Quaternary Research 65 (2006) 223 – 231
www.elsevier.com/locate/yqres
Spatial and temporal variability of Holocene temperature in the
North Atlantic region
Michael R. Kaplan a,*, Alexander P. Wolfe b
b
a
School of GeoSciences, University of Edinburgh, Edinburgh, EH8 9XP, Scotland
Department of Earth and Atmospheric Sciences, University of Alberta, Edmonton, AB, Canada T6G 2E3
Received 4 November 2004
Available online 30 January 2006
Abstract
The early Holocene climate of the North Atlantic region was influenced by two boundary conditions that were fundamentally different from the
present: the presence of the decaying Laurentide Ice Sheet (LIS) and higher than present summer solar insolation. In order to assess spatial and
temporal patterns of Holocene climate evolution across this region, we collated quantitative paleotemperature records at sub-millennial resolution
and synthesized their temporal variability using principal components analysis (PCA). The analysis reveals considerable spatial variability, most
notably in the time-transgressive expression of the Holocene thermal maximum (HTM). Most of the region, but especially areas peripheral to the
Labrador Sea and hence closest to the locus of LIS disintegration, experienced maximum Holocene temperatures that lagged peak summer
insolation by 1000 – 3000 years. Many sites from the northeastern North Atlantic sector, including the Nordic Seas and Scandinavia, either
warmed in phase with maximum summer insolation (11,000 – 9000 years ago) or were less strongly lagged than the Baffin Bay – Labrador Sea
region. These spatially complex patterns of Holocene climate development, which are defined by the PCA, resulted from the interplay between
final decay of the LIS and solar insolation forcing.
D 2005 University of Washington. All rights reserved.
Keywords: Holocene; North Atlantic Ocean; Laurentide Ice Sheet; Insolation; Principal components analysis
Introduction
Delineating spatial and temporal patterns of past climate
variability is one approach to assessing interactions between
individual climate forcing mechanisms. The history of the North
Atlantic Ocean is closely associated with changing ice sheet and
ocean dynamics, which may have global effects, making this a
critical region for understanding large-scale patterns of climate
variability (Rahmstorf, 1997; Clark et al., 2002). Although the
role of ice sheet dynamics in climate change is evident for the
glacial world (e.g., Bond et al., 1992), it is perhaps underappreciated for the early Holocene Epoch, with the notable
exception of discrete short-lived events (Bond et al., 1997;
Barber et al., 1999). Recently, the time-transgressive character of
the Holocene thermal maximum (HTM)—the interval of peak
Holocene warm—has been related directly to the proximity of
the decaying LIS (Kaufman et al., 2004). Although similar
* Corresponding author.
E-mail address: [email protected] (M.R. Kaplan).
conclusions are apparent from a number of additional sites (e.g.,
Koc˛ et al., 1993; Levac et al., 2001), these records have not yet
been integrated into a regional synthesis spanning the North
Atlantic basin.
The goal of the present study is to examine objectively the
broad spatial and temporal patterns of climate change during the
Holocene Epoch for the North Atlantic sector, beginning in the
Boreal chronozone 10,000– 9000 cal yr B.P. Our goal is to
synthesize continuous records of Holocene climate development, in contrast to the approach of temporal snapshots or ‘‘timeslices’’ (e.g., CAPE, 2001). The analysis provides a new
framework with which to consider the variability of proxy
climate records associated with regional differences in Holocene
climate evolution. In an attempt to address the combined effects
of final disintegration of the LIS and elevated summer
insolation, emphasis is placed on the early to middle Holocene
interval. Northern Hemisphere summer insolation 9000 years
ago was higher than present by 7.2% at 35-N, and by 7.5% at
70-N (Kutzbach, 1981), but winter insolation was lower by
¨6% (Berger and Loutre, 1991). However, in the high northern
0033-5894/$ - see front matter D 2005 University of Washington. All rights reserved.
doi:10.1016/j.yqres.2005.08.020
224
M.R. Kaplan, A.P. Wolfe / Quaternary Research 65 (2006) 223 – 231
latitudes with reduced (or absent) winter sunlight, summer
insolation changes are of far greater climatic significance.
Furthermore, most paleoclimate proxies, including the majority
used in the present study, primarily reflect summer conditions
(Overpeck et al., 1997).
Methods
To understand better Holocene climate trends across the
North Atlantic sector, published paleotemperature records were
collated from both land and sea (Fig. 1; Table 1). The criteria for
Figure 1. Paleogeography of the early Holocene LIS at 9000 and 8000 cal yr B.P., and major North Atlantic surface currents (a), shown alongside the Holocene
decrease of LIS area, which includes modern glaciers (b). Both panels a and b are adapted after Dyke et al. (2003). Locations of Holocene proxy records
considered in this study are shown in panel c, with bolded sites representing those included in the PCA (Table 1). In panel d are Holocene insolation anomalies,
relative to current values, for the summer months (Berger and Loutre, 1991).
