Evolution of the Scoresby Sund Fan, central East Green ...

NORWEGIAN jOURNAl OF GEOLOGY Scoresby Sund Fan, East Greenland 3

Evolution of the Scoresby Sund Fan, central East Green land - evidence from ODP Site 987

Fei sal A. Butt, Anders Elverhøi, Carl Fredrik Forsberg & Anders Solheim

Butt, F. A., Elverhøi, A., Forsberg, C. F. & Solheim A.: Evolution of the Scoresby Sund Fan, central East Greenland- evidence from ODP Site 987. Norsk Geologisk Tidsskrift, Vol81, pp. 3-15. Trondheim 200 l. ISSN 0029-196X.

Results of sedimentological analyses (grain-size distribution, bulk and day mineralogy, TC/TOC) from ODP Site 987 show that the sedimentary succession off Scoresby Sund fjord complex consists of three dominantly hemipelagic phases ( with varying input from ice rafting) and two de bris flow units. The presence of ice on continental central East Greenland throughout the last -7.5 Ma is indicated by the continuous presence of dasts in the sediments. The debris flow units, with bases at -5 Ma and -2.6 Ma respectively are interpreted to represent major glacial advances across the central East Greenland shelf. It is suggested that the shelf areas were free of any major ice during the hemipelagic phases. Based on mineralogy and dast (>l cm) compositions, a progressive upward shift in the main sediment source area from volcanic regions to the sedimentary Jameson Land is proposed.

F. A. Butt & A. Elverhø� Department of Geology, University of Oslo, P.O.Box l 047, Blindern, N-0316 Oslo, Norway; C. F. Forsberg, Norwegian Polar Institute, Polarmiljøsenteret, N-9296 Tromsø, Norway; A. Solheim, Norwegian Geotechnical Institute, P.O.Box 3930, Ullevaal Stadion, N-0806 Oslo, Norway

lntroduction

The Polar North Atlantic margins comprising of Norway, Barents Sea and Svalbard to the east and East Greenland to the west have been strongly influenced by gladal build-up and decay over the past few million years (Funder et al. 1994; Elverhøi et al. 1998; Solheim et al. 1998). The two margins, however, show distinctly different morphologies indicating different mechanisms of shelf development. The Svalbard-Barents Sea Margin is characterized by a series of thick sedimentary wedges offshore of shelf-cutting troughs and major fjord systems (Figure l) (Faleide et al. 1996; Solheim et al. 1996; Vorren et al. 1998). In contrast, no such deposits are found on the East Greenland Margin from 71 °N to 79°N (Mienert et al. 1993, 1995; Dowdeswell et al. 1996; Solheim et al. 1998). There is however one large submarine sediment deposit off the Scoresby Sund fjord system between 69°N and 71 °N known as the Scoresby Sund Fan (Funder & Larsen 1989).

This paper attempts to describe the major phases in the glacial evolution of the Scoresby Sund Fan in view of the results from Ocean Drilling Program (ODP) Site 987 (Figure 1). The site was drilled on the ocean floor off Scoresby Sund and, combined with earlier studies, provides the opportunity for a detailed study of the depositional environments and processes during the last -7.5 m.y. The Scoresby Sund Fan has previously been described by Larsen (1990), Vanneste (1995), Vanneste et al. (1995)

and Dowdeswell et al. ( 1997) based on seismic data. The sedimentary record from the area has thus far been restricted to a few shallow cores covering the last 250 kyr (Nam et al. 1995).

location and Morphology

The East Greenland Margin extends from roughly 60°N to about 80°N (Figure 1). The southern and northern parts of the margin differ in structural styles and tectonic development. The southern part of the margin is made up of Precambrian basement rocks whereas the northern half comprises a coast-parallel mountain belt of Caledonian age and Palaeozoic and Mesozoic rift basins (Larsen 1990). The Reykjanes Ridge, the Icelandic spreading centre, the Kolbeinsey Ridge and the Mohns Ridge mark the seaward boundary of the East Greenland Continental Margin (Figure 1).

