Accelerated drawdown of meridional overturning in the late-glacial
Atlantic triggered by transient pre-H event freshwater perturbation
I. R. Hall,1S. B. Moran,2 R. Zahn,1,3 P. C. Knutz,4 C.-C. Shen,5 and R. L. Edwards6 Received 8 March 2006; revised 30 June 2006; accepted 7 July 2006; published 31 August 2006.
[1] Abrupt decreases of the Atlantic meridional
overturning circulation (MOC) during the Late Pleistocene have been directly linked to catastrophic discharges of glacimarine freshwater, triggering disruption of northward marine heat transport and causing global climate changes. Here we provide measurements of excess sedimentary
231
Pa/230Th from a high-accumulation sediment drift deposit in the NE Atlantic that record a sequence of sudden variations in the rate of MOC, associated deep ocean ventilation and surface-ocean climatology. The data series reveal a sequential decrease in the MOC rate at 18.0 ka BP ago that coincides with only transient and localized freshwater inputs. This change represents a substantial, though not total, cessation in MOC that predates the major Heinrich (H1) meltwater event by at least 1,200 years. These results highlight the potential of targeted freshwater perturbations in promoting substantial MOC changes without a direct linking with catastrophic freshwater surges. Citation: Hall, I. R., S. B. Moran, R. Zahn, P. C. Knutz, C.-C. Shen, and R. L. Edwards (2006), Accelerated drawdown of meridional overturning in the late-glacial Atlantic triggered by transient pre-H event freshwater perturbation, Geophys. Res. Lett., 33, L16616, doi:10.1029/2006GL026239.
1. Introduction
[2] 231Pa (half life t1/2= 32.5 ka) and230Th (t1/2= 75.2 ka)
are particle-reactive radionuclides produced in the oceans at a constant initial 231Pa/230Th activity ratio of 0.093 and deposited in underlying sediments by attachment to settling particles. Interest in the application of excess, or unsup-ported, sediment 231Pa/230Th (decay-corrected to the time of deposition; herein referred to as 231Paxs/230Thxs) as a
palaeocirculation proxy in the Atlantic is based on the similar residence time of 231Pa (100– 200 yrs) [Nozaki and Nakanishi, 1985] and Atlantic deep waters [Broecker, 1979], whereby the rate of meridional overturning circula-tion (MOC) directly affects the lateral export of231Pa [Yu et al., 1996; Marchal et al., 2000]. Presently, approximately
half of the water column production of231Pa is exported in North Atlantic Deep Water (NADW) to the Southern Ocean [Yu et al., 1996; Marchal et al., 2000] by comparison, lateral transport of 230Th is minimized due to its shorter residence time (20– 40 yr) [Nozaki and Nakanishi, 1985; Yu et al., 1996; Marchal et al., 2000]. Thus, changes in Atlantic MOC are recorded down-core as variations in sediment 231Paxs/230Thxs, with vigorous rates of MOC
causing a reduction in231Paxs/230Thxsfrom the production
ratio. The 231Paxs/230Thxs proxy therefore enables the
re-construction of deep ocean circulation changes that goes beyond what traditional proxies such asd13C from benthic foraminifera can provide. Benthic d13C is driven by end-member variation, water mass chemical ‘‘aging’’ and mix-ing while 231Paxs/230Thxs does depend more directly on
lateral water mass advection, i.e., the physical vigor of the MOC.
2. Material and Methods
[3] We measured sedimentary231Paxs/ 230
Thxsalong
sed-iment core DAPC2 recovered from a high-accumulation drift deposit located in the northern Rockall Trough (5858.100N, 0936.750W, 1709 m water depth) (Figure 1). The site is presently influenced by Wyville-Thomson Overflow Water (WTOW) from the Norwegian Sea, a precursor to North Atlantic Deep Water (NADW), and recirculated upper NADW [New and Smythe-Wright, 2001]. This core provides high-resolution, multi-proxy, paleoceanographic records [Knutz et al., 2002] of iceberg discharge and meltwater variability that indicate a rapid response of ice sheet surges and meltwater from NW Europe over the last deglaciation.