M.R. Kaplan, A.P. Wolfe / Quaternary Research 65 (2006) 223 – 231
inclusion of a given record included (1) spanning the last 10,000
years with sufficient chronological control and sample quantity
to provide sub-millennial resolution; and (2) the expression of
proxy data explicitly as paleotemperature. This resulted in the
inclusion of thirteen published records (as of 2004), encompassing a broad range of physically, geochemically, and
biologically-inferred temperatures (Table 1). The data are almost
equally distributed between terrestrial (n = 6) and marine (n = 7)
realms. Due to the paucity of paleotemperature records in the
northwest North Atlantic sector, we included one record from
this area that is temporally discontinuous between 8000 and
7300 cal yr B.P. (Baffin Bay; Levac et al., 2001). In this case, a
linear interpolation was used to complete the record. Almost all
records in Table 1 capture summer conditions specifically, with
the exception of GISP2 ice core paleothermometry, which
probably integrates mean annual temperature (Cuffey and Clow,
1997). For the few studies in which seasonal temperatures have
been reconstructed separately (Koç et al., 1993; Penalba et al.,
1997; Levac et al., 2001), similar overall trends are apparent,
suggesting minimal seasonal differences within the paleotemperature reconstructions. Although the total number of records is
not large, we assume they nonetheless reflect regional Holocene
climatic trends as best as currently possible, given the limited
number of paleotemperature records in the literature and the
range of assumptions implicit to each methodology for deriving
temperature estimates.
All records were placed on a common timescale in calibrated
radiocarbon years before present (i.e., cal yr B.P. = calendar year
B.P. (1950)) and reconstructed temperatures recorded from the
rescaled time-series every 250 years between 9750 and 750 cal
yr B.P. The most recent 750 years have been omitted from the
analysis because many records, especially those from marine
225
cores, do not preserve intact uppermost core tops. Given the
variable quality of original 14C chronologies, the estimated
temporal precision of this study is T500 years, that is, twice the
250-year resolution of our resampling scheme. This is sufficient
to address the broad regional patterns of climate change that are
the focus of the investigation but precludes the analysis of shortlived (sub-millennial) climate oscillations, which are treated in
detail elsewhere (e.g., Mayewski et al., 2004). To alleviate
between-site differences in the amplitudes of paleotemperature
reconstructions, all resulting time-series were converted to
standardized deviations from their means, or z-scores, where
z = (x m)/r, and x, m, and r are the raw paleotemperature
data, mean values for individual time-series, and their
standard deviations, respectively. This approach follows the
rationale used in other studies that integrate different proxy
types (Overpeck et al., 1997).
In order to assess objectively the underlying patterns of
variation within the collated records, principal components
analysis (PCA) was applied to the standardized paleotemperature
time-series, in 250-year time-steps. PCA aims to reduce the
complexity of multivariate data into a few interpretable directions
of variability, or principal components (PCs), that represent
synthetic variables that explain cumulative, but independent,
proportions of variance within the raw data (ter Braak and
Prentice, 1988). Here, PCA is based on a centered covariance
matrix of the z-scores (13 sites, 37 samples per site), computed
with CANOCO version 4 (ter Braak and Šmilauer, 1998).
