Quaternary Science Reviews 29 (2010) 3399e3414 Contents lists available at ScienceDirect Quaternary Science Reviews journal homepage: www.elsevier.com/locate/quascirev Late Quaternary glaciation history of northernmost Greenland e Evidence of shelf-based ice Nicolaj K. Larsen a, b, *, Kurt H. Kjær c, Svend Funder c, Per Möller a, Jaap J.M. van der Meer d, Anders Schomacker e, f, Henriette Linge g, Dennis A. Darby h a Department of Earth and Ecosystem Sciences, Lund University, Sölvegatan 12, S-223 62 Lund, Sweden Department of Earth Sciences, University of Aarhus, C.F. Møllers Allé 4, DK-8000 Aarhus, Denmark Centre for Geogenetics, Natural History Museum of Denmark, Øster Voldgade 5-7, DK-1350 Copenhagen, Denmark d Department of Geography, Queen Mary University of London, Mile End Road, London E1 4NS, UK e University of Iceland, Institute of Earth Sciences, Askja, Sturlugata 7, IS-101 Reykjavík, Iceland f Norwegian University of Science and Technology, Department of Geology, Sem Sælands Veg 1, N-7491 Trondheim, Norway g Department of Earth Science and Bjerknes Centre for Climate Research, University of Bergen, Allégaten 41, NO-5007 Bergen, Norway h Department of Ocean, Earth and Atmospheric Sciences, Old Dominion University, Norfolk, USA b c a r t i c l e i n f o a b s t r a c t Article history: Received 13 December 2009 Received in revised form 16 July 2010 Accepted 30 July 2010 We present the mapping of glacial landforms and sediments from northernmost Greenland bordering 100 km of the Arctic Ocean coast. One of the most important discoveries is that glacial landforms, sediments, including till fabric measurements, striae and stoss-lee boulders suggest eastward ice-flow along the coastal plain. Volcanic erratic boulders document ice-transport from 80 to 100 km west of the study area. We argue that these findings are best explained by local outlet glaciers from the Greenland Ice Sheet and local ice caps that merged to form a shelf-based ice in the Arctic Ocean and possibly confirming an extensive ice shelf in the Lincoln Sea between Greenland and Ellesmere Island. It is speculated that the shelf-based ice was largely affected by the presence of thick multiyear sea ice in the Arctic Ocean that prevented it from breaking up and forced the outlet glaciers to flow eastwards. During the initial retreat the coastal area was dammed by the shelf-based ice and kame and glaciolacustrine sediments were deposited up to 50 m above the marine limit before the final deglaciation and marine transgression. The timing of the shelf-based ice is constrained on land by dating glaciolacustrine sediments with OSL and marine molluscs with radiocarbon and by re-evaluating IRD events in cores from the Fram Strait. Results show that the shelf-based ice started to build-up as early as 30 cal ka BP and reached a maximum during the Last Glacial Maximum (LGM). The shelf-based ice began to retreat ca 16 ka to 10.3 cal ka BP before the final break-up, which took place ca 10.1 cal ka BP probably as a combined result of increased inflow of warm Atlantic water through the Fram Strait, a shallower halocline and higher summer temperatures, corresponding to orbital maximum solar insolation at this time. The existence of extensive shelf-based ice north of Greenland provides an important contribution to the understanding of the LGM glaciation history of the Arctic Ocean. Ó 2010 Elsevier Ltd. All rights reserved. 1. Introduction From the distribution of glacial erratics along the coast, Koch (1923, 1928) concluded that the Greenland Ice Sheet did not reach the north coast, which instead was overrun by outlet glaciers from the “North Cap”, a local ice cap that developed over the coastal mountains. This view has been supported by later fieldwork, as * Corresponding author. Department of Earth Sciences, University of Aarhus, C.F. Møllers Allé 4, DK-8000 Aarhus, Denmark. Tel.: þ45 89422565. E-mail address: nkl@geo.au.dk (N.K. Larsen). 0277-3791/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.quascirev.2010.07.027 reviewed by Dawes (1986) and Funder (1989). To the south and west of the coastal mountains, the North Cap ice merged with the Greenland Ice Sheet (Bennike, 1987; Landvik et al., 2001). On the north coast, it was originally envisaged that piedmont glaciers emanating from the local North Cap terminated either on the coastal plain (Koch, 1923) or on the inner shelf where some of the northernmost islands (Kaffeklubben Ø, Oodaaq Ø) were interpreted as marginal moraines formed by adjacent valley glaciers (Davies, 1963). In contrast, Funder and Larsen (1982) suggested that the outlet glaciers from the North Cap, were deflected eastward by an ice stream from the Greenland Ice Sheet. This was based on coastparallel ice directional glacial landforms and on the occurrence of 3400 N.K. Larsen et al. / Quaternary Science Reviews 29 (2010) 3399e3414 Fig. 1. Overview map of North Greenland with LGM ice reconstructions according to Dawes (1986). glacial erratics with a likely source at Kap Washington 80e100 km to the west. Dawes (1986) added new observations on far-travelled erratics on the north coast, and suggested that the deflection of the glaciers on the coastal plain was caused by an ice shelf which extended farther out in the ocean along the northern coasts of both Ellesmere Island and Greenland (Fig. 1). The erratics indicate that the eastern part of the ice shelf must have been fed from ice sheet outlets in fjords to the west, and especially from Nares Strait where the Innuitian Ice Sheet over the Canadian Arctic Archipelago and the Greenland Ice Sheet joined. However, fieldwork in this region first seemed to negate the existence of a large feeder ice stream in this trough during the LGM, and the idea of the ice shelf was revoked (Funder and Hansen, 1996). But later results from the Canadian side of both the southern and northern end of Nares Strait (Funder, 1989; Blake, 1999; England, 1999) imply that rather thick ice streams did move both north and south through this trough at the LGM. This was supported by the high isostatic uplift, and hence the ice shelf reappeared on the LGM reconstruction of the Greenland Ice Sheet (Funder et al., 2004). We spent two summers on the northern Greenland coast, principally to (i) investigate and date raised marine sediments and drift- wood to reconstruct the Holocene sea ice variability (Funder et al., submitted), (ii) map glacial and glaciolacustrine deposits in valleys and on the coastal plain to determine Holocene valley glaciations (Möller et al., in this volume) and, (iii) to investigate the last glaciation and deglaciation history recorded on the coastal plain and the evidence of a possible extensive ice shelf in the Lincoln Sea between Ellesmere Island and North Greenland e the specific topic of this paper. 2. The north coast of Greenland The field area extends from Kap Morris Jesup in the west to Kap Ole Chiewitz in the east, encompassing the coastal plain (w100 km long and w5e10 km wide) that separate interior, up to 2000 m high, ice covered mountains from the Arctic Ocean (Figs. 1 and 2) The glaciation limit on the coastal fringe facing the Arctic Ocean is w200 m a.s.l. but rises to 800e1000 m a.s.l. in interior Peary Land, from which outlet glaciers descend from local ice caps towards the coastal plain and the ocean (Koch, 1928; Weidick, 2001). The coastal plain is covered mostly by a mosaic of Quaternary sediments; however, thicker deposits occupy valley mouths where rivers reach the Arctic Ocean (Fig. 2). The mean annual temperature at Kap N.K. Larsen et al. / Quaternary Science Reviews 29 (2010) 3399e3414 Fig. 2. Geomorphological map of the north coast of Johannes V. Jensen Land with outcrops and sample localities. Numbers 1e34 refer to localities mentioned in the text, figures or tables. 