Shallow banks with water depths around 100-200 m are present on the Northeast Greenland Shelf. In the southern two-thirds of the shelf, water depths generally range from 250 to 350 m. On the shelf, transverse channels originating from the major fjords show depths between 400 and 600 m (Larsen 1990).

The Scoresby Sund fjord complex is situated along the central East Greenland coast between 69°-72°N and 22°-29°W (Figure 1). It represents a major trough that

f. A. Butt et al. NORWEGIAN jOURNAL OF GEOLOG Y

Figure l. Map of the Polar North Atlantic region with location of the ODP Site 987. Bathymetry of the shelf and slope areas is shown. EGC =East Greenland Current, JMFZ =fan Mayen Fracture Zone, NAW =North Atlantic Water. (Butt et. al.)

links the modem Greenland lee Sheet with the continental shelf and slope. The fjord complex consists of a number of fjords of varying sizes and extends deep into the continent with a distance of about 350 km from the heads of the inner fjords to the mouth of the main fjord. The fjord complex is strongly branched incorporating a number of peninsulas and islands. The inner fjords are narrow (<6 km wide) and deep (over 1000 m) whereas the outer fjord arms are wide (up to 50 km) and shallow ( -600 m). There is no sill at the main fjord entrance (Funder 1989; Vanneste 1995).

The shelf region in the vicinity of the Scoresby Sund Fjord complex can be divided into two geological provinces (Larsen 1990). These include the Blosseville Kyst Basin to the south consisting of Mesozoic to Tertiary sediments underlain by volcanic basement and the Liverpool Land Basin to the north (Figure 2). The Liverpool Land Basin, which also contains a thick (-10 km) sedimentary pile over oceanic/volcanic basement (Larsen 1990), is further subdivided into an Inner Liverpool Land Basin and an Outer Liverpool Land Basin. The Scoresby Sund Fan occupies the depocentre of the Outer Liverpool Land Basin where it forms the }argest accumulation of Neogene sediments along the entire East Green-

land Margin (Figure 2). The fan has a slope of about 2° beyond the shelf break with the total sediment volume estimated at 30 000 ± 10 000 km3 (Dowdeswell et al. 1997). The Iceland Plateau and the Kolbeinsey Ridge form the eastern boundary of the basin.

The evolution of the area has been strongly influenced by the Tertiary tectonic events, which principally consist of an early phase of rift basin formation and a later p hase of seafloor spreading (Larsen 1990).

ODP Site 987

Site 987 was drilled on the eastern extremity of GGU line 82-12 on the northern flank of the Scoresby Sund Fan (Figure 2) at a water depth of -1670 m. The total drilled depth was to 859.4 metres below sea floor (mbsf). The hole did not reach the basement, but the basement, based on seismic data, is inferred to be only a few metres from the bottom of the drill hole (Jansen et al. 1996). The base of the hole is dated at -7.5 Ma (Figure 3) (Channell et al. 1999). Dating is primarily based on paleomagnetic data. The magnetostratigrap-

NORWEGIAN jOURNAL OF GEOLOGY

69°

2QO

Scoresby Sund Fan, East Greenland 5

69°

l � "'

"' l Early Tertiary

L. �--"'--"'_j_ volcanics ��=��==�� Pal7ozoic and Mesozoic -==-===-==== sed1mentary rocks

Figure 2. The Scoresby Sund Fan (SSF) with geology of the surrounding continental areas. Location of Si te 987 and seismic lines GG U 82-12 (Larsen, 1990) and 90600 (Vanneste et al., 1995) are marked. ILL = Inner Liverpool Land Basin, OLL = Outer Liverpool Land Basin. The dashed line between ILL and OLL is only an approximation of the boundary between the two basins. (Butt et.al.)

hic interpretations have been checked against a shorebased study of dinoflagellate cysts from Site 987 (M. Smelror, unpublished data). Late Cenozoic dinoflagellate zones for the Norwegian-Greenland Sea defined by Paulsen et al. (1996) from ODP Leg 151 were used for comparisons.