Results
The z-scores of the paleotemperature records are shown in
Figure 2. In almost all records, the timing of peak warmth
Table 1
Proxy records considered in this study
Site name
Latitude
Longtitude
Proxy type
Source
Paleotemperature records
Northern Baffin Bay composite
LeHave Basin adjacent to Nova Scotia
Eagle Lake, Labrador
Lake Hope-Simpson, Labrador
GISP2 Greenland summit
Core PS 21842, Kolbeinsey Ridge
Core HM 57-5, Iceland Plateau
Core NA 87-22, Rockall Plateau
Core SU 81-18 adjacent to Portugal
Northern Iberian Peninsula
Vøring Plateau
Scandes Mountains
Northern Finland
77-16VN
43-46VN
54-14VN
52-27VN
72-58VN
69-28VN
69-26VN
55-28VN
37-46VN
42-02VN
66-58VN
63-N
68-42VN
74-19VW
63-43VW
58-33VW
56-20VW
38-46VW
16-32VW
13-07VW
14-41VW
10-11VW
3-01VW
7-38VE
12-E
22-05VE
Dinoflagellate cyst transfer function
Dinoflagellate cyst transfer function
Terrestrial pollen transfer function
Terrestrial pollen transfer function
d 18O-calibrated borehole paleothermometry
Marine diatom transfer function
Marine diatom transfer function
Foraminifera transfer function
Foraminifera transfer function
Terrestrial pollen transfer function
Alkenone paleothermometry
Tree-lined and equilibrium line variations
Terrestrial pollen transfer function
Levac et al., 2001
Levac, 2001
Sawada et al., 1999
Sawada et al., 1999
Cuffey and Clow, 1997
Koç et al., 1993
Koç et al., 1993
Duplessy et al., 1992
Duplessy et al., 1992
Penalba et al., 1997
Calvo et al., 2002
Dahl and Nesje, 1996
Seppä and Birks, 2001
Additional records considered
Moon Lake, South Dakota
Elk Lake, Minnesota
Devon Ice Cap
Agassiz Ice Cap
Amakuttalik Lake, Baffin Island
Robinson Lake, Baffin Island
Qipisarqo Lake, southern Greenland
Core 23258, western Barents Sea
46-51VN
45-52VN
75-20VN
80-42VN
67-48VN
63-24VN
61-01VN
75-N
98-09VW
95-48VW
82-30VW
73-06VW
64-34VW
64-16VW
47-45VW
14-E
Diatom-inferred lake salinity
Pollen Assemblages
d 18O of glacial ice
Percent summer melt
Pollen assemblages
Pollen assemblages
Chironomid d 18O-based paleotemperatures
Foraminifera transfer function
Laird et al., 1996
Whitlock et al., 1993
Patterson et al., 1977
Koerner and Fisher, 1990
B. Fréchette, A.P. Wolfe (unpublished)
Miller et al., 1999
Wooller et al., 2004
Sarnthein et al., 2003
226
M.R. Kaplan, A.P. Wolfe / Quaternary Research 65 (2006) 223 – 231
(positive departures) shows some lag relative to the interval of
peak high-latitude summer (June/July) insolation, 11,000 –
9,000 yr ago (Fig. 1d). The earliest expressions of the HTM,
with initial warming before 8000 cal yr B.P., are especially
evident in records from the Scandes Mountains, the Vøring
Plateau (Norwegian Sea), the East Greenland and Iceland seas,
west of the Iberian Peninsula, and adjacent to Nova Scotia.
Remaining sites, notably those around the northwestern North
Atlantic Ocean, warmed considerably later, after 8000 cal yr
B.P. Despite these generalities, it is difficult to discern visually
any simple or geographically coherent patterns from the
standardized paleotemperature records shown in Figure 2.
The three leading PCs extracted from the 13 time-series
reveal distinct modes of variability that are considerably more
interpretable (Fig. 3a). The first principal component (PC1)
explains 33.9% of the variance and depicts a pattern of
progressively increasing scores to a maximum at 7000 cal yr
B.P., followed by a steady decline. High scores reflect positive
loadings from individual sites on both the west and east sides
of the North Atlantic Ocean (Fig. 3b), each of which have
positive temperature anomalies in the 8000– 6000 cal yr B.P.
interval (e.g., Lake Hope-Simpson and Vøring Plateau). PC1
thus characterizes a mode that includes significant delay in the
timing of the HTM relative to peak summer insolation.
In contrast, PC2 (25.0%) produces peak scores prior to
¨7500 cal yr B.P., with strong positive loadings from sites that
warmed earliest in the Holocene and in phase with summer
insolation, such as the Scandes Mountains, the east Greenland
Sea, and the summit of Greenland. The third and final PC
considered here captures considerably less variance (15.4%)
and, as shown in Figure 3b, illustrates pronounced variability
over the mid-Holocene interval, 7000 to 4000 cal yr B.P. PC3
has strongest loadings (both negative and positive) from sites
that are distal to the disintegrating LIS, such as the Iberian
margin, northern Scandinavia, and SE Labrador. All subsequent PCs account for <10% of variance and are not considered
further.
Discussion
Validation of the PCA results
In this section, we evaluate whether or not the three leading
PCs extracted from the paleotemperature time-series are in fact
representative summaries of variability within Holocene
climate trends. We accomplish this through a posteriori
comparisons with additional records from the circumNorth
Atlantic region that were not included in the PCA because of
failure to meet initial selection criteria immediately (Table 1,
Fig. 1). This exercise reveals a number of records that mirror
the patterns of Holocene climate development suggested by
each PC (Fig. 4).
PC1: HTM 8000 – 6000 cal yr B.P.
The temporal evolution of PC1 is broadly mirrored by
several additional proxy records from the eastern Canadian
Figure 2. Standardized deviations of Holocene paleotemperatures at 250 yeartime steps for the 13 records used in the PCA. Records are arranged from west
(top) to east (bottom). Solid horizontal lines are means and dashed lines
indicate 1 standard deviation.
Arctic (Fig. 4a), close to the northeastern margin of the LIS.