3401 3402 N.K. Larsen et al. / Quaternary Science Reviews 29 (2010) 3399e3414 Table 1 Radiocarbon dates from North Greenland 0: this study, 1: Möller et al. (in this volume), 2: Funder et al., (submitted). Site Kaffeklubben Kaffeklubben Kaffeklubben Kaffeklubben N lat. Elv Elv Elv Elv Constable Bugt 0705 Kap Ole Chiewitz Kap James Hill 4 1 1 1 W long. m a.s.l. 83.62248 83.62667 83.62197 83.62205 30.55827 30.51728 30.51458 30.51335 12 8 6 8 83.58128 32.21093 106.8 83.43093 83.65603 25.97575 31.00147 12 23 Sample No. Lab. No. Sediment Material Coventional age, yr BP Cal. Age, yr BP KKE KKE KKE KKE LuS7053 LuS7052 LuS7054 LuS7060 Marine Marine Marine Marine 9125 7055 6795 6495 9569.5 7433 7136.5 7398.5 CB 0705:1 LuS7469 Glaciolacustrine 07042 07010 AAR11908 AAR11919 Marine Marine shell Hiatella arctica Mya truncata Terrestrial plant detritus Terrestrial plant detritus Hiatella arctica Hiatella arctica 4:1 1a 1b 1c Morris Jesup is c. 18  C and the coastal plain lies in the zone of continuous permafrost. Precipitation in Peary Land is only around 50 mm/year (Davies and Krinsley, 1963). Patterned ground, including tundra polygons, are widespread and the bedrock composed of quartz-phyllite is heavily frost shattered. Furthermore, deflation surfaces with ventifacts are common, produced by prevailing winds from the west and northwest. 3. Methods Stereoscopic mapping of landforms and sediments, Digital Elevation Model (DEM) extraction and orthophoto generation was done on a digital photogrammetric workstation using aerial imagery from 1960e61 (1:50,000) and 1978 (1:150,000). Control of the imagery was obtained by transferring the absolute reference through time homologous tie points from the younger (1978), aerotriangulated imagery with datum GR96 (WGS84). A series of panoramic scenes were produced where the 1978 ortho-imagery is draped over a digital elevation model extracted from the aerial imagery, acquired by the National Cadastre and Survey (KMS). The DEM has been created using a modification of the original KMS aero-triangulation of the imagery, with emphasis on calibrating sea level. A geomorphological map was produced by manual mapping and classification of the landforms taking advantage of the 92 120 131 106 0 0 0 0 9135  120 10300  125 1 9593  48 9560  55 10124  162 10087  164 2 2     50 50 50 50 Reference     geometric and radiometric resolution of the aerial photographs. The resulting map was then tested by ground-thruthing in the field. Field surveys were conducted from small camps utilizing Twin Otter fixed-wing aircraft, helicopter and ATV’s. Emphasis was placed on stratigraphic analysis and dating of the most complete Quaternary sections. Outcrops were logged using the facies codes of Krüger and Kjær (1999) commonly at a vertical scale of 1:10. Clast fabric measurements in diamicts were made by measuring the dip and dip direction of 25 clasts with long a-axis between 0.5 and 10 cm and an a/b ratio >1.5 in a 20 cm  30 cm area (Larsen and Piotrowski, 2003). Fabric data are presented as points and twosigma Kamb contours on an equal-area, lower-hemisphere Schmidt net. Intact sediment samples were taken in steel mammoth and Kubiena tins and prepared for thin section analysis as described by van der Meer (1993). The elevation of sample sites and marine limit were measured by handheld GPS with build-in altimeter or a digital altimeter. The accuracy of measurements is 1 m after correcting for daily atmospheric pressure changes. In addition, elevations measured in the field were verified using the DEM which has a vertical RMS of 3 m at 90% confidence level. Radiocarbon dates mentioned in this paper are listed in Table 1. Lists of all available 14C-dates from the area are supplied by Funder et al. (submitted). Samples obtained during our fieldwork were age dated at the AMS radiocarbon laboratories in Lund (LuS) and Aarhus Fig. 3. Sedimentary logs documenting the glacial (A), kame and glaciolacustrine deposits (B) from Henson Bugt to Kap Ole Chiewitz. The two units in CB 8 and CB 9 with no designated number are Holocene glaciolacustrine and glacial deposits identified in Sifs Valley (see Möller et al., in this volume). N.K. Larsen et al. / Quaternary Science Reviews 29 (2010) 3399e3414 3403 Fig. 4. (a) Overview of the glacial landscape around Kap James Hill (orthophotograph draped on an elevation model). A till plain (triangles) is present in the central parts above the marine limit (white stippled line). Two outcrops (JH 0706 and JH 0711) show thick successions of basal till, usually massive (b) but also stratified (c). Below the marine limit are marine sediments (blue dots), at one locality (JH 0707) overlying glaciolacustrine deposits. The red dots represent glaciofluvial outwash fan sediments with a large dead-ice depression in the central part (brown). The sedimentary logs from the outcrops are shown in Figs. 3 and 8. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article). (AAR). All marine samples were calibrated into calendar years using the terrestrial and marine calibration curves of Reimer et al., (2004) and Hughen et al., (2004) after subtracting a reservoir age of 550 years (DR ¼ 150; Funder, 1982; Mörner and Funder, 1990). We note that the marine reservoir effect estimated from North Greenland is slightly lower than what has been estimated from Arctic Canada near carbonate outcrops that add additional dead C to the seawater (DR ¼ 335; Dyke, 2004a,b). Sediments for optically stimulated luminescence dating (OSL) were sampled in plastic tubes in sections a least 30 cm from distinct sediment boundaries. Age determinations were carried out at the Nordic Laboratory for Luminescence Dating at Risø (Denmark) using standard techniques including correction for specific water saturation and burial depth (Murray et al., 1987; Olley et al., 1996; Murray and Wintle, 2000). Incomplete bleaching by light is a potential problem for specific types of depositional environments such as ice-dammed lakes, where exposure to direct daylight is often insufficient to reset the OSL signal (Murray and Olley, 2002). In such cases there will be an age overestimation as the remnant dose is superposed on the dose acquired after burial. However, by dating several samples from the same stratigraphic unit, possible outliers caused by incomplete bleaching can often be identified (Larsen et al., 2006). 4. Results The north coast of Peary Land has experienced a complex history of both regional and local glacial advances as well as episodes of marine transgression. Based on several proxies including geomorphology, lithostratigraphy and faunal investigations combined with a firm chronology (14C and OSL dating) four depositional environments have been identified: A) glacial deposits, B) glaciolacustrine deposits, C) marine limit and deposits and D) glaciofluvial deposits. Our studies focus on the coastal plain and to a lesser degree on the northesouth oriented valleys which are often occupied by local outlet glaciers and only record evidence of Holocene valley glaciations. One exception is Sifs Valley where dislodged glaciolacustrine and marine sediments have been observed (Möller et al., in this volume). 