The lithology at the site principally consists of silty clay with varying amounts of dropstones and clasts. The clasts range in size from <l cm in diameter to over l O cm in extreme cases (Jansen et al. 1996). The shipboard scientific party (Jansen et al. 1996) subdivided the succession at the Site 987 into five principal lithostratigraphic units (Figure 3) based on visual core descriptions, smear

slide examination, measurements of bulk calcium carbonate, spectral reflectance, magnetic susceptibility and natura! gamma radiation and X-ray diffraction of bulk sediments. This lithological subdivision along with the ages of the various units is shown in Figure 3.

Methodology Around 340 samples (10-20 cm3) from hales 987D (upper 373 mbsf) and 987E (373-859 mbsf) have been analysed at regular intervals (l sample per 50 cm of recovered co re) from top to bottom.

6 f. A. Butt et al.

o

100

200

300

c

..8 400 .s

f o

500

600

700

800

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

Lithological Unit

l

11

l liA

IIIB

IV

V

Simplified Lithology

Silty clay

Silty clay with sand and gravel (debris flow)

Silty clay

Silty clay with interbedded sand

Silty clay with sand and gravel (debris flow)

Silty clay

Base of drilled section ...

900

Age (Ma)

-0.78 -0.99 -1.07

- 1.77 - 1.95

• - 2.14

2.58

c- 3.11

1- 3.58

1- 4.29 ...

1- 5.23

r- 5.89 • c- 6.57 ·:·

c- 7.43

Figure 3. Lithological units together with simplified lithology of the Site 987 succession (from ]ansen et al., 1996). The ages are magnetostratigraphically deduced (Channell et al., 1999). Dinoflagellate events defined by Paulsen et al. (1996) from ODP Leg 151, are also indicated. • LAD Filisphaera filifera, T LAD Recticulatosphaera actinocoronata, + LAD Evitosphaerula?, •:• LAD Labrinthodinium truncatum (LAD = Last Appearance Datum). (Butt et. al.)

Grain size analyses

Samples (5-20 grams, mostly approx. 10 grams) were wet sieved to remove the <63 fliD fraction. The proportion of silt ( 63-2 fliD) and clay ( <2 fliD) in the <63 fliD fraction was determined by using a Sedigraph. The >63 fliD fraction was dried and sieved to separate the >500 fliD fraction from sand (50(k)3 fliD). The results are shown as weight percentages (Figure 4).

NORWEGIAN JOURNAL OF GEOLOGY

X -ray diffraction (XRD)

Analyses were performed on both un-oriented bulk samples as well as oriented samples of the clay-sized fraction. All analyses were carried out on a Phillips PW17129 Xray diffractometer.

A normal scan between 2 and 50°29 (0.06° per step) and a slow scan between 26 and 28.S029 (0.01° per step) were recorded for each sample.

For the clay-size fraction, in addition to normal (2-45°29) and slow scans (24-26°29), analyses were performed on glycolated (2-35°29) and heat-treated samples. Heat treatment was performed in two stages. The samples were first heated for at least 2 hours at 350 °C and analysed between 2-30°29 and thereafter heated at 550 °C for at least 2 hours and analysed between 2-30°29.

Minerals were identified using manual interpretations and comparison (e.g. Brindley & Brown 1980; Moore & Reynolds 1997). Quantification was automated on a computer using a UNIX script specifically written for this purpose. The quantification is based on the total number of counts between peak half-heights under selected peaks after subtraction of the background level. Peaks from various scans were used in the quantification process and normalized to a sum of l 00% . The X-ray counts have not been converted to mineral percentages, as weighting factors for the site are unknown. The counts thus only reflect relative changes in mineralogical compositions. Relative mineral abundances suffice for the purpose of the present work. No attempt was therefore made to determine the absolute mineral percentages.