For example, both the d 18O record from the Devon Ice Cap
core (Patterson et al., 1977) and pollen assemblages from
eastern Baffin Island lake sediments (Miller et al., 1999) depict
a delayed onset of the HTM relative to maximum summer
M.R. Kaplan, A.P. Wolfe / Quaternary Research 65 (2006) 223 – 231
227
Figure 3. PCA results. (a) temporal evolution of the first three PCs, and (b) mapped loadings of sites to each PC.
insolation, in the order of 1000 –3000 years. The Devon Ice
Cap preserves evidence for an exceptionally long HTM (8000 –
5000 yr B.P.) followed by progressive Neoglacial cooling.
Arctic pollen records are sensitive to local factors that may
delay plant colonization on recently deglaciated landscapes,
thus imparting apparent lags in the response to Holocene
warming that are ecologically and not climatically driven
(Kaufman et al., 2004). However, on eastern Baffin Island, the
most salient Holocene pollen signal is the influx of extraregional shrub pollen (alder and birch) associated with
southerly atmospheric trajectories, a process that is largely
independent of local site characteristics (Miller et al., 1999).
These exotic assemblages replace grass and herb tundra
assemblages around 7000 cal yr B.P. before declining in the
late Holocene (Fig. 4a). PC1 also captures the interval of
progressive Neoglacial cooling following the HTM, which
appears to be a pervasive feature across the North Atlantic
region (Sawada et al., 1999; Jennings et al., 2002). Indeed,
most of the paleotemperature records considered here preserve
evidence of mid to late Holocene cooling (Fig. 2), and thus it is
not surprising that this trend is captured by the dominant PC of
our analysis.
PC2: HTM 10,000 – 8000 cal yr B.P.
The pattern of delayed HTM onset relative to peak
summer insolation (e.g., PC1) is not universally expressed in
the Canadian Arctic, however. Sites from Ellesmere Island
(Agassiz Ice Cap summer melt; Koerner and Fisher, 1990)
and the inter-island channels of the central archipelago (fossil
bowhead whale abundance; Dyke et al., 1996) follow more
closely the mode of PC2 (Fig. 4b). Similarly, early
expressions of the HTM are also observed in the Barents
Sea region (Sarnthein et al., 2003), including Svalbard and
Franz Josef Land (Birks, 1991; Lubinski et al., 1999). It
therefore appears that a high arctic population of sites within
the North Atlantic sector consistently enters the HTM well
before 8000 cal yr B.P., conferring enhanced sensitivity
228
M.R. Kaplan, A.P. Wolfe / Quaternary Research 65 (2006) 223 – 231
Figure 4. Additional proxy records (locations in Fig. 1) that parallel the temporal evolution of the three PCs extracted from the standardized paleotemperatures (a – c),
which are reproduced on the right. These records have been plotted on their independent chronologies following primary data sources, listed in Table 1.
among these sites to the early Holocene summer insolation
maximum. Although the exact mechanism for this geographic pattern remains unclear, perhaps these northern-most sites
(Fig. 1) are, in general, either upstream or distal to
atmospheric and oceanic circulation patterns directly influenced by the LIS (see below), which may account for their
greater sensitivity to insolation forcing. Much of the
Norwegian and Greenland seas, as well as northwestern
Europe, also attained peak warmth well before 8000 cal yr
B.P. Sea ice was much reduced over this region during the
early Holocene, most likely as the combined response to
high insolation forcing and advection of warm surface waters
associated with the North Atlantic Drift (Sejrup et al., 2001;
Birks and KoÓ, 2002; Andrews and Giraudeau, 2003). Off
Nova Scotia, the LeHave Basin also records an early HTM,
a finding that has been recently supported by a record from
the nearby Laurentian Fan in the southern Labrador Sea
(Sachs and Keigwin, 2004).
M.R. Kaplan, A.P. Wolfe / Quaternary Research 65 (2006) 223 – 231
Despite the apparent existence of broad geographic patterns,
there is also considerable evidence for very strong regional
climate gradients in the early Holocene. For example, Qipisarqo
Lake on southwestern Greenland registers an early expression of
the HTM inferred from insect chitin d 18O (Fig. 4b), yet lies only
¨500 km from sites on the opposite side of the Labrador Sea,
which have lagged manifestations of the HTM. Despite their
proximity, east and west coasts of the Labrador Sea/southern
Baffin Bay appear contrasted owing to fundamentally different
oceanographic influences (Fig. 1). While eastern Baffin Island is
flanked by cold water of the Baffin Current, the relatively warm
West Greenland Current, an arm of the North Atlantic Drift,
bathes southwestern Greenland. These differences appear to
have created exceptionally strong regional early Holocene
climate gradients in this region, namely when the Baffin Current
was directly influenced by LIS meltwaters. In addition, major
north –south differences in timing of the HTM apparently
occurred from southeast Labrador to the Scotian shelf area
(LeHave Basin, Levac, 2001; Sachs and Keigwin, 2004). It is
also noteworthy that most proxy records appear to record greater
degrees of spatial and temporal uniformity across the North
Atlantic region once the LIS is no longer a dominant force in
regional climate dynamics (CAPE, 2001; Kaufman et al., 2004).
