (Hogg et al. 1999).

Productivity in the study area is highest along the continental margin of the Argentine Basin, in particular off the mouth of the Rio de la Plata (Fig. 1B). Satellite imagery clearly shows the frequent occurrence of eddies spinning off the Brazil–Malvinas Confluence zone. To the south of the confluence zone, nutrient-rich Malvinas Current waters are thus transported offshore. As a result, average annual pigment concentration is higher in the Argentine Basin than in the oligotrophic Brazil Basin (Antoine et al., 1996). Sources of marine particulate organic carbon are therefore larger in the south and west of the study area than in the north and east. This pattern is also reflected in the organic carbon contents of the sediments (see below; Fig. 1D). Possible sources of laterally advected sediments are the shelf areas as described by Michaelovitch de Mahiques et al. (2002) for the Brazil Basin. According to Benthien and Müller (2000), suspended particles can also be advected from the south over distances of up to 1000 km. In addition, strong and dominantly northward-directed bottom currents prevailing in the Argentine Basin can lead to sediment remobilization and redistribution. High kinetic energy related to the Brazil–Malvinas Confluence may induce benthic storms resulting in extensive reworking and sediment redistribution (cf. Hollister and McCave, 1984).

Deep-sea sediments in this region are dominated by carbonates and, below approximately 4000 m water depth, by clays. Input of terrigenous material through the Rio de la Plata is approximately 92 Gt annually (Milliman and Meade, 1983), including significant amounts of terrigenous organic matter (Schlünz et al., 1999; Seiter et al., 2004). Organic carbon concentrations in central basin surface sediments are generally low (<0.5% dry weight), but a local maximum reaching values of up to 2% exists on the continental slope underlying the Malvinas–Brazil Confluence Zone (Fig. 1D; Mollenhauer et al., 2004; Seiter et al., 2004).

In the present paper, we report the first coupled U-series and molecular 14C-based study of sediment supply and deposition. We use U-series isotopes to assess sediment focusing in selected core-top sediments from the Argentine and Brazil Basins. Areas apparently unaffected by northward sediment advection were considered as well as those where the largest anomalies have been reported (Benthien and Müller, 2000; Romero and Hensen, 2002). In addition, we studied radiocarbon ages of coarse-grained foraminifera, total organic carbon (TOC) and alkenones in the same sediment samples. We show that although the 230Th-based focusing factor (Ψ) data imply the occurrence of sediment focusing in areas with large-temperature anomalies, molecular-level radiocarbon analyses do not yield a clear transport signal. We use the data, together with evidence from the literature (Lange, 1985; Conte et al., 2006), to infer that long-range lateral transport of phytoplankton detritus over several hundred kilometers can occur during, or shortly after production in surface waters.

Section snippets

Samples

We used multicorer sediment samples taken from 10 sites on the continental margin and in the deep Argentine and Brazil Basins. Out of those sites, eight have been discussed in the study of Benthien and Müller (2000). Two new sites located near previously studied locations were added (Fig. 1C, Table 1). From each site, we used core-top or near-surface (0–1, 0–2 or 1–2 cm) samples for radiocarbon dating of foraminifera, TOC and alkenones (Table 2). Additional foraminiferal radiocarbon dates were

Radiocarbon analyses

Accelerator mass spectrometer (AMS) radiocarbon ages were determined on TOC and planktic foraminifera for each of the 10 core-top samples. Where abundances were sufficient, molecular-level radiocarbon dates of C37, C38 and C39 alkenones (measured as a composite sample) were also obtained. All radiocarbon measurements were performed at the National Ocean Sciences Accelerator Mass Spectrometry (NOSAMS) facility at the Woods Hole Oceanographic Institution, USA, following standard procedures for

Results and discussion

Core-top radiocarbon ages of foraminifera, TOC and alkenones, 230Thxs normalized fluxes and sediment focusing factors based on 230Thxs are given in Table 2, Table 3 and shown in Fig. 2, Fig. 3, Fig. 4. Despite compelling evidence for lateral transport as an important influence on the sedimentation in the Argentine Basin, in particular on the sedimentation of alkenones (Benthien and Müller, 2000), our data based on organic matter radiocarbon ages do not immediately imply sediment redistribution.

