Distribution, thickness and origin of Heinrich layer 3 in the Labrador Sea

https://doi.org/10.1016/S0012-821X(02)01047-6Get rights and content

Abstract

The presence of Heinrich layer 3 (HL-3) in the northwest Labrador Sea has been debated in the literature. Calypso giant piston core MD99-2233, five new standard piston cores, and re-interpretation of 34 cores from previous cruises confirm the presence of HL-3 in the Labrador Sea. It is identified by high total carbonate concentration (up to 45%), an increase in coarse fraction content, and lighter δ18O values in polar species planktonic foraminifer Neogloboquadrina pachyderma (left-coiling) as low as 3.1‰. The age of HL-3 of ∼27 ka was bracketed in the various cores by about 50 14C-accelerator mass spectrometer dates. Where it is present in ice-proximal regions, it consists of nepheloid-flow deposits at the base and mud turbidites at the top. The thickness of HL-3 varies between 4.8 m (proximal to Hudson Strait) and 0.9 m (distal), decreasing rapidly seaward. On the upper continental slope, HL-3 was too deeply buried to be sampled. Elsewhere, HL-3 is absent in some cores, probably due to slumping or erosion associated with sandy turbidity currents or debris flows.

Introduction

Late Pleistocene sediments from the North Atlantic Ocean contain a sequence of layers of ice-rafted debris (IRD) known as Heinrich layers [1], [2], which are interlayered with hemipelagic and pelagic sediments. The Heinrich layers document episodes of massive iceberg discharge from the Laurentide Ice-Sheet. Ages of Heinrich events have been determined in the North Atlantic Ocean by radiocarbon dating and extrapolation of sedimentation rates and are summarized by Bond et al. [3]. Heinrich layers 1 and 2 were first systematically described in the Labrador Sea by Andrews and Tedesco [4] and deeper Heinrich layers were reported by Hillaire-Marcel et al. [5], Hiscott et al. [6], and Andrews et al. [7]. The origin, presence, and distribution of Heinrich layer 3 (HL-3) in the Labrador Sea has been debated in the past. For example, Grousset et al. [8] stated that HL-3 is restricted to the central and eastern part of the northern North Atlantic Ocean. Based on a limited number of cores, Andrews et al. [7] and Kirby and Andrews [9] suggested that HL-3 is absent in the northwest Labrador Sea. In a re-examination of the cores from the North Atlantic studied by Bond et al. [10], Bond and Lotti [11] identified a narrow peak of carbonate grains at the level of HL-3 in Orphan Knoll core EW9303-GGC31 and subsequently found equivalent narrow detrital carbonate peaks in DSDP Site 609 and core V23-81 in the eastern North Atlantic [3] not seen in earlier work on those cores [10]. Hillaire-Marcel et al. [5] also noticed the presence of HL-3 in two cores (Hu90-13-12 and 13) from the Greenland slope.

In the northwest Labrador Sea, eight Heinrich events, H0 to H6 (including H5a [12]), have been identified between 10 and 60 ka by their known ages, distinct nepheloid-flow deposit structure, high total carbonate concentration, increased coarse fraction content, mostly light oxygen isotope ratios, and low magnetic susceptibility values [13]. Using 40 cores (Fig. 1; Table 1), 34 of which are from our core collection, this study attempts to delineate the regional extent of HL-3 and map its thickness. The significance of establishing the distribution and thickness of HL-3 in the Labrador Sea is that it would demonstrate the extent to which iceberg discharge was similar and from the same sources during different Heinrich events, and in particular whether Hudson Strait was a primary source also for HL-3. For the absence of HL-3 in certain cores, alternative causes rather than the lack of iceberg discharge from the Hudson Strait have been considered, which are discussed in this contribution.

HL-3 is identified by bracketing 14C-accelerator mass spectrometer (AMS) dates of 26 and 28 ka, by high total carbonate content (up to 55%) and high L-values of reflectance, low magnetic susceptibility, δ18O values mostly lighter than 3.1‰ in polar species Neogloboquadrina pachyderma (left-coiling; hereafter N. pachyderma (l)), and an increase in IRD [4], [13]. Sedimentologically, it is characterized by typical nepheloid-flow deposit structure [14], consisting of graded mud layers that contain dispersed IRD and are occasionally interrupted by thin laminae of IRD. The basal boundary with the underlying hemipelagic sediment is sharp (in some cores erosional). The upper boundary is either gradual to overlying hemipelagic sediments or in sharp contact with overlying parallel laminated, carbonate-rich, mud turbidites devoid of IRD.