PC3: unstable mid-Holocene 8000 – 4000 cal yr B.P.
Although considerably less variance is captured by PC3
(15.4%), it nonetheless appears to reflect real, but secondary,
aspects of Holocene climate development expressed in the North
Atlantic region. PC3 highlights pronounced mid-Holocene
climate variability in the 7000 to 5000 cal yr B.P. interval, a
period of considerable change in many regions at both high and
low latitudes (Steig, 1999; Mayewski et al., 2004). Additional
proxy records that preserve parallel shifts to PC3 include
paleoecological records from the North American midwest
(Fig. 4c). In these cases, aridification of the mid-continent was
initiated following final drainage of proglacial lakes, leading to
the replacement of forests by grasslands, and concomitant
increases in the salinity of closed-basin lakes, during the middle
Holocene (Whitlock et al., 1993; Laird et al., 1996).
In summary, our analysis broadens the conclusion of marked
variability with respect to the timing of the HTM (Kaufman et
al., 2004), in addition to reinforcing the notion of spatial and
temporal complexity of Holocene climate change in general
(Mayewski et al., 2004). PCA efficiently captures two dominant
modes that relate closely to final disintegration of the LIS and
summer insolation, respectively (Fig. 1), and thus these are
considered the dominant forcing mechanisms underlying multicentury to millennial-scale climate changes. The dominant PC
indicates that a 1000 – 3000 year lag of the HTM relative to peak
summer insolation is truly a widespread feature of Holocene
climate in the North Atlantic region.
Decay of the LIS and early Holocene climate forcing
The most plausible explanation for delayed expressions of
the HTM in many regions of the North Atlantic sector is the
229
sustained influence of the decaying LIS, both proximally and
distally to its last vestiges in the eastern Canadian Arctic.
Earlier disintegration of the Fennoscandian Ice Sheet implies
that it was unlikely involved as a forcing in the North Atlantic
sector after 10,000 cal yr B.P., given the deglacial chronology
on land and the timing of the final drainage of the Baltic Ice
Lake (Nesje et al., 2004). On the other hand, the LIS still
covered an area of Foxe Basin and Labrador – Québec of about
1.5 106 km2 at 8000 cal yr B.P., representing approximately
10% of its surface area at the Last Glacial Maximum (Fig. 1b;
Dyke et al., 2003). Final demise of the LIS occurred by 6000
cal yr B.P.
Even in these reduced configurations (Fig. 1), the LIS
impacted climate via several processes. First, even a vestigial
ice sheet may have influenced the atmosphere by enhancing cold
northwesterly flow towards the Labrador Sea and displacing
southward the jet stream, as shown by various climate
simulations for 9000 cal yr B.P. (COHMAP, 1998; CAPE,
2001). Second, positive feedbacks associated with ice-surface
albedo enabled the persistence of sea-ice on the adjacent ocean,
primarily downstream of Baffin Bay and Hudson Strait
(Andrews et al., 1999). Third, demise of the LIS remained a
major source of meltwater, of sufficient magnitude to disturb
North Atlantic Deepwater (NADW) formation (Licciardi et al.,
1999; Clark et al., 2001), well into the Holocene.
Early Holocene background meltwater fluxes associated
with LIS melting have been estimated in the order of
0.0114 – 0.834 Sv (Licciardi et al., 1999). Although these
values are more than an order of magnitude lower than those
inferred for abrupt deglacial climatic events, such as
proglacial lake drainage 8200 cal years B.P. (5– 6 Sv;
Barber et al., 1999), they are of comparable magnitude to
the modern freshwater flux to the North Atlantic Ocean that
indeed appears sufficient to influence contemporary NADW
formation (Rahmstorf, 1995; Belkin et al., 1998; Clark et al.,
2002). At present, freshwater is routed from the Arctic
Ocean through Fram Strait and the Canadian Arctic
Archipelago, where it affects convective activity and hence
the climate of the North Atlantic region (Dickson et al.,
1996; Belkin et al., 1998). The decaying LIS would have
provided a comparable sustained meltwater fluxes until
¨6000 cal yr B.P., primarily influencing fiords and marine
embayments throughout eastern Arctic Canada, and ultimately Baffin Bay and the Labrador Sea.