[4] Radiochemical analyses of 231Pa and 230Th were
made by isotope dilution using a thermal ionization mass spectrometer for 231Pa and a high-resolution magnetic sector inductively coupled plasma mass spectrometer for
230
Th [Shen et al., 2002, 2003]. Diatom abundance as a possible modulator of 231Paxs/230Thxs was quantified
fol-lowing Scherer [1995]. The chronology of DAPC2 is con-strained by 10 calibrated accelerator mass spectrometry14C dates (age scale of Knutz et al. [2002] augmented by 4 further 14C-AMS dates) that provide a tight temporal framework for the abrupt MOC changes indicated in the records and their linking with sudden changes of surface ocean climatology. The resulting time step along our stable isotope, faunal and ice rafted debris (IRD) records during the last glacial maximum (LGM) and H1 is less than 50 years, mean time step along the 231Paxs/230Thxsis less
than 120 years. At this temporal resolution, and because of its geographic location, DAPC2 provides an ideal sedimen-tary record in which to investigate the sequence of events
Here for
Full Article
1School of Earth, Ocean, and Planetary Sciences, Cardiff University, Cardiff, UK.
2Graduate School of Oceanography, University of Rhode Island, Narragansett, Rhode Island, USA.
3
Now at Institut de Ciencia i Tecnologia Ambientals, Universitat Auto`noma de Barcelona, Bellaterra, Spain.
4
Geological Survey of Denmark and Greenland, Copenhagen, Denmark.
5
Department of Geosciences, National Taiwan University, Taipei, Taiwan.
6
Department of Geology and Geophysics, University of Minnesota, Minneapolis, Minnesota, USA.
Copyright 2006 by the American Geophysical Union. 0094-8276/06/2006GL026239$05.00
surrounding a major past freshwater perturbation, the H1 meltwater event, and its consequences for the MOC.
3. Results and Discussion
[5] The231Paxs/230Thxsrecord displays (Figures 1 and 2)
minimum values during the LGM and a multi-step transition to higher values that includes the Heinrich 1 (H1) meltwater event (H-1MW). As non-carbonate clay and carbonate fluxes
were high during the LGM (see auxiliary material1Figure S1), conditions typically associated with elevated
231
Paxs/230Thxsratios [Bacon, 1988], the low231Paxs/230Thxs
ratios prior to18 ka BP imply a vigorous MOC rate and export of231Pa from this location. At this timed18O values in DAPC2 of surface (G. bulloides [Schiebel et al., 1997]) and deep-dwelling (thermocline depths, N. pachyderma sinistral [Kohfeld et al., 1996]) planktonic foraminifera converge suggesting a weakly developed thermocline and decreased vertical stability, indicative of a well mixed ocean upper layer, plausibly marking conditions favorable for convective overturn in the region, in line with elevated benthic d13C in our core [Knutz et al., 2002]. In contrast, following the Younger Dryas stadial and throughout the Holocene, 231Paxs/230Thxs ratios remain close to the
pro-duction ratio which we attribute to preferential removal of
231
Pa by biogenic silica [Guo et al., 2002; Luo and Ku, 2004; Siddall et al., 2005], as evidenced by enhanced
diatom accumulation in the uppermost 50 cm (12 ka BP) of DAPC2 (Figure 1c). The absence of diatom accu-mulation at the DAPC2 site during the LGM and early deglaciation indicates that231Paxs/230Thxsis not dominated
by variations in biogenic silica flux during this period. [6] Fine scale variability is observed in DAPC2 on
cen-tennial to sub-cencen-tennial time scales (Figure 2). These changes are directly coupled to atmospheric and surface
Figure 1. Study area and core DAPC2 data. (a) Map of the North Atlantic showing the location of core DAPC2 from Rockall Trough (5858.100N, 0936.750W, 1709 m water depth). (b)230Th normalized biogenic silica flux (diatoms), note axis break (c) sedimentary record of 231Paxs/230Thxs.
Horizontal dashed line represents the 231Pa/230Th produc-tion ratio. AMS14C ages denoted by solid black triangles. Error bars represent uncertainties in 231Pa and230Th data (2s), which range from 3 – 7% and are generally <5%. Shaded area highlights data not considered here for the reconstruction of paleocirculation due to the likely effects of preferential removal of 231Pa by the increased biogenic silica flux.
Figure 2. Paleoceanographic time-series from DAPC2 (a) Neogloboquadrina pachyderma sinistral (Nps) abundance. (b) Abundance of ice-rafted debris, quartz grains (Quartz, black) and detrital carbonate (DC, brown), in the >250mm fraction (c) N. pachyderma s. (Nps., cyan), Globigerina bulloides d18O (G. bull., blue) and benthic d18O (Cibici-doides wuellerstorfi, Cw, orange). (d) Benthic d13C (Cw). (e) Sediment 231Paxs/230Thxs. Acronyms are: H-1mw,
Heinrich 1 meltwater event; E-1a-c, European sourced
glacimarine events [Knutz et al., 2002]. These E-events and additional IRD pulses that do not display measurable planktonic d18O meltwater excursions at this location are highlighted by the grey bands. Heinrich Event 1 (H1) indicated by yellow band. Arrows between d and e mark onset of a multi-step MOC slowdown.