Conclusions

Our study of core-top radiocarbon ages of co-occurring foraminifera, total organic carbon (TOC) and alkenones, and 230Thxs-based sediment focusing factors has helped characterize particle supply and deposition in a dynamic sedimentary environment. Comparison of core-top radiocarbon ages and sediment fluxes has shown that vertical flux is the most important factor in determining core-top ages of foraminifera and, to a certain extent, also of TOC. However, 230Thxs inventories suggest that

Acknowledgments

The crews and captains of R/V Meteor cruises M23/2, M29/1, M29/2, M46/2 and M46/3 are thanked for support during sampling. Technical support was provided by Daniel Montluçon and Li Xu (WHOI, organic geochemistry), Alan Fleer and Susan Brown-Leger (WHOI, radioisotope sample preparation and ICP/MS analysis) and by Hella Buschoff (University of Bremen, TOC and DBD analyses). The NOSAMS staff is thanked for the radiocarbon analyses. Helpful discussions with Roger François and Michiel Rutgers van

References (76)

  • W.R. Martin et al.

    Organic matter cycling in sediments of the continental margin in the northwest Atlantic Ocean

    Deep-Sea Research I

    (2004)
  • J.F. McManus et al.

    Radiometrically determined sedimentary fluxes in the sub-polar North Atlantic during the last 140,000 years

    Earth and Planetary Science Letters

    (1998)
  • A.P. McNichol et al.

    TIC, TOC, DIC, DOC, PIC, POC—unique aspects in the preparation of oceanographic samples for 14C-AMS

    Nuclear Instruments and Methods in Physics Research B

    (1994)
  • L. Mémery et al.

    The water masses along the western boundary of the south and equatorial Atlantic

    Progress in Oceanography

    (2000)
  • M. Michaelovitch de Mahiques et al.

    Post-LGM sedimentation on the outer shelf—upper slope of the northernmost part of the São Paolo Bight, southeastern Brazil

    Marine Geology

    (2002)
  • G. Mollenhauer et al.

    Asynchronous alkenone and foraminifera records from the Benguela Upwelling System

    Geochimica et Cosmochimica Acta

    (2003)
  • G. Mollenhauer et al.

    Organic carbon accumulation in the South Atlantic Ocean: its modern, mid-Holocene and last glacial distribution

    Global and Planetary Change

    (2004)
  • P.J. Müller et al.

    Calibration of the alkenone paleotemperature index U37K′ based on core tops from the eastern South Atlantic and the global ocean

    Geochimica et Cosmochimica Acta

    (1998)
  • D.B. Olson et al.

    Temporal variations in the separation of Brazil and Malvinas Currents

    Deep-Sea Research

    (1988)
  • R.G. Peterson

    The boundary currents in the western Argentine Basin

    Deep-Sea Research

    (1992)
  • R.G. Peterson et al.

    Upper-level circulation in the South Atlantic Ocean

    Progress in Oceanography

    (1991)
  • B. Ransom et al.

    Comparison of pelagic and nepheloid layer marine snow: implicatons for carbon cycling

    Marine Geology

    (1998)
  • O. Romero et al.

    Oceanographic control of biogenic opal and diatoms in surface sediments of the Southwestern Atlantic

    Marine Geology

    (2002)
  • J.A. Sanchez-Cabeza et al.

    Sediment accumulation rates in the southern Barcelona continental margin (NW Mediterranean Sea) derived from 210Pb and 137Cs chronology

    Progress in Oceanography

    (1999)
  • P.H. Santschi et al.

    Boundary exchange and scavenging of radionuclides in continental margin waters of the Middle Atlantic Bight: implications for organic carbon fluxes

    Continental Shelf Research

    (1999)
  • B. Schlünz et al.

    Terrestrial organic carbon accumulation on the Amazonas deep sea fan during the last glacial sea level low stand

    Chemical Geology

    (1999)
  • K. Seiter et al.