Summary and conclusions
Our analysis suggests that considerable spatial variability in
proxy records of Holocene climate can be expected across the
North Atlantic region. Although there is some a priori basis for
predicting that certain regions will express early Holocene
warming alternately in phase or lagged with respect to summer
insolation, the reality is that both early and late expressions of
the HTM are observed on both sides of the North Atlantic
Ocean. Part of this complexity arises from the profound and
prolonged effects of LIS disintegration, which influenced
climate development via atmospheric impacts, sustained
230
M.R. Kaplan, A.P. Wolfe / Quaternary Research 65 (2006) 223 – 231
meltwater discharge, and attendant impacts on NADW. These
effects persisted as late as 6000 cal yr B.P. based on the current
LIS deglacial chronology (Dyke et al., 2003). In many regions
of the North Atlantic sector, cooling associated with the
remnant LIS offset the influence of high summer insolation
locally, producing the pattern of Holocene climate development
illustrated, for example, by PC1. During the early Holocene,
existence of the residual LIS would have caused strong climatic
gradients between middle and high latitudes along the western
North Atlantic sector, and W – E towards the Greenland
Norwegian seas. The spatial and temporal variability of
Holocene climate development summarized here support the
general notion that North Atlantic heat transport is highly
sensitive to freshwater inputs from ice-sheet collapse, and that
the interplay of this forcing with insolation resulted in
geographically variable climate responses.
Acknowledgments
We thank John Andrews for general advice, especially
concerning the relevance of the 20th century towards
understanding past climates. Eric DeWeaver, Colin Prentice,
Darrell Kaufman, and journal reviewers provided thoughtful
comments. Supported by a Royal Society of London Postdoctoral Fellowship (MRK), the Natural Sciences and Engineering
Research Council of Canada (APW), and the University Centre
in Svalbard.
References
Andrews, J.T., Giraudeau, J., 2003. Multi-proxy records showing significant
Holocene environmental variability: the inner N. Iceland shelf (Hunaflói).
Quaternary Science Reviews 22, 175 – 193.
Andrews, J.T., Keigwin, L.D., Hall, F.R., Jennings, A.E., 1999. Abrupt
deglaciation events and Holocene palaeoceanography from high-resolution
cores, Cartwright Saddle, Labrador Shelf, Canada. Journal of Quaternary
Science 14, 383 – 397.
Barber, D.C., Dyke, A.S., Hillaire-Marcel, C., Jennings, A.E., Andrews, J.T.,
Kerwin, M.W., Bilodeau, G., McNeely, R., Southon, J., Morehead, M.D.,
Gagnon, J.-M., 1999. Forcing of the cold event of 8,200 years ago by
catastrophic drainage of Laurentide lakes. Nature 400, 344 – 348.
Belkin, I.M., Levitus, S., Antonov, J.I., Malmberg, S.-A., 1998. FGreat Salinity
Anomalies_ in the North Atlantic. Progress in Oceanography 41, 1 – 68.
Berger, A., Loutre, M.F., 1991. Insolation values for the climate of the last 10
million years. Quaternary Science Reviews 10, 297 – 318.
Birks, H.H., 1991. Holocene vegetational history and climatic change in west
Spitsbergen-plant macrofossils from Skardtjørna, an arctic lake. The
Holocene 1, 209 – 218.
Birks, C.J.A., Koç, N., 2002. A high-resolution diatom record of lateQuaternary sea-surface temperatures and oceanographic conditions from
the eastern Norwegian Sea. Boreas 31, 323 – 344.
Bond, G., Heinrich, H., Broecker, W., Labeyrie, L., McManus, J., Andrews, J.,
Huon, S., Jantschik, R., Clasen, S., Simet, C., Tedesco, K., Klas, M.,
Bonani, G., Ivy, S., 1992. Evidence for massive discharges of icebergs into
the North Atlantic Ocean during the last glacial. Nature 360, 245 – 249.
Bond, G., Showers, W., Cheseby, M., Lotti, R., Almasi, P., deMenocal, P.,
Priore, P., Cullen, H., Hadjas, I., Bonani, G., 1997. A pervasive millennialscale cycle in North Atlantic Holocene and glacial climates. Science 278,
1257 – 1266.
Calvo, E., Grimalt, J., Jansen, E., 2002. High resolution UK37 sea surface
temperature reconstruction in the Norwegian Sea during the Holocene.
Quaternary Science Reviews 21, 1385 – 1394.
CAPE, 2001. Holocene paleoclimate data from the Arctic: testing models of
global climate change. Quaternary Science Reviews 20, 1275 – 1287.
Clark, P.U., Marshall, S.J., Clarke, G.K.C., Hostetler, S.W., Licciardi, J.M.,
Teller, J.T., 2001. Freshwater forcing of abrupt climate change during the
last glaciation. Science 293, 283 – 287.
