Glacial-to-interglacial variations in the deep water at the Bermuda Rise inferred from a Nd isotope record covering the last million years

<p>The formation of North Atlantic Deep Water (NADW) in the North Atlantic is an important modulator of the climate system, as it drives the global termohaline circulation, responsible for the distribution of heat, salts and nutrients across the oceans. ODP Site 1063 (4584 m), on the deep Bermuda Rise, is located in the mixing zone between NADW and Antarctic Bottom Water (AABW) and appears to be a good location to study how ocean circulation and climate interconnect. Here we present a new record based on Nd isotope ratios that covers ~1 Ma at that Site. Our data shows Nd isotope ratios during parts of interglacials that are much lower than present day NADW. These results are coherent with recent published studies on the last interglacial–glacial cycle that show that the deep North Atlantic Nd isotope ratios are also lower than NADW during the early interglacial. However, Nd isotope values from the shallower DSDP Site 607 (3427 m), within the core of NADW, have remained similar to modern NADW during interglacials over the same time interval. Site 607 is thought to represent the deep North Atlantic, as shown by an Atlantic meriodional transect that displays Nd isotopes ratios for glacial and interglacial maxima over the last ~1 Ma. We suggest that Nd isotope ratios at Site 1063 do not fully represent the North Atlantic endmember of the AMOC during interglacials, but regional or local processes. However, glacial values at Site 1063 fitting those of Site 607 suggest that Nd isotope ratios represent, indeed, water mass mixing during glacial periods. The low Nd-isotope ratios in the deep Bermuda Rise during interglacials would be the result of particle-seawater exchange derived from the arrival of freshly ground, poorly weathered bedrock from the Canadian shield to the North Atlantic during major ice sheet retreats, such as deglaciations as well as stadial-to-interstadial transitions. Consequently, a deep, regionally constrained layer of seawater is tagged with this extreme Nd isotope signature that is not representative of the AMOC. We suggest that a benthic nepheloid layer, whose development is driven by a deep-recirculating gyre system regulated by the interaction between the northward flowing Gulf Stream and the southward flowing deep western boundary current, facilitates the periodical masking of the deep Atlantic Nd isotope signature at Site 1063. The intermittence of the masking allows for a speculation on how the deep-recirculating gyre system might have changed over the last ~1 Ma glacial-to-interglacial cycles.</p>

231Pa/230Th in the northwestern Atlantic: circulation versus particles?

<p>In 2004 McManus et al. published their famous <sup>231</sup>Pa/<sup>230</sup>Th record from the Bermuda Rise revealing millennial-scale changes in circulation strength back to the Last Glacial Maximum. This record marks the boost of this proxy as a kinematic circulation change proxy for the Atlantic Ocean and the initial rising slope on the ‘Elderfield-Curve’. However, the up-to-date data base of Atlantic sedimentary <sup>231</sup>Pa/<sup>230</sup>Th records gives a rather inconsistent picture of changes in the circulation strength in the Atlantic throughout the past 25 ka (Ng et al., 2018). Since both radioisotopes are strongly particle reactive it is obvious that scavenging processes may play a major role in their cycling as well. At ocean margins such processes do have a major impact on <sup>231</sup>Pa/<sup>230</sup>Th, leading to increased values and thus potentially overprinting the circulation signal. In contrast, records from open ocean sites are assumed to show a less biased circulation signal. In addition, the GEOTRACES program (Schlitzer et al., 2018) provided valuable seawater data allowing for examining the cycling of both radioisotopes under today’s circulation regime in more detail. A transect across the North Atlantic by Hayes et al. (2015) revealed that nepheloid layers contribute to strong bottom scavenging of <sup>231</sup>Pa and <sup>230</sup>Th in the northwestern Atlantic basin. Surprisingly, sedimentary core-top values do not mirror predominant scavenging effects but rather indicate a strong export of <sup>231</sup>Pa and therefore a circulation signal. With our modern proxy toolbox, it is impossible to reconstruct the occurrence and intensity of past nepheloid layers and hence their potential effect on recorded <sup>231</sup>Pa/<sup>230</sup>Th variations. Therefore, isotope-enabled models may help to better decipher the interwoven processes controlling <sup>231</sup>Pa/<sup>230</sup>Th (Rempfer et al., 2017; Lerner et al., 2019). Here an up-to-date compilation of northwestern Atlantic <sup>231</sup>Pa/<sup>230</sup>Th data will be presented. Our findings base on records covering the last 25 ka and will be interpreted in the context of recent model simulations as well as compared to seawater data. Thus, we aim for a deeper understanding of <sup>231</sup>Pa and <sup>230</sup>Th cycling in the northwestern Atlantic.</p><p>References:</p><p>Hayes, C., et al. (2015), Deep-Sea Res. Pt. II, 116, 29-41.<br>Lerner et al. (2020), Deep Sea. Res. Pt. I, 155, 1-41.<br>McManus, J. F., et al. (2004), Nature, 428, 834-837.<br>Ng, H., et al. (2018), Nat. Comm., 9, 1-10.<br>Rempfer et al. (2017), EPSL, 468, 27-37.<br>Schlitzer, R., et al. (2018), Chem. Geol., 493, 210-223.</p><p> </p>