1
Auxiliary materials are available in the HTML. doi:10.1029/ 2006GL026239.
ocean (buoyancy) forcing revealed by the planktonic records that document a sequence of transient warming events fol-lowed by cooling/freshwater pulses, displayed in briefd18O and N. pachyderma (sin) abundance anomalies, prior to and during the onset of deglaciation (Figures 2a and 2c). Con-comitant IRD peaks (Figures 2a – 2c) suggest that the latter are linked with brief but intense pulses of icebergs discharged from the EIS (Figure 2b), previously labeled as E-events [Knutz et al., 2002], most likely triggered by short-lived collapses of marine-based ice margins caused by the slow but steady global warming at the end of the LGM. Similar lithic ‘‘precursor’’ events preceding Heinrich events have been reported from North Atlantic core sites and inferred to originate from European and Icelandic glacial sources [Grousset et al., 2000; Scourse et al., 2000]. The DAPC2
231
Paxs/230Thxsrecord indicates that these events went along
with measurable slowdown in MOC. In particular the event centered at19.1 ka BP stands out that is associated with a cooling and freshwater pulse, labeled as E-1cevent (Figure 2).
The transient MOC slow-down during E-1cthat is evident in 231
Paxs/230Thxs is around half the magnitude of the major
MOC drawdown starting at 18 ka BP, although a value similar to those around H1 is reached in a single data point at 19.0 ka. MOC rapidly recovered from the slow down during the E-1cevent, within200 years returning to similar rates as
before. A sustained step-wise increase in231Paxs/230Thxsthen
starts at18.0 ka BP, within 170 yr after the warm pulse that followed the European sourced glacimarine E-1b event
[Knutz et al., 2002]. This signifies the onset of reduced lateral
231
Pa escape and a slow-down of deep water export from the region (Figures 2a and 2e). The first step of apparent MOC slow-down is coincident with a centennial scale IRD pulse that suggests a coeval EIS surge event. About one third of the total MOC drawdown, relative to the 18.0 ka BP
231
Paxs/230Thxmaxima, is accounted for in this initial step.
Approximately 270 years later, a second step of increasing
231
Paxs/230Thxsdirectly coincides with the incursion of
fresh-water/IRD pulse E-1a. This phase of further MOC slow-down
is accompanied by a coeval stepwise benthicd13C decrease that is also recorded at other core sites in the wider North Atlantic region and has been used, in conjunction with benthic trace element ratios, to infer decreasing chemical ventilation from northern sources and increased advection of southern hemisphere water masses [Zahn et al., 1997; Will-amowski and Zahn, 2000; Zahn and Stu¨ber, 2002; Rickaby and Elderfield, 2005]. The lack of equivalent negative benthic d18O excursions during the E-1b-aevents supports the view
that these events did not cause globally significant changes in ice volume and sea level. The main H1 event lithologically identified in DAPC2 by the presence of a discrete detrital (dolomitic) carbonate IRD layer that is embedded within the quartz IRD deposit [Knutz et al., 2002] and the associated meltwater surge then occur as MOC rates approach peak minimum rates, some 1,200 yr after the onset of the MOC drawdown (Figure 2, H-1MW).231Paxs/230Thxsratios
through-out the H1 interval remain below production ratio indicating that MOC never achieved a total cessation.
[7] The changes in deep water export traced by 231
Paxs/230Thxs in DAPC2 are also recorded at deep
sub-tropical Atlantic sites (core OCE326-GGC5, lower Ber-muda Rise, 4550m water depth [McManus et al., 2004]; SU81-18, western Iberian margin, 3135 m water depth
[Gherardi et al., 2005]). The inference drawn at these sites was a significant slow-down in MOC coincident with the catastrophic iceberg discharge H1 [McManus et al., 2004; Gherardi et al., 2005] dated to 16.8 ka BP in the North Atlantic [Hemming, 2004]. The fine-scale time series of complementary proxies of ice rafted detritus, planktonic d18O, benthic d13C, and 231Paxs/230Thxs (Figure 2) in
DAPC2 from within the more immediate region of iceberg drift reveals that such MOC decrease occurred significantly before the H-1 event. The fact that all our records are from the same single sediment core ensures that the sequence of events and the temporal offsets between them are robust features independent of age modeling. In particular, we emphasize the similarity between the 231Paxs/230Thxs and
benthicd13C signals that underscores the significance of the changes in the rate of overturning inferred from
231
Paxs/230Thxs. Based on our age model the MOC decrease
pr ec e de s t h e H 1 ev e nt b y a s m u ch as 1,200 (231Paxs/230Thxs) to1,300 yrs (benthic d13C).