Clark, P.U., Pisias, N.G., Stocker, T.S., Weaver, A.J., 2002. The role of the
Atlantic thermohaline circulation in abrupt climate change. Nature 415,
863 – 869.
COHMAP, 1998. Climatic changes of the last 18,000 years: observations and
model simulations. Science 241, 1043 – 1052.
Cuffey, K.M., Clow, G.D., 1997. Temperature, accumulation and ice sheet
elevation in central Greenland through the last deglacial transition. Journal
of Geophysical Research 102, 26383 – 26396.
Dahl, S.O., Nesje, A., 1996. A new approach to calculating Holocene winter
precipitation by combining glacier equilibrium-line altitudes and pine-tree
limits: a case study from Hardangerjøkulen, central southern Norway. The
Holocene 6, 381 – 398.
Dickson, R.R., Lazier, J., Meincke, J., Rhines, P., Swift, J., 1996. Long-term
coordinated changes in the convective activity of the North Atlantic.
Progress in Physical Oceanography 38, 241 – 295.
Duplessy, J.C., Labeyrie, L., Arnold, M., Paterne, M., Duprat, J., van Weering,
T.C.E., 1992. Changes in surface salinity of the North Atlantic Ocean
during the last deglaciation. Nature 358, 485 – 487.
Dyke, A.S., Hooper, J., Savelle, J.M., 1996. A history of sea ice in the
Canadian Arctic Archipelago based on postglacial remains of the Bowhead
Whale (Balaena mystecetus). Arctic 49, 235 – 255.
Dyke, A.S., Moore, A., Roberson, L., 2003. Deglaciation of North America.
Geological Survey of Canada Open File Report 1574 (CD ROM).
Jennings, A.E., Knudsen, K.L., Hald, M., Hansen, C.V., Andrews, J.T., 2002. A
mid-Holocene shift in Arctic sea-ice variability on the East Greenland
Shelf. The Holocene 12, 49 – 58.
Kaufman, D.S., Ager, T.A., Anderson, N.J., Anderson, P.M., Andrews, J.T.,
Bartlein, P.J., Brubaker, L.B., Coats, L.L., Cwynar, L.C., Duvall, M.L.,
Dyke, A.S., Edwards, M.E., Eisner, W.R., Gajewski, K., Geirsdottir, A.,
Hu, F.S., Jennings, A.E., Kaplan, M.R., Kerwin, M.N., Lozhkin, A.V.,
MacDonald, G.M., Miller, G.H., Mock, C.J., Oswald, W.W., Otto-Bliesner,
B.L., Porinchu, D.F., Rühland, K., Smol, J.P., Steig, E.J., Wolfe, B.B.,
2004. Holocene thermal maximum in the western Arctic (0 to 180-W).
Quaternary Science Reviews 23, 529 – 560.
Koç, N., Jansen, E., Haflidason, H., 1993. Paleoceanographic reconstructions
of surface ocean conditions in the Greenland, Iceland, and Norwegian Seas
through the last 14 ka based on diatoms. Quaternary Science Reviews 12,
115 – 140.
Koerner, R.M., Fisher, D.A., 1990. A record of Holocene summer climate from
a Canadian high-Arctic ice core. Nature 343, 630 – 631.
Kutzbach, J.E., 1981. Monsoon climate of the early Holocene: climate
experiment with the Earth’s orbital parameters for 9000 years ago. Science
214, 59 – 61.
Laird, K.R., Fritz, S.C., Grimm, E.C., Mueller, P.G., 1996. Century-scale
paleoclimatic reconstructions from Moon Lake, a closed-basin lake in the
northern Great Plains. Limnology and Oceanography 41, 890 – 902.
Levac, E., 2001. High resolution Holocene palynological record from the
Scotian shelf. Marine Micropaleontology 43, 179 – 197.
Levac, E., de Vernal, A., Blake Jr., W., 2001. Sea-surface conditions in
northernmost Baffin Bay during the Holocene: palynological evidence.
Journal of Quaternary Science 16, 353 – 363.
Licciardi, J.M., Teller, J.T., Clark, P.U., 1999. Freshwater routing by the
Laurentide Ice Sheet during the last deglaciation. In: Clark, P.U., Webb,
R.S., Keigwin, L. (Eds.), Mechanisms of Global Climate Change at
Millennial Time Scales, Geophysical Monograph Series 112. American
Geophysical Union, Washington, DC, pp. 177 – 201.
Lubinski, D.J., Forman, S.L., Miller, G.H., 1999. Holocene glacier and climate
fluctuations on Franz Josef Land, Arctic Russia, 80-N. Quaternary Science
Reviews 18, 85 – 108.