4.1. Glacial deposits (A) On the coastal plain, glacial landforms and deposits are indistinct and often overprinted by younger glacial events and periglacial activity. However, in a few places above the marine limit (w45 m a.s.l.) and outside the large moraines that mark the maximum extent of local Holocene glacier advances (Möller et al., 3404 N.K. Larsen et al. / Quaternary Science Reviews 29 (2010) 3399e3414 in this volume), glacial sediments and landforms have been identified (Figs. 2 and 3). From Kap Morris Jesup to Henson Bugt no significant traces of pre-Holocene glacial landforms or sediments were found on the 4e10 km wide coastal plain (Fig. 2). Here beach ridges, a thin cover of marine sediments or fluvial and deltaic sediments drape the frost shattered bedrock. On the eastern side of Henson Bugt towards Constable Bugt, the dominant landscape is a till plain and at higher elevation dead-ice topography dissected by meltwater streams. Despite intense slumping and solifluction a few outcrops reveal successions of grey, clast-rich and compact diamict. Well exposed outcrops are located around Constable Bugt and Kap James Hill on the large till plain (Fig. 4a) composed of 2e6 m thick clast-rich grey diamict (Fig. 4b). Many clasts are stoss/lee-shaped and heavily striated, and amongst them far-travelled erratics like reddish granites have been identified. The diamict is mainly massive and fissile although it may be stratified, consisting of alternating layers of diamict and sand (Fig. 4c). Six fabric analyses show strong preferred eastewest orientation except at one site at the outer coast where the clasts were aligned northwestesoutheast (Fig. 3). Overlying this diamict we encountered younger glaciolacustrine, marine or glacial sediment, especially near Sifs Valley (Möller et al., in this volume). Associated with the diamict or lying directly on local bedrock are volcanic and pink granitic erratic boulders (Fig. 5a), particularly below the highest shoreline and mainly near Kaffeklubben Ø. Farther to the east in the Bliss Bugt area large stoss-lee boulders, often with a long axis of more than 2 m (Figs. 3 and 5b) are scattered on the discontinuous thin cover of diamict. Considering the low relief, flat terrain and boulder size, it is assumed that they are still in their original position. Striation measurements on these large boulders show a strong eastewest alignment (n ¼ 64), as are their long axes (Fig. 5c). In front and superimposed by the large end moraine at Moore Glacier in Bliss Bugt are two, ca 5 km long, single crested ridges that are ESEeWNW orientated (Fig. 2). On the coast from Bliss Bugt to Kap Ole Chiewitz a conspicuous set of coast-parallel landforms several kilometres long and a few metres high were identified on aerial photographs (Fig. 5d). In the field it was realized that these landforms, especially in Bliss Bugt area, were structural lineaments in the local bedrock, except at Kap Ole Chiewitz where these landforms are composed of or covered by a thin layer of diamict. 4.2. Glaciolacustrine deposits (B) Kame and glaciolacustrine sediments have been identified above the highest marine limit (40e45 m a.s.l.) in a few places along the coastal plain (Fig. 2). One of them is at Henson Bugt (Fig. 6a) where glaciolacustrine sediment lies as an erosional remnant at 110 m a.s.l. at the foot of the mountain (Fig. 6b and c). The glaciolacustrine succession is dominantly composed of laminated silt with frequent dropstones, interbedded with gravel and sand stringers and scours (Figs. 3 and 6d). Convolute bedding is common in the lower part of the sequence and the entire succession is cut by a number of normal faults. Glaciolacustrine sediments were also documented in Sifs Valley farther to the east and at much lower elevations (up to 40 m a.s.l.). They are glacially overridden and to a large extent incorporated into the Constable Bugt moraine during an early Holocene valley glacier advance (Möller et al., in this volume). At Kap James Hill (Fig. 2) there is a large area completely covered with thick successions of glaciolacustrine sediments (Fig. 7a). The sequence sits in an onlapping position on the coastal plain but ends with a steep, up to 10 m high scarp towards the Arctic Ocean. North of the scarp we have not observed more glaciolacustrine deposits. The complex has been extensively modified by postglacial fluvial erosion (Fig. 7b and c), which allows good access to the sediments. Gullies cutting through the sediments reveal finely laminated, occasionally graded, silt and fine sand beds (Fig. 7d) with dropstones (Fig. 7e) and iceberg dumps. At one site the glaciolacustrine sediments are overlain by a thin layer of marine sediments with Fig. 5. a) Pink granitic erratic boulder. b) Patchy till plain with scattered boulders and frost shattered bedrock in the Bliss Bugt area. c) Eastewest orientated striations on a large boulder in Bliss Bugt. d) Large lineations in bedrock, overlain by scattered erratic boulders (orange dots). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article). N.K. Larsen et al. / Quaternary Science Reviews 29 (2010) 3399e3414 3405 Fig. 6. Overview of the area from Henson Bugt to Constable Bugt. Between the valleys is a dead-ice landscape (triangles) at higher elevation above the coastal plain. To the west (right) is a relative thin cover of marine sediment (blue dots) on top of bedrock and near the coast there are up to 10 km long beach ridges (blue dotted lines). The Henson and Constable Bugt moraines are indicated by red hatched lines. b) Glaciolacustrine sediments 110 m a.s.l. in Henson Bugt. c) Glaciolacustrine sediments at 110 m a.s.l. in Henson Bugt (site 2 in (a)), clinging to the mountain slope as an erosional remnant. Note man for scale. d) Close-up of fine grained laminated glaciolacustrine sediments with dropstones. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article). molluscs (Fig. 8). Thin section analysis of the glaciolacustrine sediments shows distinct and abrupt boundaries and in places much finer laminations (Fig. 9a). Grading is present, but nongraded laminae make up the bulk of the sediments. Evidence of bioturbation is sparse, but can be pervasive in some strata (Fig. 9b and c), though not noticed macroscopically. No fossils have been observed and thus the organisms responsible for the bioturbation are unknown. Most likely, however, they must have been scavengers as the samples regularly contain tiny, but clearly recognizable slivers of moss-like plant material. West of Bliss Bugt and at Kap Ole Chiewitz, kames were observed close to the highest shoreline (51 m a.s.l.), continuing as small remnants of such up to w100 m a.s.l. No recordable sections were available. Thus, samples for OSL dating were taken from shallow pits excavated down to the permafrost table (Table 2). 4.3. The marine limit and marine deposits (C) The ice retreat on the coastal plain was followed by marine transgression up to the marine limit. This is recorded in sand- and gravel-delta terraces at the valley mouths on the inner plain. The distal delta terrace margin, which reflects sea level at the time of the delta and marine limit formation, is 40e45 m a.s.l. along the entire stretch of coastline. Wherever exposed the delta terraces are 3406 N.K. Larsen et al. / Quaternary Science Reviews 29 (2010) 3399e3414 Fig. 7. a) Overview of Kap James Hill, composed of thick successions of glaciolacustrine sediments. b, c) The sediments are draped by a lag of cobbles and boulders and cut by numerous fluvial gullies. d) Close-up of the laminated and graded sediments with dropstones, some of which have glacial abrasion marks. of Gilbert-type and composed of ca 5 m high foresets. 