[8] Transient231Paxs/230Thxsincreases associated with the
E-1a-b meltwater events suggest a direct link between
circulation changes and local European sector derived meltwater perturbations. Taking the E-1a-bplanktonicd
18
O anomalies at face value suggests a salinity reduction of 0.3– 1.0 (freshwater endmember d18O35% VSMOW), probably higher if accounting for the effects of sea surface cooling on planktonic d18O. While salinity anomalies ap-pear close to the magnitude of the H1mw perturbation
recorded at the DAPC2 site, at 90– 150 year duration these localized episodes are extremely short-lived. The possibility of additional sources of freshwater input at the time of MOC collapse remains for instance, from the small Iceland Ice Sheet and the much larger Laurentide and Greenland Ice Sheets [Hagen and Hald, 2002; Voelker et al., 2000] but the European provenance for the E-1a-b
meltwater perturbations is consistent with the reconstruction of a widespread salinity reduction west of Ireland at 15.0 – 17.714C ka BP (17.3 – 20.4 ka BP) that likewise has been linked to European melt sources [Sarnthein et al., 1995]. Moreover, the close linkage of the multi-step231Paxs/230Thxs
increase documented in DAPC2 with the sequence of IRD and negative planktonicd18O events in the same core make a compelling case that while additional freshwater sources may have existed elsewhere in the region it were these brief freshwater pulses that played a decisive role in the sus-tained, but not total, drawdown of meridional overturning. Thus the H1 event and its associated major meltwater surge (H-1MW) appear to have been embedded into an already
slowed-down MOC and acted to sustain rather than initiate minimum MOC rates. Such lead of MOC slow down is supported by a similar temporal offset between the initial
231
Paxs/230Thxsincrease after 19 ka BP in the deep
subtrop-ical Atlantic [McManus et al., 2004] and the initiation of surface ocean heat being retained in the tropical [Ru¨hlemann et al., 1999] and subtropical [Flower et al., 2004] Atlantic that each significantly precede the age of H1 event. Continued surface ocean stratification is sug-gested in DAPC2 in the enhanced offset between the paired planktonicd18O records following H1 (Figure 2c) [Knutz et al., 2002] most likely indicating enhanced poleward trans-port of warm surface waters to the Nordic Sea, via the North Atlantic Drift, and that the location of deep convection has
shifted to north of the Iceland-Scotland Ridge [Rahmstorf, 1995], leaving the DAPC2 site under the influence of WTOW and an increasing recirculation of upper deep water [Knutz et al., 2002].
4. Conclusions
[9] Our data demonstrate that catastrophic freshwater
surges, e.g., during the past from the disintegrating Lauren-tide Ice Sheet, are not the sole component involved with major MOC change. Model simulations suggest that the MOC is sensitive to small-scale freshwater perturbations, on order of 0.1 Sv [Stocker and Wright, 1991; Manabe and Stouffer, 1997]. Our data from core DAPC2 indeed demonstrate that transient localized freshwater episodes occurred between 1,200 – 1,300 years prior to the large-scale H1 event and coincided with likewise transient phases of increasing
231
Paxs/230Thxsand decreasing benthicd13C, both fully
con-sistent with decreasing deep water ventilation and export. The events formed part of a multi-stepped shift toward a collapsed MOC, lending credence to numerical models that demon-strate the role of targeted freshwater events in forcing the MOC toward and across a threshold from where on convec-tion enters a collapsed mode [Rahmstorf, 1994, 1995]. From this it appears that accelerated (yet non-catastrophic scale) melting of the Greenland Ice Sheet [Zwally et al., 2002], may indeed bear significance for future MOC stability and climate in the wider North Atlantic region.
[10] Acknowledgments. We thank H. Medley, G. Bianchi, and J. Pike for technical assistance. The helpful comments of two anonymous reviewers improved this manuscript. This work was supported by funding from the U.K. National Environmental Research Council, U.S. National Science Foundation, Ministerio de Educacio´n y Ciencia, Spain, and the Taiwan ROC National Science Council.