Total Carbon (TC), Total Organic Carbon (TOC) and carbonate analyses

The carbon and carbonate contents were found by analysing both untreated and acid leached samples of dried and ground sediments in a LECO 5344 furnace. The carbon content in the acid leached samples represents the total organic carbon (TOC). Total carbonate was calculated by using the form ula CaC03 = (TC-TOC) x 8.33

Results

Grain-size distribution Among the main aims of carrying out grain-size analyses, in addition to determining the general distribution of sand, silt and clay, was the identification of ice-rafted debris (IRD). The limiting size for a particle to be classified as IRD varies depending on the part of the basin from where the sediments are retrieved. In deepsea sediments from the Norwegian-Greenland Sea, particles >0.125 mm in diameter have been classified


s • Q. 8

� l:i Q ... l l

! l

Scoresby Sund Fan, East Greenland

� l

7

Figure 4. Grain-size distribution together with bulk and clay mineralogy data obtained by XRD analyses. Unit boundaries (from Jansen et al., 1996) as well as age estimates (Ma) (Channell et aL, 1999) for various horizons are shown. Grain-size data, including particles >0.5 mm, is plotted as weighto!b whereas the mineralogical data are presented as relative percent counts. (Butt et.aL)

8 f. A. Butt et al.

as IRD (He.prich et al. 1995). Whereas in sediment cores from the outer Scoresby Sund fjord system and shelf, a size of > 2 mm has previously been used as the lower limit for IRD (Nam et al. 1995; Dowdeswell et al. 1998). Si te 98 7 was drill ed at the base of the con tinental slope (Figures l and 2). For the present work, the term IRD is applied to particles >0.5 mm in diameter that are interpreted to have been transported by icebergs. However, as described in the following sections, all particles >0.5 mm in diameter do not represent IRD.

Unit V contains up to 70% sand and silt, but is devoid of any major >0.5 mm particle peaks - less than l% barring a few exceptions (Figure 4). Unit IV contains abundant >0.5 mm particles, up to 27%. Sand (avg. 30%) and silt (avg. 34%) form the bulk of sediments in this unit whereas the amount of clay is lower (avg. 30%) than in any other unit. Units IIIA and IIIB do not show any major differences in the grain-size record. In general the amounts of sand and >0.5 mm particles in Unit IIIA appear to be less than in Unit IIIB but the differences are not very pronounced. Unit Il again shows abundant sand (avg. 20%) and particles >0.5 mm (up to 5 l%). A considerable part of the record in this unit was not recovered. However, similar trends at the beginning and the end of the 'gap' suggest that the trend is continuous throughout the unit. This is confirmed by the uniform bulk density log trends in the unit (Jansen et al. 1996, p. 386-387). Unit I has on the average the lowest amount of sand (-5%). However scattered peaks of >0.5 mm particles (up to 27%) can be seen. The interval 120-180 mbsf shows more or less continuous presence of particles >0.5 mm and probably represents a single rafting phase.

Bulk Mineralogy The total clay mineral content reflects the grain-size changes associated with the unit boundaries with smaller amounts in the sand- and gravel-rich Units IV and Il and higher amounts in the other units (Figure 4). The amount of quartz remains more or less constant with slightly lower values in Units IV and IL Plagioclase is the dominant feldspar. However a considerable amount of K-feldspar is seen in Unit IV. Unit Il also shows higher K-feldspar content when compared with Units l, Ill and V. Carbonate is a significant mineralogical component only in the uppermost two units. The first major carbonate peak (over 80% carbonate) is seen in Unit IIIA (Figure 4). However, this peak is not seen in the carbonate record of the TC/TOC analyses (Figure 5) which otherwise is very similar to the XRD carbonate record. The high carbonate peak in Unit IIIA may thus be a measurement error and not a true value. The carbonate content in Units 1-11 is higher than that in Units III-IV. Carbonate content shows high values in the upper part of Unit L


Clay mineralogy Smectite is the dominant clay mineral in all units with the exception of Unit I. Clay in Unit Il appears almost exclusively to consist of smectite. Smectite continues to be significant in Unit I but kaolinite+illite become dominant. Illite and kaolinite show upward increasing trends in Unit V. A considerable increase in these two minerals is seen in Unit IV. The values fall in Unit Ill but show a slight upward increasing trend within the unit. There is sharp fall in the relative amounts of these minerals in Unit Il, in particular illite. Thereafter illite and kaolinite become significant in Unit I. Chlorite is present in low amounts in Units II-V though some isolated peaks can be seen. It registers a considerable increase in Unit I but still remains the !east abundant of the studied clay minerals.