14C dating of in situ mollusc shells from prodeltaic silt date the end of terrace construction and e assuming that the terraces were built rapidly e the time of deglaciation of the coastal plain. Both at Constable Bugt in the west and at Kap Ole Chiewitz in the east this occurred at c. 10.1 cal ka BP (Table 1). A peculiar landscape on the coastal plain is the occurrence of extensive, barren areas of marine mud with a thin deflation armour of pebbles and boulders at the surface. These are especially conspicuous between Kap James Hill and Bliss Bugt, where raised beaches and signs of wave action are absent. Funder et al. (submitted) suggest that the coast in these areas has been permanently beleaguered by sea ice since deglaciation. Only in embayments with a large inflow of meltwater did sizeable leads arise, allowing wave action and beach formation. In outcrops the marine sediments are uniformly fine grained with only few and diffuse traces of lamination, while the top is reworked by frost action (Fig. 10). In thin sections the marine sediments show diffuse, usually disrupted lamination with very rare grading (Fig. 9d). As in the glaciolacustrine sediments the thin sections of marine deposits show tiny slivers of organic material similar to moss (Fig. 9e). Only one of the samples showed bioturbation similar to, but much less dense, than that sporadically found in the lacustrine deposits (Fig. 9f). 4.4. Glaciofluvial deposits (D) A sandur on both sides of Kaffeklubben River covers a w2 km2 large area (Figs. 2 and 11a) with a flat main plain at an altitude of w54 m a.s.l. (i.e. above the marine limit). The deposits drape underlying bedrock and/or till with a thickness of 5e25 m, as shown in sections along the present river bed of the Kaffeklubben River (Fig. 11b). The sandur sediments consist of a stacked sequence of clast-supported cobble gravel as the main constituent, interbedded with thin beds of massive to planar parallel-laminated sand and thinner beds of clast-supported open work gravels (Fig. 11d). The fluvial feeder system for the constructional part of the sandur is a series of presently dry channels, cut down into till and bedrock and having hanging entrances along the current river valley. Some channels widen in their distal end and merge with the main braidplain, whereas other channels widen and emerge on N.K. Larsen et al. / Quaternary Science Reviews 29 (2010) 3399e3414 Fig. 8. Sedimentary logs of glaciolacustrine (B) and marine sediments (C) at Kap James Hill, Kaffeklubben River (KKR) and Bliss Bugt west. 3407 3408 N.K. Larsen et al. / Quaternary Science Reviews 29 (2010) 3399e3414 Fig. 9. Thin section of glaciolacustrine (aec) and marine (def) sediments. a) The lamination and grading is very distinct in the lacustrine sediments and contain abundant evidence of bioturbation (b, c). d) The marine sediments are massive or diffusively laminated and contain organic (moss) detritus (e). f) Some intervals have been bioturbated but not as intensely as the glaciolacustrine sediments. All images taken in plain light; horizontal field of view is 13.2 mm except for d (9.4 mm) and e (2.6 mm). a hanging terrace w2 m above the main braidplain (Fig. 11a). In distal direction (towards the north), the flat upper surface becomes undulating, forming hummocks and short ridges, intercalated with open and closed, 2e5 m deep depressions (Fig. 11c). 4.5. Chronology The age of the glacial phase has mainly been constrained by dating the glaciolacustrine and marine sediments with OSL and by radiocarbon dating of mollusc shells from the marine sediments. A lower age estimate of the glacial phase is very poorly determined as we have been unable to find any sediment below the till. Old-finite 14 C ages (i.e. ages between 20,000 and 40,000 years) have been shown to be mixtures of non-finite and Holocene, or probably due to contamination (Funder, 1982; Funder et al., submitted), and are therefore omitted from this paper. However, they do show that older marine sediments have been widespread in the area. The glaciolacustrine sediments have been dated with OSL and reveal ages with a large spread between 135e12.4 ka, with a cluster around 20 ka (Fig. 12). In addition, there is one radiocarbon date from Sifs Valley giving an age of 10.3 cal ka BP (Möller et al., in this volume). Some retrieved OSL ages are significantly older than others within the same stratigraphic unit and age reversals are common (Fig. 3). We expect that the “old” population of dates are erroneous because of incomplete bleaching in the turbid muddy water, a dating problem often encountered in ice-proximal sediments (Fuchs and Owen, 2008; Alexanderson and Murray, in press). We only regard the younger ages as reliable and propose that the glaciolacustrine sediments were deposited between ca 16 ka to 10.3 cal ka BP. The deglacial age of the coastal plain is also indicated from two exposure ages of boulders (12 and 16.5 ka) above the marine limit on the coastal plain outside Sifs Valley (Möller et al., in this volume). The termination of the glacial and glaciolacustrine phases is furthermore constrained by radiocarbon dates of molluscs from marine sediments and of drift-wood. The oldest radiocarbon age of a marine mollusc is 10.1 cal ka BP providing a minimum age of the marine transgression as there is a time lag for a fauna to establish (Fig. 12). Marine sediments have also been dated with OSL and they are all a few to several thousand years older than what is expected based on the radiocarbon chronology (e.g. KKR 5, Fig. 8). Apparently also the marine sediments have not been sufficiently bleached before final deposition and the group of “old” ages is probably recording a previous bleaching event. The most likely source of the marine deposits is the fine grained sediment in Sifs Valley deposited during the retreat phase. The OSL ages of these source sediments fall within the same interval as the OSL ages of the marine sediments. It is, therefore, likely that the OSL signal from the marine sediments actually records the time of deposition of glaciolacustrine sediments in Sifs Valley. Associated with the marine transgression is the formation of the large sandur around Kaffeklubben River, terminating close to the marine limit. These sediments gave OSL ages ranging from 24.8 to 5.1 ka with intermediate ages of 12.1 and 13.5 ka (Fig. 12). The 5.1 ka age is most likely too young and this might be related to the sampling location close to a lithological boundary. The intermediate OSL ages are slightly older than expected from radiocarbon dates of marine molluscs (10.1 cal ka BP) and are probably related to incomplete bleaching. In summary, the chronological data from this study suggest that there was a glacial phase during LGM along the north coast of Greenland that was followed by or contemporaneous with the deposition of glaciolacustrine sediments from ca 16 ka to 10.3 cal ka BP, before the coastal plain was inundated around 10.1 cal ka BP. 5. Interpretation and discussion 5.1. The glacial phase The massive diamict on the coastal plain with abraded and striated clasts, fissile structure and high fabric strength is interpreted as a basal till deposited beneath a warm-based glacier with strong interaction between the ice and substratum (Larsen et al., 2004; Evans et al., 2006). Directional elements such as till fabric, striae, the shape of boulders and their axial orientation, together with volcanic erratics originating from Kap Washington, 80e100 km west of the study area, all indicate an eastward ice-flow direction. Eastward ice-flow is also supported by the geographic distribution of glacial deposits and landforms outside and to the Table 2 OSL dates from North Greenland. 