Section snippets

Cores from which HL-3 has been reported as absent in the Labrador Sea

Andrews et al. [7] and Kirby and Andrews [9] in their study of four cores, Hu75-54, Hu75-55, Hu75-56, and Hu87-033-09 (from water depths of 1169, 2410, 2434, and 1120 m, respectively), from the northern Labrador slope suggested that HL-3 was absent in these cores. In all four cores, HL-0, HL-1, and HL-2, where present, were identified by these authors by their prominent peaks in total carbonate content, increase in coarse fraction content, low magnetic susceptibility, lighter δ18O values in N.

Records of HL-3 in the Labrador Sea

From the 40 cores used in this study, three cores are chosen for graphic presentation of available data based on the location, quality, and integrity of the stratigraphic record, and the availability of 14C-AMS dates. Cores MD99-2233 and Hu90-29 were retrieved from the Labrador rise at water depths of 2350 and 2895 m and are 2462 and 1100 cm long, respectively (Fig. 1, Fig. 3). Core Pa96-06 was collected from Flemish Pass in the southern Labrador Sea at a water depth of 1085 m and is 789 cm

Distribution and thickness of HL-3 in the Labrador Sea

Data presented here (Fig. 4) clearly show that HL-3 is present in 24 out of 40 cores. Eight cores had insufficient penetration to reach HL-3 and in eight cores HL-3 appeared to be missing. Radiocarbon dates (14C-AMS) were obtained wherever possible from both the base and the top to identify HL-3 and bracket its duration. With the aid of this tool it can be shown that HL-3 is absent in three areas: in cores Hu92-01 and 02 on the upper Labrador Slope, in cores Hu92-04 and 05, Hu90-25 and 26,

Re-interpretation of previously studied cores from the northwest Labrador Sea

Of the cores previously studied by Andrews et al. [7] and Kirby and Andrews [9], the Heinrich layer identified as HL-4 in two cores (i.e., the Hu75-56 and Hu87-033-09), could be either HL-3 or HL-4. In the remaining cores, HL-3 may in fact be missing. In core Hu75-56, the two radiocarbon ages of 37 935±1020 and 33 615±600 at depths of 350 and 375 cm, respectively, show a stratigraphic reversal (Fig. 2), implying that at least one of them is unreliable. If the younger age were reliable, the

Why is HL-3 absent in some cores and why was it not found in some of the earlier NW Labrador Sea cores?

HL-3 is absent in a number of cores for various reasons.

  • 1.

    In high sedimentation rate cores, Heinrich layers older than HL-2 were not penetrated. This is the case for the upper Labrador slope cores Hu97-16, 14, 13, 10 and 09 but also for lower slope and rise cores Hu75-55, 56, 60, 61, and 62 (Fig. 2).

  • 2.

    Heinrich layers are absent from cores Hu92-01 and -02 from the upper slope, perhaps because these locations were landward of the Labrador Sea iceberg trajectory or may have experienced permanent

Discussion

The detrital carbonate content identified in the Labrador Sea and North Atlantic Heinrich layers originated largely from the Hudson Strait outlet of the Laurentide Ice Sheet, which acted as a major conduit for the subglacial drainage of the Hudson Bay, Foxe Basin, Baffin Island and surrounding areas. These areas are mostly underlain by Paleozoic carbonates. However, petrographic and geochemical studies [11], [24], [25] have implicated East Greenland as another source in addition to the

Conclusions

The debate about HL-3, caused by suggestions concerning its absence in some Labrador Sea cores, has led to the identification of HL-3 in a large number of additional cores. The provenance of HL-3 in the Labrador Sea is from the same Hudson Strait source as that of the other Heinrich layers. It is identified based on age, its typical nepheloid-flow deposit sedimentary structure together with high total carbonate concentration, high coarse fraction content and lowered δ18O values. On the upper

Acknowledgments

This research was supported by grants from the Natural Sciences and Engineering Research Council of Canada (NSERC), and the National Science Foundation (NSF) of the United States to R.H., D.J.W.P., and H.R. H.R. thanks NSERC, NSF, Fonds pour la Formation de Chercheurs et l’Aide à la Recherche (FCAR), Québec, and McGill University (Carl Reinhardt major fellowship), for financially supporting his research through doctoral scholarships and funding his research. Piston cores were obtained on the

References (31)

  • M.E. Kirby et al.

    Mid-Wisconsin Laurentide Ice Sheet growth and decay implications for Heinrich events-3 and 4

    Paleoceanography

    (1999)
  • G.C. Bond et al.

    Evidence for massive discharges of icebergs into the North Atlantic Ocean during the last glacial period

    Nature

    (1992)
  • G.C. Bond et al.

    Iceberg discharges into the North Atlantic on millennial time scales during the last glaciation

    Science

    (1995)
  • H. Rashid, R. Hesse, D.J.W. Piper, Evidence for an additional Heinrich event between H5 and H6: global implication,...
  • H. Rashid, R. Hesse, D.J.W. Piper, Origin of unusually thick ice-proximal Heinrich layers H-1 to H-3 in the northwest...
  • Cited by (0)

    View full text