Total Carbon/Total Organic Carbon The amount of TC (Figure 5) remains below l% in Units III-V barring a couple of exceptions. Most of this carbon is organic in nature. The amount of TC dramatically increases in Unit Il but declines again in Unit I. The values in Unit I are generally higher when compared to Units III-V particularly in the uppermost 50 m. Although organic carbon remains the dominant carbon for most of Unit l, there are several intervals where inorganic carbon is the major component. This is indicated by the carbonate peaks in Figure 5. In the few measurements available for Unit Il, inorganic carbon dominates in the lower part whereas organic carbon forms the bulk of TC in the upper part. Shipboard measurements of C/N ratios indicate that most of the organic matter in Units I and Il are of marine origin (Jansen et al. 1996).

Discussion

Depositional environments Jansen et al. (1996) describe sediments of Unit V as being highly indurated with many small-scale features such as flaser-bedding and discrete burrow traces. Thin clayey laminations are also common. These characteristics are indicative of low-energy environments. The grain-size record of the unit shows very low amounts of particles >0.5 mm and clasts > 1.0 cm (Figures 4 and 5). Slumps and turbidities have been identified in the unit (Jansen et al. 1996) but the overall nature of the unit represents a dominantly hemipelagic depositional environment.

Unit IV shows pronounced peaks of particles >0.5 mm in diameter (Figure 4). The unit consists of firm, compacted, heterogeneous sediments with numerous clasts and without any signs of bioturbation. The shipboard measurements of physical properties indicate an increase in bulk density, compressional velocity and magnetic susceptibility and a corresponding decrease in


o o

100

150

200

250

øoo

760

Qlz/K-føld 3 8 o

K·feldiPiag 1

Sllleltl 2 o 1{1 20

TC% o 1 2 3 o

Scoresby Sund Fan, East Greenland

TOC% 2 3 o 5 10 15 2D

- 0.78

1.07

1.95

2.14

- - - 3.58

6.23

6.89

6.67

- - - 7.43

9

Figure 5. Variations in relative percent count ratios of selected minerals together with the TC/TOC and carbonate analyses. Unit boundaries (from Jansen et al., 1996) as well as age estimates (Ma) (Channell et al., 1999) for various horizons are marked (Qtz=quartz, K-feld=K-feldspar, Plag=plagioclase, Sme=smectite, Ill=illite). (Butt et. al.)

porosity and natural gamma radiations compared to over- and underlying units (Jansen et al. 1996). The unsorted, compact nature of the sediments together with frequent occurrence of clasts, mainly < 1.0 cm, indicate a debris flow origin for the unit and therefore the >0.5 mm fraction cannot be classified as IRD. The high percentage of smectite among the clay minerals (Figure 4)

also favours downslope transport of sediments as a cohesive mass in the form of debris flows.

Unit Ill is characterized by a lowering in the sand content, reduction in particles >0.5 mm in diameter (Figure 4) and drop in the sedimentation rate (mostly 65-150 m/m.y. from over 200 m/m.y. in Unit IV, Channell et al. 1999). Jansen et al. (1996) report fewer clasts,

10 F. A. Buttet al

thin clayey, parallel laminations and low to moderate bioturbation throughout the unit. These characteristics indicate a return to quieter conditions comparable to those of Unit V.