References
Bacon, M. P. (1988), Tracers of chemical scavenging in the ocean: Bound-ary effects and large scale chemical fractionation, Philos. Trans. R. Soc. London, Ser. A, 325, 147 – 160.
Broecker, W. S. (1979), A revised estimate of the radiocarbon age of the North Atlantic Deep Water, J. Geophys. Res., 84, 3218 – 3226. Flower, B. P., D. W. Hastings, H. W. Hill, and T. M. Quinn (2004), Phasing
of deglacial warming and Laurentide Ice Sheet meltwater in the Gulf of Mexico, Geology, 32, 597 – 600.
Gherardi, J. M., L. Labeyrie, J. F. McManus, R. Francois, L. C. Skinner, and E. Cortijo (2005), Evidence from the Northeastern Atlantic basin for variability in the rate of the meridional overturning circulation through the last deglaciation, Earth Planet. Sci. Lett., 240, 710 – 723.
Grousset, F. E., C. Pujol, L. Labeyrie, G. Auffret, and A. Boelaert (2000), Were the North Atlantic Heinrich events triggered by the behavior of the European ice sheets?, Geology, 28, 123 – 126.
Guo, L., M. Chen, and C. Gueguen (2002), Control of Pa/Th ratio by particulate chemical composition in the ocean, Geophys. Res. Lett., 29(20), 1961, doi:10.1029/2002GL015666.
Hagen, S., and M. Hald (2002), Variation in surface and deep water circulation in the Denmark Strait, North Atlantic, during marine iso-tope stages 3 and 2, Paleoceanography, 17(4), 1061, doi:10.1029/ 2001PA000632.
Hemming, S. R. (2004), Heinrich events: Massive late Pleistocene detritus layers of the North Atlantic and their global climate imprint, Rev. Geo-phys., 42, RG1005, doi:10.1029/2003RG000128.
Knutz, P. C., I. R. Hall, R. Zahn, T. L. Rasmussen, A. Kuijpers, M. Moros, and N. J. Shackleton (2002), Multidecadal ocean variability and NW European ice sheet surges during the last deglaciation, Geochem. Geo-phys. Geosyst., 3(12), 1077, doi:10.1029/2002GC000351.
Kohfeld, K. E., R. G. Fairbanks, S. L. Smith, and I. D. Walsh (1996), Neogloboquadrina pachyderma (sinistral coiling) as paleoceanographic tracers in polar oceans: Evidence from Northeast Water Polynya plankton tows, sediment traps, and surface sediments, Paleoceanography, 11(6), 679 – 699.
Luo, S., and T.-L. Ku (2004), On the importance of opal, carbonate, and lithogenic clays in scavenging and fractionating230Th,231Pa, and10Be in the ocean, Earth Planet. Sci. Lett., 220, 201 – 211.
Manabe, S., and R. J. Stouffer (1997), Coupled ocean-atmosphere model response to freshwater input: Comparison to Younger Dryas event, Paleoceanography, 12(2), 321 – 336.
Marchal, O., R. Franc¸ois, T. F. Stocker, and F. Joos (2000), Ocean thermo-haline circulation and sedimentary231Pa/230Th ratio, Paleoceanography, 15(6), 625 – 641.
McManus, J. F., R. Francois, J. M. Gherardi, L. D. Keigwin, and S. Brown-Leger (2004), Collapse and rapid resumption of Atlantic meridional cir-culation linked to deglacial climate changes, Nature, 428, 834 – 837. New, A. L., and D. Smythe-Wright (2001), Aspects of the circulation in the
Rockall Trough, Cont. Shelf. Res., 21, 777 – 810.
Nozaki, Y., and T. Nakanishi (1985),231Pa and230Th in the open ocean water column, Deep Sea Res., Part A, 32, 1209 – 1220.
Rahmstorf, S. (1994), Rapid climate transitions in a coupled ocean-atmosphere model, Nature, 372, 82 – 85.
Rahmstorf, S. (1995), Bifurcation of the Atlantic thermohaline circulation in response to changes in the hydrological cycle, Nature, 378, 145 – 149. Rickaby, R. E. M., and H. Elderfield (2005), Evidence from the high-latitude North Atlantic for variations in Antarctic Intermediate water flow during the last deglaciation, Geochem. Geophys. Geosyst., 6, Q05001, doi:10.1029/2004GC000858.
