Application of elemental and isotopic ratios of bulk sediment and detrital carbonate to identify and source Heinrich events in IODP Exp 303/306 sites.

Lead Research Organisation: University of Cambridge
Department Name: Earth Sciences


Heinrich events, named after Hartmut Heinrich who first described them in 1988, are layers of ice rafted debris (IRD) found in North Atlantic sediment cores dating to the last glaciation. They are believed to have formed by massive discharges of ice bergs during surging of the Laurentide Ice Sheet in the region off Hudson Strait in northern Canada. The ice bergs produced by these events traveled across the entire North Atlantic between ~40 and 55 oN, melted, and dropped sediment onto the seafloor (known as 'ice rafted debris'). The diagnostic feature of Heinrich events is they contain detrital carbonate (limestone and dolomite) derived from large lower Paleozoic basins in northern Canada and northwest Greenland. The melting icebergs produced meltwater that decreased the salinity of North Atlantic surface waters and increased surface water stratification and water column stability. In turn, this increased sea ice formation and diminished rates of deep-water production, resulting in large decreases in temperatures over Greenland and Europe. Several mechanisms have been proposed to explain these remarkable events, but none have met with universal acceptance. Although Heinrich events have been studied for the last glacial cycle, little is known about their occurrence in previous glacial periods of the Pleistocene. Such studies have been hampered by the availability of long continuous cores with high sedimentation rates from the North Atlantic IRD belt. To address this shortcoming, Integrated Ocean Drilling Program (IODP) Expedition 303 recovered long cores from multiple locations in the North Atlantic with high sedimentation rates (5 to 20 cm/kyr). To study these long records, we are faced with the prospect of having to make tens of thousands of measurements at multiple sites to identify Heinrich and other types of IRD events. Such an effort would be impractical in terms of both time and expense for the hundreds of meters of section collected during IODP Exp. 303/306. Instead, our approach has been to utilize automated, non-destructive core logging and scanning techniques to identify Heinrich-like events in previous glacial periods. For example, we have found that some Heinrich events are marked by peaks in Ca/Sr ratio measured by core scanning x-ray fluorescence (XRF), an analytical techniques that measures semi-quantitive elemental data along the surface of the core at high spatial resolution. Here we propose to test and calibrate the XRF elemental ratios and develop new techniques (oxygen isotopes of bulk sediment) for recognizing Heinrich events in North Atlantic sediments using cores recovered by IODP Exp. 303/306. These methods should have widespread application for tracing Heinrich events throughout the North Atlantic IRD belt between ~40 and 55 N. In addition to identifying Heinrich and other IRD events, it is necessary to determine the source of the IRD to characterize the dynamics of the specific ice sheets that have advanced, surged, or collapsed. It is often possible to fingerprint the bedrock over which an ice sheet has moved by analyzing geochemically the IRD in marine sediment cores. Most studies have focused on the silicate fraction of IRD in Heinrich events to determine provenance, but the detrital carbonate grains on which Heinrich (H) events are defined have received less attention. Here we propose to test new methods (using oxygen, strontium, and lead isotopes) to identify the source of detrital carbonate mineral grains contained in IRD layers in North Atlantic sediments. These methods will then be applied to identify and source detrital carbonate layers (Heinrich-like events) in older glacial periods of the Pleistocene to help constrain the origin and cause of these remarkable events in the North Atlantic.


10 25 50
Description The five original objectives of the proposal were successfully achieved:

1. Oxygen isotopic values measured in 166 individual detrital carbonate grains from six Heinrich layers at three sites average _5.6‰ ±1.5‰ (1_; n=166). Because the _18O of detrital carbonate is significantly lower than biogenic carbonate, the _18O of bulk carbonate in North Atlantic sediments is sensitive to variations in the proportion of detrital-biogenic carbonate.

Oxygen isotopes of bulk or fine-fraction carbonate is a powerful tool for tracing sediment layers rich in detrital carbonate in the North Atlantic. Recognition that detrital carbonate contained in Heinrich layers has very low _18O values has important implications for cleaning of foraminifera for stable isotope analysis. Fine-grained detrital carbonate not removed from the inner test chambers will lower foraminifer _18O, leading to overestimation of the reduction in salinity associated with Heinrich events. We have the magnitude of 18O depletion is reduced when foraminifera are rigorously cleaned prior to analysis, which may require the community to reconsider the amount of meltwater delivered to the North Atlantic during Heinrich events. Results are published in Hodell and Curtis (2008).

2. We measured the elemental and strontium isotopic composition of 20 detrital carbonate grains using an ICP-OES and MC-ICP-MS, respectively. Mg/Ca ratios of individual grains
ranged from 9 to 925 mmol mol-1 (nearly pure dolomite), reflecting the relative proportion of dolomite-to-calcite in each grain. Sr/Ca is low, averaging 0.5 mmol mol-1, and ranges from
0.12 to 2.37 mmol mol-1. These values are much lower than ratios of biogenic carbonate (foraminifera and coccoliths), thereby supporting the interpretation that Ca/Sr measured by
core scanning XRF can be used to detect Heinrich layers in North Atlantic sediment cores.

87Sr/86Sr of detrital carbonate in Heinrich events averages 0.70800 +/- 0.00015 (n=20), which is significantly less than modern seawater values (0.70917). We measured lead isotopes of detrital carbonate grains from 12 layers and derived a lead isochron age of 450 +/- 54 Ma, consistent with ages reported for detrital carbonates from Hudson Strait and Baffin Bay. We
conclude that Sr and Pb isotopes in detrital carbonate grains are useful tools for establishing the origin of these grains in North Atlantic sediments.

3. We successfully calibrated the scanning XRF data from the split-core sediment surface by analyzing dried, homogenized powders of 50 samples from Sites U1308 using a conventional

XRF analyzer (Spectro Xepos EDPXRF) at the University of Bremen. The samples selected included a range of Ca, Si and Sr values and ratios and included Heinrich events 1 through 6. Comparisons of (a) Ca/Sr and (b) Si/Sr data obtained by scanning XRF of the split core surface and dried homogenized samples measured by conventional XRF showed excellent agreement. The calibration data were published as an appendix in Hodell et al. (2008).

4. We applied bulk _18O and Ca/Sr measured by scanning XRF to identify detrital carbonaterich IRD layers in older glacial periods of Site U1308 (re-occupation of DSDP Site 609). Our
study has provided some of the first evidence for the occurrence of Heinrich events in older glacial period of the Pleistocene. The first Heinrich layer appeared in the sedimentary record at ~640 ka during MIS 16, coinciding with increased dominance of the 100-ka cycle in the benthic d18O signal at end of the Middle Pleistocene Transition. Prior to 650 ka, IRD delivery to Site U1308 was more frequent and dominated by silicate minerals (quartz, feldspar, etc.).

The first occurrence of Heinrich layers in MIS 16 is significant because it provides evidence for a possible change in the dynamic behavior of the North American Ice Sheet with the onset
of the 100-kyr cycle at the end of the MPT. We have successfully obtained a NERC standard grant to further study changes in ice volume and ice sheet dynamics across the MPT.

5. We have begun to apply the newly developed elemental and isotopic tools to test if detrital carbonate layers in older glacial periods and in other North Atlantic cores have the same elemental/isotopic fingerprint as those derived from Hudson Strait. We are also investigating the source of the IRD prior to 650 ka, because IRD was more frequent, lacked detrital carbonate, and was dominated by silicate minerals prior to the MPT.
Exploitation Route see above
Sectors Environment