0: this study, 1: Möller et al. (in this volume). Map No. N lat. Bliss Bugt W Bliss Bugt W 0702 Bliss Bugt W 0704 Kaffeklubben River Kaffeklubben River Kaffeklubben River Kaffeklubben River Kaffeklubben River Kaffeklubben River Kaffeklubben River Kaffeklubben River Kaffeklubben River Bliss Bugt W Constable Bugt Constable Bugt Constable Bugt Constable Bugt Constable Bugt Constable Bugt Constable Bugt Henson Bugt James Hill James Hill James Hill 0707 James Hill 0707 Kap Ole Chiewitz Kap Ole Chiewitz Kap Ole Chiewitz 8 6 7 18 19 15 16 10 10 10 17 17 5 33 32 31 30 27 28 26 34 24 25 20 20 3 1 2 83.56457 83.55472 83.55750 83.61950 83.62156 83.62082 83.62011 83.61116 83.61116 83.61116 83.61024 83.61024 83.54607 83.58128 83.58128 83.55883 83.56742 83.57818 83.69077 83.58848 83.60843 83.64306 83.63834 83.64250 83.64194 83.41817 83.40182 83.41517 2 3 5 5 5 8 8 W long. 29.97395 29.88028 29.89917 30.80018 30.79866 30.58512 30.60233 30.56297 30.56297 30.56297 30.70187 30.70187 29.76568 32.21093 32.21093 32.00837 32.02294 31.92252 31.94940 31.99593 32.75962 31.40361 31.46299 31.02639 30.95528 25.81592 25.67002 25.77934 Sample No. Lab No. Sediment m a.s.l. n w.c. Equivalent dose (Gy) Dose Rate (Gy/ka) OSL age (ka) Reference 07e038 070801e03 070730e01 06809 06810 06805 06806 06807 06808 06819 06811 06812 07e004 CB 0705:2 CB 0705:3 CB 2:1 CB 3:1 CB 6:1 CB 7:1 CB 9:1 Hensonbugt 2:1 070806e07 070808e11 070805e05 070805e06 07e050 07e051 070813e13 080212 080217 080215 070217 070218 070213 070214 070215 070216 070221 070219 070220 080211 080203 080204 070205 070206 070222 070223 070212 080222 081056 081057 081054 081055 080213 080214 080219 Marine Marine Marine Marine Marine Marine Marine Marine Marine Marine Glaciofluvial Glaciofluvial Kame Glaciolacustrine Glaciolacustrine Glaciolacustrine Glaciolacustrine Glaciolacustrine Glaciolacustrine Glaciolacustrine Glaciolacustrine Glaciolacustrine Glaciolacustrine Glaciolacustrine Glaciolacustrine Kame Kame Kame 43 17 19 40 40 16 15 32 38 30 40 36 37 106.9 108.9 14 14 1.9 22.9 18 108.4 17 20 21 23.8 51 36 33 17 29 18 24 24 20 23 29 39 44 20 52 22 13 20 21 23 25 27 26 17 15 15 15 32 41 49 32 37 52 34 22 27 18 24 30 16 45 36 41 48 46 31 28 37 21 24 33 4.15  4.11  4.17  2.03  2.64  4.00  4.53  4.07  3.77  4.39  3.66  2.22  1.67  3.63  4.01  3.04  3.96  4.11  3.85  4.25  5.07  3.52  3.05  3.71  15.2 21.3 15.6 13.5 12.1 15.8 10.8 15.2 13.6 9.4 5.1 24.8 31 29 45 19.4 39.8 14.2 12.4 50 20 23.4 33.5 26.7 17 16 24 17 14 28 63.0  1.8 88  4 65  2 27  3 32  2 63  3 49  2 62  2 51  3 41  2 18.5  1.2 55  2 52  3 106  17 181  16 60  1 159  17 59  3 48  1 212  16 103  12 82.61  3.04 102.21  3.40 99.26  3.56 no signal 278  15 39  3 60  4 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0.15 0.14 0.13 0.07 0.09 0.13 0.16 0.15 0.14 0.17 0.14 0.09 0.07 0.12 0.14 0.11 0.13 0.13 0.14 0.16 0.21 0.17 0.14 0.16 2.06  0.08 1.85  0.08 3.07  0.12  0.8  1.3  0.7  1.6  0.9  0.9  0.7  0.9  1.1  0.6  0.4  1.4 2 5 4  0.8  4.6  0.8  0.6 4 2  1.5 2  1.6 135  9 21  2 19.6  1.4 N.K. Larsen et al. / Quaternary Science Reviews 29 (2010) 3399e3414 Site 3409 3410 N.K. Larsen et al. / Quaternary Science Reviews 29 (2010) 3399e3414 Fig. 10. Sedimentary logs from the outcrops along Kaffeklubben River, showing marine sediments (C) above glaciofluvial sediments (B) from the thinning phase. east of valley mouths on the coastal plain such as the ESEeWNW orientated ridges in front of Moore Glacier, which are interpreted as eskers (Fig. 2). In theory these landforms and sediments could be attributed to local outlet glaciers, forming piedmont lobes on the coastal plain (cf. Davies, 1963). However, this would require that the local outlet glaciers were deflected eastward; possibly by the presence of thick multiyear sea ice, which has been inferred to have existed during the LGM and during the early part of deglaciation (Bradley and England, 2008). It would also require that the erratic boulders were transported by sea ice to the study area and reworked by local glaciers; as suggested by Funder and Larsen (1982). However, the eastewest orientated till fabric and striae just outside valley mouths are difficult to reconcile with local outlet glaciers, but are better explained by a more regional ice-flow pattern parallel to the coastal plain. Another fundamental problem is the glaciolacustrine sediments (up to 100 m a.s.l.) that occasionally occur above the marine limit (45 m a.s.l.) on the coastal plain and in the local valleys. Obviously these sediments are not marine and require the presence of ice on or outside the coastal plain to form a dam against the mountains when the local valleys were partly deglaciated (see below). Accordingly, we argue that the observations are best explained by the existence of shelf-based ice, nourished by outlet glaciers from North Greenland that locally would provide the dynamics needed to produce glacial landforms and deposits on the coastal plain farther to the east. Dawes (1986) suggested that this ice shelf occupied the entire Lincoln Sea between Ellesmere Island and North Greenland and was fed by glaciers all along the north coast of Greenland and possibly by some of the major outlet glaciers draining the Innuitian Ice Sheet on Ellesmere Island (Fig. 1). We find that idea compelling and it also seems to comply with an ice stream operating in the northern part of the Nares Strait during LGM as suggested by England et al. (2006). However, more data from especially Northwest Greenland is needed to confirm such an extensive ice shelf in the Lincoln Sea. From North Greenland, we find evidence of shelf-based ice on the coastal plain nourished by outlet glaciers that might have formed an ice shelf on the outer shelf, but this has yet to be resolved. Offshore multibeam mapping and a sediment core on the Morris Jesup Rise a ca 200 km north of our study area (Jakobsson et al., in this volume) show that ice was not grounded here during the LGM. Iceberg scours have been identified on as much as 1045 m water depth but are dated to MIS 6. However, these observations do not exclude the existence of an ice shelf north of Greenland and in the Lincoln Sea during the LGM, but simply show that the ice shelf was too thin to become grounded on the deeper parts of Morris Jesup Rise. It should also be noted that only a restricted part of the Morris Jesup Rise was mapped and the landward and shallower areas are unexplored and may contain evidence of ice grounding. N.K. Larsen et al. / Quaternary Science Reviews 29 (2010) 3399e3414 3411 Fig. 11. a) Aerial photograph of the Kaffeklubben River sandur. Open yellow ovals: collapsed, hummocky distal part of the sandur plain. Red line demarcates the boundary between sandur sediments and till/bedrock areas. Two till “islands” are protruding through sandur sediments distal to one of the feeder channels (yellow). b) The flat proximal part has an elevation of w54 m a.s.l. In the foreground Kaffeklubben River cuts the 5e20 m thick sandur deposits and underlying bedrock. c) The collapsed, distal part of the sandur plain. d) Sedimentary log from the sandur (KKR 8). