Laurentide Ice Sheet Meltwater and the Atlantic Meridional Overturning Circulation During the Last Glacial Cycle: A View From the Gulf of Mexico - Abrupt Climate Change: Mechanisms, Patterns, and Impacts - page 46

1999; Sarnthein, 2000; Hagen and Keigwin, 2002; Rahm-
storf,2002;Clark et al.,2002;Hoogakker et al.,2007].
Other studies have inferred that circulation changes were
greatest during Heinrich stadials [Boyle and Keigwin,1987;
Vidal et al., 1997; Curry et al., 1999; McManus et al., 2004].
In particular, AMOC reduction during Heinrich stadial 1
appears to have been more severe than during the Younger
Dryas [McManus et al., 2004]. It is important to stress that
there are uncertainties in the degree of inferred Atlantic
Ocean circulation change based on different proxies, includ-
ing the Pa/Th proxy data for AMOC. Opal production may
signi cantly affect sedimentary Pa/Th and lead to overesti-

mates of AMOC reduction [Chase et al., 2002; Keigwin and
Boyle, 2008; Lippold et al., 2009]. Nevertheless, it is clear
from 14 C records in Cariaco Basin [Hughen et al., 2000] and
14 C[Keigwin, 2004; Robinson et al., 2005; Keigwin and
Boyle, 2008] and Nd records in the deep Atlantic Ocean
[Roberts et al., 2010] that signi cant AMOC reduction oc-

curred during the H1 stadial and the Younger Dryas relative
to the Bølling/Allerød and the Holocene. This contention
agrees with older nutrient-based proxy data for NADW
volume reduction and Southern Component Water increase,
such as Cd/Ca and δ

13 C records in benthic foraminifera from
the deep western North Atlantic [Boyle and Keigwin, 1987;
Keigwin et al., 1991; Curry and Oppo, 1997]. The latter
records do not require, but are consistent with, AMOC re-
duction. In particular, decreased benthic

13 Cvaluesare
recorded on Ceara Rise during the Heinrich 3, 4, and 5
stadials and interpreted as decreased NADW production
[Curry and Oppo, 1997].
It is now clear that the Heinrich stadials correlate with
some of the major AIM warming events in Antarctica [EPI-
CA Community Members et al., 2006]. Indeed, AIM 1, 4, 8,
and 12 are associated with Heinrich events 1, 3, 4, and 5,
respectively. Heinrich events are classically interpreted as
increased iceberg discharge during particularly severe D-O
stadials [Bond et al., 1993; Hemming, 2004]. Although
Heinrich events appear to have discharged into a cold North
Atlantic, subtropical warming has been observed [Mix et al.,
1986; Rühlemann et al., 1999; Arz et al., 1999; Grimm et al.,
2006; Williams et al., 2010], and “ hidden warming ” in the
Northern Hemisphere has long been suspected as a contrib-
utor to cryospheric instability (S. Hemming, personal com-
munication, 2009). Our reconstructions from the Gulf of
Mexico suggest that LIS meltwater spanned the AIM events
that were associated with Heinrich events 1, 3, 4, and 5
(Figure 8). These records represent perhaps the best evidence
for bipolar warming during AIM events.
Comparing the temporal patterns of meltwater input dur-
ing the last deglaciation and MIS 3 leads to the observation
that major LIS melting was associated with AMOC reduction

δ

18 Osw from
MD02-2551 [Hill et al., 2006; this study] to NGRIP δ

Figure 7. Comparison of ice volume-corrected δ

18 Oand

18 O from 48 to 28 ka. Also shown are Heinrich events 3 - 5,
D-O events 4 - 12 and AIM 4, 8, and 12. Modern Gulf of Mexico
δ

EDML δ

18 Osw is indicated by horizontal line.

7. ICE SHEET MELTWATER AND THE ATLANTIC
OCEAN DEEP CIRCULATION

Modeling studies and North Atlantic sediment records
have suggested that changes in deep circulation were linked
to the D-O events [Dokken and Jansen, 1999; Bond et al.,