Disentangling the effects of particles and circulation on 231Pa/230Th during Heinrich Stadials

<p>It has been shown that during Heinrich stadials northern deep water production ceased leading to an enhanced inflow of southern sourced water. Although Heinrich events are not considered to represent the primary trigger of Heinrich stadials the reorganisation of Atlantic ocean dynamics during their occurrences is an active field of research. In particular, Heinrich stadial 2 (HS2) is of high interest, based on the observation that the interplay with the climate system was very different during HS2 compared to HS1, although the magnitude of iceberg and freshwater discharge was similar (Hemming, 2004). During HS2 sea-level was still decreasing while the atmospheric CO­<sub>2</sub> content was relatively stable unlike the climatic evolution during Heinrich HS1.</p><p>The notion of a reduced Atlantic Meridional Overturning Circulation (AMOC) during Heinrich Stadials is mainly strengthened by the <sup>231</sup>Pa/<sup>230</sup>Th records from the Bermuda Rise. However, other influencing factors, capable of increasing the sedimentary <sup>231</sup>Pa/<sup>230</sup>Th without according decreases in AMOC strength, need to be considered as well. Besides biogenic opal, high dust fluxes may also result in enhanced scavenging rate of both radionuclides and consequently higher sedimentary <sup>231</sup>Pa/<sup>230</sup>Th signals, since another distinct feature that accompanies Heinrich Stadials is the high atmospheric concentration of dust in the northern hemisphere. Furthermore, high dust concentrations might be an indicator of a vigorous wind system and therefore strong ocean mixing, which can lead to the enhanced formation of nepheloid layers These layers are suspected to cause strong bottom scavenging and consequently high sedimentary <sup>231</sup>Pa/<sup>230</sup>Th. Very high dust fluxes were observed e.g. during HS2 and MIS4. Here, we compare <sup>231</sup>Pa/<sup>230</sup>Th with dust records in order to disentangle the effects of scavenging and circulation on the recorded sedimentary <sup>231</sup>Pa/<sup>230</sup>Th from the northwestern Atlantic.</p>

Late Cenozoic architecture of the St. Pierre Slope

The late Cenozoic seismic stratigraphy of the continental slope south of western Newfoundland is interpreted using new seismic reflection profiles. New Miocene–Pliocene biostratigraphic (palynology) age determinations on the Hermine E-94 well on the northwestern Grand Banks of Newfoundland are correlated to the study area. The Quaternary section of St. Pierre Slope is disrupted by numerous failure scarps and mass-transport deposits, but correlation from the mid- slope to the continental rise is achieved using major mass-transport deposits as markers. On the upper slope, stacked downslope-thinning wedges of acoustically incoherent sediment are interpreted as till deposits of mid- to late Pleistocene age. Sedi mentation rates in the youngest part of the succession are estimated from a 30 ka radiocarbon date 25 m below the horizon of the youngest till tongue, which is exposed on a 60 m deep failure surface. Extrapolation of sedimentation rates and comparison with dated sections on the J-Anomaly Ridge and Bermuda Rise provides a consistent interpreted age model for the till tongues that corresponds to marine isotope stages 2, 4, 6, 8, 10, and 12.