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article). 5.2. Ice retreat and marine transgression Both kame and glaciolacustrine sediments have been observed above the marine limit along the coastal plain. The coarse-grained kame sediments are often located just outside and eastward of valley mouths and border the coastal mountains. The non-fossiliforous glaciolacustrine sediments located both above and below the marine level at several places along the coastal plain and in Sifs Valley suggest deposition primarily from suspension, while the scattered dropstones record episodes of ice-rafting. Coarser Fig. 12. OSL and radiocarbon ages. a) Glaciolacustrine and kame deposits. Grey bar ¼ inferred age. b) Marine and glaciofluvial sediments. Grey bar ¼ inferred age of marine transgression. 3412 N.K. Larsen et al. / Quaternary Science Reviews 29 (2010) 3399e3414 interbeds of sand and gravel are likely to reflect underflows or sediment gravity flows from adjacent valley slope. It is possible to distinguish the glaciolacustrine sediments, occurring below the marine limit, from the marine sediments on the basis of differences in the frequency of lamination, grain size, nature of boundaries (Ó Cofaigh and Dowdeswell, 2001), and fossil content. Furthermore the glaciolacustrine sediments are restricted in their extent, whereas the marine deposits are much more widespread over the coastal plain. We thus conclude that the glaciolacustrine sediments most likely reflect temporary and disconnected embayments between the coastal mountains, their outlet glaciers and the shelfbased ice occupying the inner shelf. The marine sediment consists of diffusively laminated to massive and often shell-bearing silt and clay and reflects sedimentation from suspension in a shallow glaciomarine environment. Dropstones in the marine sediments, many of which are striated, indicate that icebergs were frequent in this area (Funder et al., submitted). The coarsening upward succession of marine sediments at Kaffeklubben River (Fig. 10) is interpreted to reflect a forced regression caused by the isostatic uplift of the land, ending in foreshore/beach face sands and gravels. Associated with the marine limit is a sandur at Kaffeklubben River (Fig. 11) that was deposited from a glaciofluvial feeder system emanating from the valley emerging from the mountain range (ca 13 km) south-westward from the present river mouth. We interpret the succession of kame, glaciolacustrine, glaciofluvial and marine deposits as a continuum in plaeoenvironmental change, related to the deglaciation of shelf-based ice nourished by outlet glaciers. The kames were deposited at an early stage when ice on the coastal plain had downwasted and retreated sufficiently to allow lateral ice-dammed lakes to form between the ice and the coastal mountains. Later, as ice retreat continued, glaciolacustrine sediments was deposited in larger ice-marginal lakes on the coastal plain and in the south-trending valleys, e.g. as in Sifs Valley, implying that the local outlet glaciers had abandoned the outer parts of the valleys when ice was still present on the coastal plain. In a last stage, when ice totally disappeared from the coastal plain there was a marine transgression and deltas and shallow marine sediments were deposited up to the marine limit at ca 45 m a.s.l. The transition from late-stage glaciolacustrine to marine conditions was gradual as suggested from similar lake- and sea-level altitude. The large sandur around Kaffeklubben River seems to be associated with the marine limit and the pitted appearance in its distal part suggests melt-out of buried dead-ice. This dead-ice could possibly be remnants of the outlet glaciers feeding the shelf-based ice into which the sandur prograded during its build-up, filling out the depositional spaces between the ice blocks and covering them. Due to the insulating effect of the sediments, the melt-out process was slow and it probably continued long after the depositional phase of this system. A somewhat similar scenario has been described from Arctic Canada where a large ice stream impinged over a 400 km stretch up against the coastal mountains (e.g. England et al., 2009). Here proglacial lakes initially were formed, later transformed into epishelf lakes when the ice stream had thinned and was replaced by an ice shelf. However, in North Greenland the damming was initially caused by eastward flowing outlet glaciers, forming shelf-based ice deflected eastwards by sea ice in the Arctic Ocean. The initiation area of this ice must have been in the west; the glaciolacustrine sediments found in for instance Sifs Valley, demonstrate that local outlet glaciers did not feed the ice on the shelf after LGM but rather outlets farther to the west, thus also testifying the regional character of the shelf-based ice. Fig. 13. a) Temporal evolution of the shelf-based ice: main phase (light grey), retreat phase (dark grey) and marine transgression (darkest grey). Arrow indicates inflow of warm Atlantic water (Bradley and England, 2008). b) OSL and radiocarbon dates of kame and glaciolacustrine sediments (black) and marine and glaciofluvial sediments (red). c) Detrital Fe grains from PS1230 (Darby et al., 2002; Darby and Zimmerman, 2008) in the Fram Strait matched to North Greenland (black), Laurentide Ice Sheet (green) and Innuitian Ice Sheet (red). Dashed line represents the minimum percentage of Fe grains for a statistically significant match to the North Greenland area. Percents are weighted as in earlier papers (e.g. Darby, 2003) so as to avoid misleading percentages where low numbers of grains occur. The weighted percent is the percent of Fe grains matched to a source multiplied by the number of these grains divided by 10, which is a conservative estimate of the number of Fe grains required for a definitive match. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article). N.K. Larsen et al. / Quaternary Science Reviews 29 (2010) 3399e3414 5.3. Regional implications Ice shelves in the western part of the Arctic Ocean are to a large degree controlled by the presence of thick multiyear sea ice that prevented them from calving into the ocean (Dawes, 1986; Lemmen et al., 1988; Funder et al., 2004; Engels et al., 2008; Jakobsson et al., in this volume). Such close interaction suggests that the stability of ice shelves is intimately linked to the presence of sea ice, which in turn is controlled by a complex of factors, including the variable inflow of warm Atlantic water to the Arctic Ocean (e.g. Bradley and England, 2008). It has also been discussed by Jakobsson et al. (in this volume) how Atlantic water inflow may have been altered during past glacial periods, and they suggest that while Atlantic water still entered the Arctic through the Fram Strait, its reduced volume and a deeper halocline resulting from reduced glacial stage input of freshwater to the Arctic, may have prevented basal melting of shelf-based ice and sea ice. As sea level recovered during deglaciation, the volume of Atlantic water into the Arctic Ocean would increase and the halocline become shallower and this may have led to a rapid