No. of clasts o 10 20 30 40 50 60 70 80

100

160

200

250

300 Unit 11

350

l 400 Unit Ill

§. 460 CD c

500

550

600 Unit IV

650 Unit V

700

750

800

850

Figure 6. A plot showing the lithological composition of clasts (>l cm) recorded by the shipboard scientific party. The individual bars show clast data in 10 metre intervals (i.e., 0-10 m, 10-20 m and so on). The term 'unspecified' refers to the horizons where only the num ber of clasts was recorded without describing their lithology. Identified igneous rocks include granite, diorite, syenite, basalt and pegmatite; metamorphic include quartzite, metaquartzite, coal, and garnet schist, while sedimentary rocks include, siltstone, limestone, shale, sandstone, mudstone and dolomite. Unit boundaries are indicated (data from Jansen et al., 1996). (Butt et.al.)


Unit Il shows a high percentage of sand and particles >0.5 mm in size. Although the mineralogical composition is somewhat different, the grain-size distribution pattern including fraction >0.5 mm is similar to Unit IV. The physical property measurements by the shipboard party also confirm this similarity. In addition, a number of clasts (>l cm) (Figure 6) are reported (Jansen et al. 1996). This unit thus may also represent debris flows and hence the coarser particles (>0.5 mm) may not be classified as IRD.

Unit I is again of fine-grained nature and slight to moderate bioturbation is seen. However, several clasts and sandy turbidites have been recorded (Jansen et al. 1996) and a number of peaks of >0.5 mm particles can be seen (Figure 4). The dominant fine-grained nature of the sediments suggests that the overall environment was hemipelagic but interrupted by frequent episodes of icerafting and downslope transport.

Glacial-interglacial changes The occurrence of IRD in the marine record of the North Atlantic has been used as an indicator of icegrowth on continental masses in the area and their subsequent build-up to form large ice sheets with extensive marine margins (Jansen et al. 1996; Wolf-Welling et al. 1996; Dowdeswell et al. 1998).

Larsen et al. (1994) date the onset of major glaciations on East Greenland to 7 Ma based on data from ODP Site 918 off Southeast Greenland (located -63°N38"W). Unit V has its base at -7.5 Ma (Figure 3) ( Channell et al. 1999). However, as described in the preceding section, the unit is predominantly hemipelagic. Minor amounts of clasts (>l cm) (Figure 6) together with low percentage of particles >0.5 mm (Figure 4) suggest absence of major glacial input. There are however indications of cold conditions. In addition to the scattered dropstones reported in this unit by the shipboard party, some dropstones at Site 9 18 were identified to represent rock formations in central East and Northeast Greenland (Larsen et al. 1994). This indicates the presence of glaciers (icebergs) on central East Greenland at this time. From the hemipelagic nature of the sediments, it is likely though that they were restricted in extent.

Unit IV is identified to be comprised of debris flows. The presence of large number of de bris flows in glacierfed submarine sediment deposits has previously been interpreted to result from glaciers reaching the shelf break and delivering large quantities of sediments (Dowdeswell et al. 1996; King et al. 1996; Solheim et al. 1998; Vorren et al. 1998). Assuming a similar relationship at Site 987, Unit IV can be interpreted to represent a glacial advance across the central East Greenland Shelf. The base of Unit IV is dated at -5 Ma (Channell et al. 1999) and this advance may thus indicate response to the initial cooling phase in the Northern Hemisphere dated at 5.5 Ma (Jansen & Sjøholm 1991).

NORWEGIAN jOURNAl OF GEOLOGY

Unit Ill marks a return of dominantly hemipelagic environments similar to those during deposition of Unit V but with a higher average rate of sedimentation. A slightly modified mineralogy and presence of more and varied clasts, indicate a more extensive and varied drainage basin than Unit V.

Unit Il is another debris flow unit with base dated to 2.58 Ma (Channell et al. 1999). Once again assuming the debris flows to represent a glacial advance, the deposition of this unit can be linked to the onset of major Northern Hemisphere glaciations at 2.6 Ma reported from the Norwegian-Greenland Sea and the North Atlantic (Shackleton et al. 1984; Jansen et al. 1988; Jansen & Sjøholm 1991). Data for the most part of this unit are missing owing to poor sample recovery but similar tren ds at the base and top of the unit together with uniform log trends (Jansen et al. 1996) indicate that the unit represents a single phase likely to be coincident with a major cooling trend in the Northern Hemisphere.