Mayewski, P.A., Rohling, E.J., Stager, J.C., Karlén, W., Maasch, K., Meeker,
L.D., Meyerson, E., Gasse, F., Van Kreveld, S., Holmgren, K., Lee-Thorp,
J., Rosqvist, G., Rack, F., Staubwasser, M., Schneider, R.R., Steig, E.J.,
2004. Holocene climate variability. Quaternary Research 62, 243 – 255.
M.R. Kaplan, A.P. Wolfe / Quaternary Research 65 (2006) 223 – 231
Miller, G.H., Mode, W.N., Wolfe, A.P., Sauer, P.E., Bennike, O., Forman,
S.L., Short, S.K., Stafford Jr., T.W., 1999. Stratified interglacial
lacustrine sediments from Baffin Island, Arctic Canada: chronology
and paleoenvironmental implications. Quaternary Science Reviews 18,
789 – 810.
Nesje, A., Dahl, S.O., Bakke, J., 2004. Were abrupt Late glacial and earlyHolocene climatic changes in northwest Europe related to freshwater
outbursts to the North Atlantic and Arctic oceans? The Holocene 14,
299 – 310.
Overpeck, J.T., Hughen, K.A., Hardy, D., Bradley, R.S., Case, R., Douglas,
M.S.V., Finney, B., Gajewski, K., Jacoby, G., Jennings, A.E., Lamoureux,
S., Lasca, A., MacDonald, G.M., Moore, J.J., Retelle, M., Wolfe, A.P.,
1997. Arctic environmental change of the last four centuries. Science 278,
1251 – 1256.
Patterson, W.S.B., Koerner, R.M., Fisher, D.A., Johnsen, S.J., Clausen, H.B.,
Dansgaard, W., Butcher, P., Oeschger, M., 1977. An oxygen-isotope
climatic record from the Devon Island Ice Cap, Arctic Canada. Nature
266, 508 – 511.
Penalba, M.C., Arnold, M., Guiot, J., Duplessey, J.-C., de Beaulieu, J.-L.,
1997. Termination of the last glaciation in the Iberian Peninsula inferred
from the pollen sequence of Quintanar de la Sierra. Quaternary Research
48, 205 – 214.
Rahmstorf, S., 1995. Bifurcations of the Atlantic thermohaline circulation in
response to changes in the hydrological cycle. Nature 378, 145 – 149.
Rahmstorf, S., 1997. Risk of sea-change in the Atlantic. Nature 388,
825 – 826.
Sachs, J.P., Keigwin, L.D., 2004. Extreme Cooling of Slopewaters East of the
USA and Canada During the Holocene. EOS Transactions AGU, 85 (47),
Fall Meeting Supplement, Abstract PP13A – 0591.
231
Sarnthein, M., van Kreveld, S., Erlenkeuser, H., Grootes, P.M., Kucera, M.,
Pflaumann, U., Schulz, M., 2003. Centennial to millennial-scale periodicities of Holocene climate and sediment injections off the western Barents
shelf, 75-N. Boreas 32, 447 – 461.
Sawada, M., Gajewski, K., de Vernal, A., Richard, P.J.H., 1999. Comparison of
marine and terrestrial Holocene climatic reconstructions from northeastern
North America. The Holocene 9, 267 – 277.
Sejrup, H.P., Haflidason, H., Flatebø, T., Kristensen, D., Grøsfjeld, K., Larsen,
E., 2001. Late-glacial to Holocene environmental changes and climate
variability: evidence from Voldafjorden, western Norway. Journal of
Quaternary Science 16, 181 – 198.
Seppä, H., Birks, H.J.B., 2001. July mean temperature and annual precipitation
trends during the Holocene in the Fennoscandian tree-line area: pollenbased climate reconstructions. The Holocene 11, 527 – 539.
Steig, E.J., 1999. Mid-Holocene climate change. Science 286, 1485 – 1487.
ter Braak, C.J.F., Prentice, I.C., 1988. A theory of gradient analysis. Advances
in Ecological Research 18, 271 – 317.
ter Braak, C.J.F., Šmilauer, P., 1998. CANOCO Reference Manual and User’s
Guide to Canoco for Windows: Software for Canonical Community
Ordination (version 4) Ithaca. Microcomputer Power, New York.
Whitlock, C., Bartlein, P.J., Watts, W.A., 1993. Vegetation history of
Elk Lake. In: Bradbury, J.P., Dean, W.E. (Eds.), Elk Lake, Minnesota:
Evidence for Rapid Change in The North-Central United States, Special
Paper - Geological Society of America, vol. 276. Geological Society of
America, Boulder, Colorado, pp. 251 – 274.
Wooller, M.J., Francis, D., Fogel, M.L., Miller, G.H., Walker, I.R., Wolfe, A.P.,
2004. Quantitative paleotemperature estimates from d 18O of chironomid
head capsules preserved in arctic lake sediments. Journal of Paleolimnology
31, 267 – 274.