    Organic carbon content in surface sediments—defining regional provinces

    Deep-Sea Research I

    (2004)
  • D.A. Siegel et al.

    Trajectories of sinking particles in the Sargasso Sea: modeling of statistical funnels above deep-ocean sediment traps

    Deep-Sea Research I

    (1997)
  • D.O. Suman et al.

    Variations in Holocene sedimentation in the North Atlantic basin determined from 230Th measurements

    Deep-Sea Research

    (1989)
  • L. Thomsen et al.

    Sediment erosion thresholds and characteristics of resuspended aggregates on the western European continental margin

    Deep-Sea Research I

    (2000)
  • C. Thomsen et al.

    Seasonal variability of the long-chain alkenone flux and the effect on the UK’37-index

    Organic Geochemistry

    (1998)
  • J. Thomson et al.

    A comparison of sediment accumulation chronologies by the radiocarbon and 230Thexcess methods

    Earth and Planetary Science Letters

    (1995)
  • J. Thomson et al.

    Implications for sedimentation changes on the Iberian margin over the last two glacial/interglacial transitions from (230Thexcess)0 systematics

    Earth and Planetary Science Letters

    (1999)
  • D.M. Anderson

    Attenuation of millenial-scale events by bioturbation in marine sediments

    Paleoceanography

    (2001)
  • D. Antoine et al.

    Oceanic primary production, 2. Estimation at global scale from satellite (coastal zone color scanner) chlorophyll

    Global Biogeochemical Cycles

    (1996)
  • R.A. Armstrong et al.

    A new, mechanistic model for organic carbon fluxes in the ocean based on the quantitative association of POC with ballast minerals

    Deep-Sea Research II

    (2002)
  • H.W. Arz et al.

    Millenial-scale changes of surface- and deep-water flow in the western tropical Atlantic linked to Northern Hemisphere high-latitude climate during the Holocene

    Geology

    (2001)
  • E. Bard

    Paleoceanographic implications of the difference in deep-sea sediment mixing between large and fine particles

    Paleoceanography

    (2001)
  • Cited by (44)

    • Deep-sea fluid and sediment dynamics-Influence of hill- to seamount-scale seafloor topography

      2013, Earth-Science Reviews
      Citation Excerpt :

      Interactions between the seafloor and its microtopography on the one hand and fluid flow on the other hand, and their influence on sediment dynamics are reviewed in a companion paper and will not be addressed here (Turnewitsch et al. (in prep): Deep-sea fluid and sediment dynamics—Influence of the turbulent boundary layer). For the largest scales, it has been known for several decades that the topographic deep- and open-ocean features, in concert with the Coriolis force, steer deep basin-scale flows (e.g., DWBCs, Antarctic Circumpolar Current (ACC)) (e.g., van Aken, 2007) and influence sediment dynamics in the deep sea by constricting, constraining and deflecting flows near the seafloor (see, e.g., Heezen et al., 1966; Chapter 9 of Heezen and Hollister, 1971; McCave et al., 1980; Gardner and Sullivan, 1981; Richardson et al., 1981; Gardner et al., 1983, 1985; McCave, 1986; Gross et al., 1986, 1988; Ledbetter and Klaus, 1987; Gross and Nowell, 1990; DeMaster et al., 1991; Gross and Williams, 1991; Anderson et al., 1993; Stow et al., 1996; Bianchi and McCave, 2000; Hunter et al., 2007; Koenitz et al., 2008; Mollenhauer et al., 2006,2011; Rebesco and Camerlenghi, 2008, and papers therein). In addition, certain upper continental-slope sections and continental-slope canyons have been known to be associated with the formation and/or reflection of internal waves, in particular internal tides, which may be related to sediment redistribution and the formation of intermediate nepheloid layers (e.g., Hotchkiss and Wunsch, 1982; Cacchione and Drake, 1986; Dickson and McCave, 1986; Gardner, 1989a,b; Cacchione et al., 2002; Shanmugam, 2003; Puig et al., 2004).

    View all citing articles on Scopus
    1

    Now at: Alfred-Wegener-Institute for Polar and Marine Research, Bremerhaven, Germany.

    View full text