Unit I indicates alternating warmer and colder environments where periods of hemipelagic sedimentation are interrupted by deposition of IRD. The interval 120-180 mbsf, dated to slightly older than 1.0 Ma (Figure 4) apparently represents a rafting phase since it shows a continuously high >0.5 mm particle (IRD in this case) content. A few other IRD peaks can also be seen but owing to their isolated nature, they more likely represent 'events:

Oceanic conditions Carbonate in the Norwegian-Greenland Sea is generally taken as an indicator of water-mass conditions. High carbonate contents are generally interpreted to reflect the influence of warm waters, while low values are associated with cold surface waters and increasing dilution by terrigenous material (Kellogg 1980; Henrich 1989; Henrich & Baumann 1994).

The carbonate content remains negligible throughout Units V-III (Figure 5). These units are also barren of fossils (Jansen et al. 1996) indicating conditions unfavourable for biogenic productivity. However, low carbonate content in sediments may also be due to dissolution which is regulated by water mass characteristics (e.g. temperature, salinity, pressure, degree of seawater saturation with respect to calcite etc.) (Broecker & Takahashi 1978) as well as other factors such as increased supply of organic matter to the seafloor. The organic matter, when oxidized, leads to increased carbonate dissolution (Berger et al. 1982). Scattered occurrence of IRD in Units V and Ill, nevertheless, indicates glacial presence on continental Greenland and it is inferred that the low carbonate in Units V-III is primarily indicative of cold water conditions.

There is a sharp increase in carbonate content in Units I and Il (Figure 5). Common occurrences of planktonic foraminifera (N. pachyderma sin.) have been

Scoresby Sund Fan, East Greenland 1 1

reported in this part of the Site 987 succession (Jansen et al. 1996). The presence of planktonic foraminifera indicates periods of increased surface-water productivity.

Huber et al. (2000), based on ultrastructural investigations on the tests of N. pachyderma sin. from Site 987 covering the last l m.y., report a high degree of dissolution of the tests in the entire investigated section. They have identified three dissolution peaks at O.l, 0.4 and 0.6 Ma, which they relate to three very strong, short-term meltwater events that led to interruption of the deepwater formation (i.e. reduced bottom-water conditions) in the Norwegian-Greenland Sea.

It is thus likely that the oceanic regime along the East Greenland Margin has undergone significant changes through time, especially since 2.6 Ma.

Site 987 and Plio-Pieistocene climatic variability The last 2.6 m.y. have been a period of climatic fluctuations in the Norwegian-Greenland Sea with interglacialglacial cycles alternating with 40/100 kyr cyclicity (Jansen et al. 1988; Ruddiman & Raymo 1988; Berger & Jansen 1994). Units I and Il to a certain extent reflect this variability through the intermittent rafting phases in predominantly hemipelagic conditions and the high variability in clay mineralogy and carbonate content after an initial phase of glacial expansion (Figures 4 and 5). There is, however, o ne aspect to be considered. Funder et al. ( 1994) have described five glacial periods in East Greenland since 250 ka. The upper part of the Site 987 succession though shows isolated sand peaks, does not contain signals such as pronounced IRD peaks and/or mineralogical shifts that would suggest such frequent and noticeable changes in climate. A possible explanation could be the sampling resolution {50 cm), which may be too coarse to capture these well-documented fluctuations. However, this is not very likely as earlier glacial advances have been recorded at the same resolution. A corollary of this may be that the earlier glaciations were much more extensive than the Plio-Pleistocene ones. The limited nature of Plio-Pleistocene glaciations in central East Greenland have earlier been proposed by Funder et al. ( 1994) and Dowdeswell et al. {1997). Glaciers may have covered coastal areas producing icebergs but without any major thrust of the inland ice across the shelf. The ice advances that did extend across the shelf during this time were shortlived events (Funder et al. 1994) and may have been missed at the current sampling resolution.

