Geochemical response of the mid-depth Northeast Atlantic Ocean to freshwater input during Heinrich events 1 to 4

Abstract

Introduction 3The climate of the last glacial period was punctuated by a number of pronounced events with near global impacts, known as Heinrich (or H-) events (e.g. Heinrich, 1988;Hemming, 2004). During these events, transient catastrophic collapses of the North American Laurentide Ice Sheet (LIS), centered over the Canadian Hudson Bay, produced armadas of icebergs that travelled through the Hudson Straight into the North AtlanticOcean, eventually depositing large volumes of ice-rafted debris (IRD), including distinctive detrital limestone clasts, in a belt across the North Atlantic (Ruddiman, 1977).LIS surging, together with iceberg calving from other circum-North Atlantic and Arctic ice sheets during H-events added large volumes of fresh (low density) water to the (sub)polar oceans with important implications for northern-sourced deep water formation and the Atlantic Meridional Overturning Circulation (AMOC). The AMOC exerts a strong control on regional and global climate on orbital to suborbital timescales, influencing latitudinal heat budgets (e.g. Boyle and Keigwin, 1987;Broecker and Denton, 1989) and partitioning of carbon between the atmosphere and deep ocean (e.g. Adkins, 2013;Sigman and Boyle, 2000;Sigman et al., 2010). Understanding the response of AMOC to fresh water addition is therefore crucial, particularly in light of recently accelerating mass loss from the Greenland Ice Sheet and other circum-Atlantic ice masses (Gierz et al., 2015; Vaughan et al., 2013).Previously, it was suggested that fresh water inputs at high latitudes during Hevents caused a cessation of deepwater formation in the North Atlantic, resulting in a shoaling of the northern component water overturning cell by ~1 km (e.g. Alley et al., 1999;Sarnthein et al., 1994;Swingedouw et al., 2009). Support for this concept of a near complete shutdown of AMOC during H-events came from two main lines of evidence: (i) 4 a dramatic drop in circulation vigour (e.g. McCave et al., 1995a;McManus et al., 2004) and (ii) water mass provenance reconstructions (including those based on δ 13 C, Cd/Ca and 14 C) indicating an increased presence of nutrient-rich southern-source waters (SSW) below ~2 -2.5 km depth in the North Atlantic Ocean (e.g. Keigwin et al., 1991;Robinson et al., 2005;Stern and Lisiecki, 2013;Vidal et al., 1997). More recently, however, both of these arguments have been questioned. 231 Pa/ 230 Th xs records from sites across a wide range of water depths suggest that overturning persisted at shallower depths during H1 (Bradtmiller et al., 2014;Gherardi et al., 2009;Lippold et al., 2016), with values compatible with a near complete shutdown only identified in the SSW cell and during H-events close to glacial maxima Lippold et al., 2009;McManus et al., 2004). A further recent development is the documentation of a poorly ventilated water mass in the Nordic Seas during the last glacial period that overflowed the Greenland-Scotland Ridge (GSR) into the Atlantic basin during the deglaciation (Thornalley et al., 2015). This discovery means...

The North Atlantic Glacial Eastern Boundary Current as a Key Driver for Ice‐Sheet—AMOC Interactions and Climate Instability

“…These sedimentological results agree with an active upper AMOC cell, although vertically contracted, persisting during the last glacial period as previously shown by various, independent geochemical proxies (i.e., benthic δ 13 C, ε Nd , Cd/Ca, 231 Pa/ 230 Th; Boyle & Keigwin, 1987; Curry & Oppo, 2005; Duplessy et al., 1988; Gherardi et al., 2009; Lippold et al., 2012; Marchitto & Broecker, 2006; Oppo & Lehman, 1993). This is also supported by our benthic δ 13 C data that show better ventilated waters during the glacial than the Holocene above ∼2,000 m whereas the opposite is observed below this depth (Figure ), in agreement with studies from both the European margin (Crocker et al., 2016; Peck et al., 2006; Willamowski & Zahn, 2000; Zahn et al., 1997) and the western North Atlantic (Curry & Oppo, 2005; Keigwin, 2004; Lynch‐Stieglitz et al., 2014; Zahn & Stüber, 2002).…”

“…Our foraminiferal ε Nd records reveal core-top (i.e., upper Holocene) signatures of −10.6 ± 0.1 (n = 1) at ∼1,000 m and −11.3 ± 0.2 (n = 3) at 2,450 m (Figure 4d and Table S6). These ε Nd values are consistent with those of modern regional seawater and recent coral data from the French Atlantic margin (Copard et al, 2011;Dubois-Dauphin et al, 2017) as well as with foraminiferal data from the northern NE Atlantic (Crocker et al, 2016;Roberts & Piotrowski, 2015). Our results thus capture the modern regional water mass structure of the NE Atlantic with MOW overlying well-mixed NADW ( Figure S9).…”

“…Using multiple independent geochemical tracers of water mass properties (including foraminiferal ε Nd values of −9.7 ± 1.1 on the ∼45–20 ka interval) at ODP Site 980 (Rockall Bank), Crocker et al. (2016) have shown persistent ventilation of the glacial NE Atlantic down to ∼2,200 m by NSW. The consistency of the ε Nd signatures at ODP Site 980 with those recorded on the French Atlantic slope at ∼1,000–2,450 m (−9.2 ± 0.5 on the same interval; n = 46), and the similarity of our carbon isotopic data ( C. wuellerstorfi ; δ 13 C of ∼1.2–1.6‰ at ∼1,000 m and ∼0.6‰ at 2,450 m; Figure ) with the LGM distribution of benthic foraminiferal δ 13 C in the North Atlantic (Curry & Oppo, 2005; Duplessy et al., 1988; Sarnthein et al., 1994), strengthen the hypothesis that the French Atlantic margin at intermediate depths was bathed by NSW during the last glacial period.…”

Origin of Abyssal NW Atlantic Water Masses Since the Last Glacial Maximum

“…Taking into account the southward shifted and reduced deep water formation in the Greenland Sea during the LGM, Iceland Scotland Overflow Water (ISOW), and Denmark Strait Overflow Water may have been weakened and shoaled (Fagel et al, ; Millo et al, ). Two studies suggested persistent overflow from the GIN Seas during the LGM reporting water with relatively high εNd passing the Wyville‐Thomson Ridge (Crocker et al, ; Crocket et al, ). However, if this water mass indeed descended deep enough in the subpolar North Atlantic moving as a deep boundary current analogous to today, it would first have passed through the Labrador Sea where it would have been admixed into less radiogenic deep waters.…”

On climate and abyssal circulation in the Atlantic Ocean during late Pliocene marine isotope stage M2, ∼3.3 million years ago

“…This suggestion invokes a deep Eastern Atlantic route for this water-20 mass to the Angola Basin. This possibility cannot be ruled out for our study intervals based on our Nd data alone because today both Nordic Sea overflows and AABW share a radiogenic Nd signature (of ~ -7 to -9; Crocker et al, 2016). Yet [CO3 2-] is markedly higher for modern NGS overflow water (>115 µmol kg -1 ) than for SCW (~85 µmol kg -1 ) and this contrast would have been greater during the Last Glacial Maximum than today (Yu et al, 2008;Crocker et al, 2016).…”