reduction of sea ice and subsequent disintegration of ice shelves including the shelf-based ice occupying the Arctic Ocean north of Greenland. Based on the terrestrial chronology it is proposed that the shelf-based ice began to build-up prior to the LGM around 30 ka, after which it started to degrade between ca 16 ka to 10.3 cal ka BP, finally giving way to a marine transgression at ca 10.1 cal ka BP (Fig. 13). Additional information about the timing has been obtained by identifying ice-rafted Fe oxide grains from the North Greenland study area in the well-dated core PS1230 from the Fram Strait (Darby et al., 2002; Darby and Zimmerman, 2008). These sand-sized Fe grains (45e250 mm but mostly about 100 mm) can be precisely matched to samples from the entire circum-Arctic, including the study area, by their unique chemical composition as measured by electron probe microanalysis of 12 elements in each Fe grain and matched to all compositions of the same mineral in the circum-Arctic database, which consists of more than 16,000 grain analyses that were clustered and tested for compositional uniqueness for each of nine Fe oxide mineral types (Darby, 2003). The North Greenland Fe grains begin to increase above background levels at w30 cal ka BP with larger peaks that are coincident with peaks in Fe grains and IRD from the Laurentide and Innuitian Ice Sheets (Darby et al., 2002; Darby and Zimmerman, 2008). The Fe grains from North Greenland rapidly decrease at 22 cal ka BP, dropping to below significant levels for valid matches to North Greenland by 12e11 cal ka BP (Fig. 13c). These data show that outlet glaciers from the Greenland Ice Sheet and local ice caps in North Greenland delivered icebergs to the Arctic Ocean and the Fram Strait outflow area of core PS1230 by w30 ka and we suggest that the shelf-based ice started to grow at this time. The early build-up towards a full sized shelfbased ice around 20 ka compares well with geophysical modelling of postglacial rebound that infer a fully established Innuitian Ice Sheet at w19.5 ka (Tarasov and Peltier, 2004). The reduction in North Greenland Fe grains from 22 to 16 cal ka BP probably reflects that local outlet glaciers initially were replaced and blocked by ice coming from farther to the west. From 16 cal ka BP until the final break-up of the shelf-based ice at 10.1 cal ka BP local outlet glaciers began to retreat from the coastal areas and ice-dammed lakes were formed between the shelf-based ice and the coastal mountains. Final disintegration of the shelf-based ice began ca 10.1 cal ka BP, which is also well documented by the timing of the marine transgression and from the termination of ice-rafted Fe grain from North Greenland (Fig. 13c). The lack of renewed influx of North Greenland Fe grains once the shelf-based ice broke-up seems to supports our interpretation that local 3413 outlet glaciers already had retreated at this time. The marine transgression coincided with increased inflow of warm Atlantic water which occurred from 13e11 ka (Bradley and England, 2008), a shallower halocline (Jakobsson et al., in this volume) and this along with the recorded summer insolation maximum (Berger and Loutre, 1991), leading to the Holocene Thermal Maximum, served as primary triggers for the disintegration of the shelf-based ice (Fig. 13). 6. Conclusions On the northernmost land stretch in Greenland facing the Arctic Ocean we found evidence of a shelf-based glaciation during the LGM. Glacial landforms and ice-flow indicators such as clast fabric and striations on boulders and the presence of erratic boulders from Kap Washington, 80e100 km away, all points toward eastward ice-flow along the coastal plain. This ice was nourished by outlet glaciers coming from further west along the coast. It is speculated that the ice shelf dynamics was to a large degree controlled by the presence of thick multiyear sea ice in the Arctic Ocean, preventing it from breaking up and forcing it to flow along the northern coast of North Greenland. The shelf-based ice probably started to build-up as early as 30 ka and it reached a maximum during the LGM. It started to retreat at ca 16 ka to 10.3 cal ka BP but calved large amounts of icebergs at around 22 ka. During this period kames and glaciolacustrine sediments were deposited between the ice shelf and the coastal mountains to the south. The final break-up of shelf-based ice and the consecutive marine transgression took place ca 10.1 cal ka BP, probably as a combined result of increased inflow of warm Atlantic water through the Fram Strait, a shallower halocline and higher summer temperatures, corresponding to orbital maximum solar insolation at this time. Acknowledgements The LongTerm project, with field seasons on northernmost Greenland in 2006 and 2007, was conducted under the umbrella of the International Polar Year (IPY) endorsed APEX (Arctic Palaeoenvironments and its Extremes) project. A substantial part of the logistic effort was funded by the Danish Research Council (FNU) and the Commission for Scientific Research in Greenland (KVUG). Logistic planning and support came from the Danish Polar Centre. The Swedish participation was funded by the Swedish Research Council (VR) with additional funding from the Crafoord Foundation, Ymer-80, Kungliga Fysiografiska Sälskapet and Stiftelsen Lars Hiertas Minne. Additional logistic support was also provided by the Swedish Polar Research Secretariat. Jaap van der Meer acknowledges a leave of absence granted by Queen Mary University while Head of Department. We thank Prof. John H. England and an anonymous reviewer for valuable comments that significantly improved the paper. References Alexanderson, H., Murray, A.S. Problems and potential of OSL dating Weichselian and Holocene sediments in Sweden. Quaternary Science Reviews, in press. Bennike, O., 1987. Quaternary geology and biology of the Jørgen Brønlund Fjord area, north Greenland. Meddelelser Om Grønland Geosciences 18, 1e23. Berger, A., Loutre, M.F., 1991. Insolation values for the climate of the last 10 million years. Quaternary Science Reviews 10, 297e317. Blake, W., 1999. Glaciated landscapes along Smith Sound, Ellesmere island, Canada and Greenland. Annals of Glaciology 28, 40e46. Bradley, R.S., England, J.H., 2008. The Younger Dryas and the sea of ancient ice. Quaternary Research 70, 1e10. Darby, D.A., 2003. Sources of sediment found in sea ice from the western Arctic Ocean, new insights into processes of entrainment and drift patterns. Journal of Geophysical Research 108 (C8), 3257. doi:10.1029/2002JC001350. 13-1 to 13-10. 3414 N.K. Larsen et al. / Quaternary Science Reviews 29 (2010) 3399e3414 Darby, D.A., Bischof, J.F., Spielhagen, R.F., Marshall, S.A., Herman, S.W., 2002. Arctic ice export events and their potential impact on global climate during the late Pleistocene. Paleoceanography 17. doi:10.1029/2001pa000639. Darby, D.A., Zimmerman, P., 2008. Ice-Rafted detritus events in the Arctic during the last glacial interval and the timing of the Innuitian and Laurentide ice sheet calving events. Polar Research 27, 114e127. Davies, W.E., 1963. Glacial geology of northern Greenland. Polarforschung 5, 94e103. Davies, W.E., Krinsley, D.B., 1963. The Recent Regimen of the Ice Cap Margin in North Greenland, vol. 58. International Association of Scientific Hydrology, Publication. 