Ru¨hlemann, C., S. Mulitza, P. J. Mu¨ller, G. Wefer, and R. Zahn (1999), Warming of the tropical Atlantic Ocean and slowdown of thermohaline circulation during the last deglaciation, Nature, 402, 511 – 514. Sarnthein, M., et al. (1995), Variations in Atlantic surface ocean
paleocea-nography, 50 – 80N: A time-slice record of the last 30,000 years, Pa-leoceanography, 10(6), 1063 – 1094.
Scherer, R. P. (1995), A new method for the determination of absolute abundance of diatoms and other silt-sized sedimentary particles, J. Pa-leolimnol., 12, 171 – 179.
Schiebel, R., J. Bijma, and C. Hemleben (1997), Population dynamics of the planktic foraminifer Globigerina bulloides from the eastern North Atlantic, Deep Sea Res., Part I, 44, 1701 – 1713.
Scourse, J. D., I. R. Hall, I. N. McCave, J. R. Young, and C. Sugdon (2000), The origin of Heinrich layers: Evidence from H2 for European precursor events, Earth Planet. Sci. Lett., 182, 187 – 195.
Shen, C.-C., et al. (2002), Uranium and thorium isotopic and concentration measurements by magnetic sector inductively coupled plasma mass spec-trometry, Chem. Geol., 185, 165 – 178.
Shen, C.-C., et al. (2003), Measurement of attogram quantities of231Pa in dissolved and particulate fractions of seawater by isotope dilution thermal mass spectrometry, Anal. Chem., 75, 1075 – 1079.
Siddall, M., G. M. Henderson, N. R. Edwards, S. A. Mu¨ller, T. F. Stocker, F. Joos, and M. Frank (2005),231Pa/230Th fractionation by ocean trans-port, biogenic particle flux and particle type, Earth Planet. Sci. Lett., 237, 137 – 155.
Stainforth, D. A., et al. (2005), Uncertainty in predictions of the climate response to rising levels of greenhouse gases, Nature, 433, 403 – 406. Stocker, T. F., and D. G. Wright (1991), Rapid transitions of the ocean’s
deep circulation induced by changes in surface water fluxes, Nature, 351, 729 – 732.
Voelker, A. H. L., P. M. Grootes, M.-J. Nadeau, and M. Sarnthein (2000), Radiocarbon levels in the Iceland Sea from 25 – 53 kyr and their link to the Earth’s magnetic field intensity, Radiocarbon, 42, 437 – 452. Willamowski, C., and R. Zahn (2000), Upper ocean circulation in the
glacial North Atlantic from benthic foraminiferal isotope and trace ele-ment fingerprinting, Paleoceanography, 15(5), 515 – 527.
Yu, E.-F., R. Francois, and M. P. Bacon (1996), Similar rates of modern and last-glacial ocean thermohaline circulation inferred from radiochemical data, Nature, 379, 689 – 694.
Zahn, R., and A. Stu¨ber (2002), Suborbital intermediate water variability inferred from paired benthic foraminiferal Cd/Ca andd13C in the tropical West Atlantic and linking with North Atlantic climates, Earth Planet. Sci. Lett., 200, 191 – 205.
Zahn, R., J. Scho¨nfeld, H. Kudrass, M. Park, H. Erlenkeuser, and P. Grootes (1997), Thermohaline instability in the North Atlantic during meltwater events: Stable isotope and ice-rafted detritus records from core SO75-26KL, Portuguese margin, Paleoceanography, 12(5), 696 – 710. Zwally, H. J., et al. (2002), Surface melt-induced acceleration of Greenland
ice-sheet flow, Science, 297, 218 – 222.
R. L. Edwards, Department of Geology and Geophysics, University of Minnesota, 310 Pillsbury Drive SE, Minneapolis, MN 55455, USA.
I. R. Hall, School of Earth, Ocean and Planetary Sciences, Cardiff University, Main Building, Park Place, Cardiff CF10 3YE, UK. (hall@cardiff.ac.uk)
P. C. Knutz, Geological Survey of Denmark and Greenland, Øster Voldgade 10, 1350 Copenhagen, Denmark.
S. B. Moran, Graduate School of Oceanography, University of Rhode Island, South Ferry Road, Narragansett, RI 02882-1197, USA.
C.-C. Shen, Department of Geosciences, National Taiwan University, No 1 Sec 4, Roosevelt Road, Taipei 106, Taiwan.
R. Zahn, Institut de Ciencia i Tecnologia Ambientals, Universitat Auto`noma de Barcelona, Edifici Cn - Campus UAB, Bellaterra E-08193, Spain.