119e130. Dawes, P.R., 1986. Glacial erratics on the Artic Ocean margin of North Greenland: implications for an extensive ice-shelf. Bulletin of the Geological Society of Denmark 35, 59e69. Dyke, A.S., 2004a. An outline of North American Deglaciation with emphasis on central and northern Canada. In: Ehlers, J., Gibbard, P.L. (Eds.), Quaternary Glaciations-Extent and Chronology. Part II. North America. Developments in Quaternary Science, vol. 2b. Elsevier, Amsterdam. Dyke, A.S., 2004b. An Outline of North American Deglaciation with Emphasis on Central and Northern Canada. Engels, J.L., Edwards, M.H., Polyak, L., Johnson, P.D., 2008. Seafloor evidence for ice shelf flow across the Alaska-Beaufort margin of the Arctic Ocean. Earth Surface Processes and Landforms 33, 1047e1063. England, J., 1999. Coalescent Greenland and Innuitian ice during the last glacial maximum: revising the Quaternary of the Canadian high Arctic. Quaternary Science Reviews 18, 421e456. England, J., Atkinson, N., Bednarski, J., Dyke, A.S., Hodgson, D.A., Cofaigh, C.O., 2006. The Innuitian Ice Sheet: configuration, dynamics and chronology. Quaternary Science Reviews 25, 689e703. England, J.H., Furze, M.F.A., Doupé, J.P., 2009. Revision of the NW Laurentide ice sheet: implications for paleoclimate, the northeast extremity of Beringia, and Arctic Ocean sedimentation. Quaternary Science Reviews 28, 1573e1596. Evans, D.J.A., Phillips, E.R., Hiemstra, J.F., Auton, C.A., 2006. Subglacial till: formation, sedimentary characteristics and classification. Earth Science Reviews 78, 115e176. Fuchs, M., Owen, L.A., 2008. Luminescence dating of glacial and associated sediments: review, recommendations and future directions. Boreas 37, 636e659. Funder, S., 1982. C-14 dating of samples collected during the 1979 expedition to North Greenland. Grønlands Geologiske Undersøgelse Rapport 110, 9e14. Funder, S., 1989. Quaternary geology of the ice-free areas and adjacent shelves of Greenland. In: Fulton, R.J. (Ed.), Quaternary Geology of Canada and Greenland. Geological Survey of Canada, Geology of Canada, pp. 743e792. Funder, S., Hansen, L., 1996. The Greenland ice sheeteea model for its culmination and decay during and after the last glacial maximum. Bulletin of the Geological Society of Denmark 42, 137e152. Funder, S., Jennings, A.E., Kelly, M., 2004. Middle and late Quaternary glacial limits in Greenland. In: Ehlers, J., Gibbard, P.L. (Eds.), Quaternary Glaciations-Extent and Chronology. Part II. North America. Developments in Quaternary Science, vol. 2b. Elsevier, Amsterdam. Funder, S., Kjær, K.H., Korsgaard, N.J., Larsen, N.K., Linderson, H., Lyså, A., Möller, P., Olsen J. A Holocene sea ice record from Northern Greenland based on raised marine shorelines and driftwood. (submitted). Funder, S., Larsen, O., 1982. Implications of volcanic erratics in Quaternary deposits of North Greenland. Bulletin of the Geological Society of Denmark 31, 57e61. Hughen, K.A., Baillie, M.G.L., Bard, E., Beck, J.W., Bertrand, C.J.H., Blackwell, P.G., Buck, C.E., Burr, G.S., Cutler, K.B., Damon, P.E., Edwards, R.L., Fairbanks, R.G., Friedrich, M., Guilderson, T.P., Kromer, B., McCormac, G., Manning, S., Ramsey, C.B., Reimer, P.J., Reimer, R.W., Remmele, S., Southon, J.R., Stuiver, M., Talamo, S., Taylor, F.W., van der Plicht, J., Weyhenmeyer, C.E., 2004. Marine04 marine radiocarbon age calibration, 0-26 cal kyr BP. Radiocarbon 46, 1059e1086. Jakobsson, M., Nilsson, J., O’Regan, M.A., Backman, J., Löwemark, L., Dowdeswell, J.A., Colleoni, F., Marcussen, C., Anderson, L., Björck, G., Darby, D.A., Eriksson, B., Hanslik, D., Hell, B., Mayer, L., Polyak, L., Sellén, E., Wallin,  A. An Arctic Ocean ice shelf during MIS 6 constrained by new geophysical and geological data. Quaternary Science Reviews, in this volume. Koch, L., 1923. Preliminary report upon the geology of Peary land, Arctic Greenland. American Journal of Science 5, 190e199. Koch, L., 1928. Contributions to the glaciology of north Greenland. Meddelelser Om Grønland 65, 181e464. Krüger, J., Kjær, K.H., 1999. A data chart for field description and genetic interpretation of glacial diamicts and associated sediments e with examples from Greenland, Iceland and Denmark. Boreas 28, 386e402. Landvik, J.Y., Weidick, A., Hansen, A., 2001. The glacial history of the Hans Tausen Iskappe and the last glaciation of Peary land, north Greenland. Meddelelser Om Grønland Geoscience 39, 27e44. Larsen, E., Kjær, K.H., Demidov, I.N., Funder, S., Grøsfjeld, K., Houmark-Nielsen, M., Jensen, M., Linge, H., Lyså, A., 2006. Late Pleistocene glacial and lake history of northwestern Russia. Boreas 35, 394e424. Larsen, N.K., Piotrowski, J.A., 2003. Fabric pattern in a basal till succession and its significance for reconstructing subglacial processes. Journal of Sedimentary Research 73, 725e734. Larsen, N.K., Piotrowski, J.A., Kronborg, C., 2004. A multiproxy study of a basal till: a time-transgressive accretion and deformation hypothesis. Journal of Quaternary Science 19, 9e21. Lemmen, D.S., Evans, D.J.A., England, J., 1988. Canadian landform example 10: ice shelves of northern Ellesmere Island, NWT. Canadian Geographer 32, 363e367. Murray, A.S., Marten, R., Johnston, A., Martin, P., 1987. Analysis for naturally-occurring radionuclides at environmental concentrations by gamma spectrometry. Journal of Radioanalytical and Nuclear Chemistry-Articles 115, 263e288. Murray, A.S., Olley, J.M., 2002. Precision and accuracy in the optically stimulated luminescence dating of sedimentary quartz: a status review. Geochronometria 21, 1e16. Murray, A.S., Wintle, A.G., 2000. Luminescence dating of quartz using an improved single-aliquot regenerative-dose protocol. Radiation Measurements 32, 57e73. Möller, P., Larsen, N.K., Kjær, K.H., Funder, S., Schomacker, A., Linge, H., Fabel, D. Early to middle Holocene valley glaciations on northernmost Greenland. Quaternary Science Reviews, in this volume. Mörner, N.-A., Funder, S., 1990. C-14 dating of samples collected during the NORQUA 86 expedition, and notes on the marine reservoir effect. In: Funder, S. (Ed.), Late Quaternary Stratigraphy and Glaciology in the Thule Area. Geoscience, Northwest Greenland, Meddelelser om Grønland, pp. 57e59. Ó Cofaigh, C., Dowdeswell, J.A., 2001. Laminated sediments in glacimarine environments: diagnostic criteria for their interpretation. Quaternary Science Reviews 20, 1411e1436. Olley, J.M., Murray, A., Roberts, R.G., 1996. The effects of disequilibria in the uranium and thorium decay chains on burial dose rates in fluvial sediments. Quaternary Science Reviews 15, 751e760. Reimer, P.J., Baillie, M.G.L., Bard, E., Bayliss, A., Beck, J.W., Bertrand, C.J.H., Blackwell, P.G., Buck, C.E., Burr, G.S., Cutler, K.B., Damon, P.E., Edwards, R.L., Fairbanks, R.G., Friedrich, M., Guilderson, T.P., Hogg, A.G., Hughen, K.A., Kromer, B., McCormac, G., Manning, S., Ramsey, C.B., Reimer, R.W., Remmele, S., Southon, J.R., Stuiver, M., Talamo, S., Taylor, F.W., van der Plicht, J., Weyhenmeyer, C.E., 2004. IntCal04 terrestrial radiocarbon age calibration, 0-26 cal kyr BP. Radiocarbon 46, 1029e1058. Tarasov, L., Peltier, W.R., 2004. A geophysically constrained large ensemble analysis of the deglacial history of the North American ice-sheet complex. Quaternary Science Reviews 23, 359e388. van der Meer, J.J.M., 1993. Microscopic evidence of subglacial deformation. Quaternary Science Reviews 12, 553e587. Weidick, A., 2001. Neoglacial glaciations around Hans Tausen Iskappe, Peary land, north Greenland. Meddelelser Om Grønland Geosciences 39, 5e26.