The scenario of limited glacial expansions during Plio-Pleistocene is in sharp contrast to the SvalbardBarents Sea Margin, which contains evidence of numerous ice-advances across the shelf during this time (Solheim et al. 1996; Solheim et al. 1998). This contrast in glacial histories could be related to different ice dynamics at the two margins. Fast-flowing ice streams are believed to be the mechanism supplying sediments to the shelf/upper slope during Plio-Pleistocene glacial

12 F. A. Butt et al.

periods at the Svalbard-Barents Sea Margin (Solheim et al. 1996, 1998; Dowdeswell et al. 1998). This would suggest a glacial regime with basal melting. The ice-sheet margin in the present-day Scoresby Sund region is restricted to the higher elevations of the coastal regions with a few glacial outlets terminating at the sea level as calving fronts at fjord heads. The areas of basal melting are limited (Funder et al. 1998).

The oceanographic conditions along the two margins play an important part in regulating the climates at the respective margins. East Greenland is influenced by the southward flowing cold East Greenland Current whereas the northward flowing warm Atlantic Water affects the Svalbard-Barents Sea Margin (Figure 1). The result is increased moisture transport to the higher latitudes along the Svalbard-Barents Sea Margin. Consequently, the Svalbard-Barents Sea Margin is interpreted to have been both warmer and more moist compared to the East Greenland Margin even during full glacial conditions (Hebbeln et al. 1998).

Another relevant factor that should be kept in mind regarding the somewhat subdued representation of the Plio-Pleistocene glaciations in the Site 987 record is the shift in the depocentre proposed by Dowdeswell et al. ( 1997). According to them the south em flank of the Scoresby Sund Fan was the most active depositional area during Plio-Pleistocene due to the presence of a crossshelf trough that acts as a conduit for sediments and secondly the southward flowing East Greenland Current would direct any calving icebergs in a southerly direction. Site 987 was drilled on the northern flank (Figure 2) and thus may not contain the Plio-Pleistocene record in its entirety.

o km 25

w


Correlation of seismic stratigraphy and Site 987 data Two key seismic stratigraphic studies have been carried out on the Scoresby Sund Fan. These include:

l. Geological Survey of Greenland (GGU) line 82-12 (Larsen 1990) on the north-eastern part (Figures 2 and 7).

2. Alfred Wegener Institute of Marine and Polar Research (AWI) line 90600 (Vanneste 1995; Vanneste et al. 1995) on the southern flank of the fan (Figures 2 and 8).

Larsen (1990) identified 12 seismic sequences along line GGU 82-12 where the deposits reach a thickness of over 3 km on the outer shelf (Figure 7). The earliest sedimentation was dated at Eocene-Early Oligocene (sequences 1-2) and the major sedimentation phase (sequences 3-8) to Miocene. The uppermost units 9-12 were interpreted to represent Plio-Pleistocene deposition. Site 987 was drilled on the eastern extremity of GGU 82-12 line (Figure 2). The age model for Site 987 suggests that although sedimentation may have started during Late Miocene times, the section is dominated by Plio-Pleistocene deposits (Figure 3) (Channell et al. 1999). In Figure 7, Sequence 9 represents a major progradation of the Liverpool Land shelf accompanied by erosion of the underlying units. The stratigraphic position of the lower boundary of this sequence at Site 987 (slightly above the halfway mark) makes it possible to correlate it to the base of Unit Il (at -375 mbsf out of a total depth of -850 mbsf). The deposition of the sequence can thus be related to the major Northern Hemisphere cooling at 2.6 Ma.

E

lnner Liverpool Land Bas in

Outer Liverpool Land Bas in

12 Site 987

2

� 4 • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • Basement • • • • • • • •

.. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. ... .. .. .. .. .. .. .. .. .. .. .. .. .. ... .. .. .. .. .. .. .. .. 5 .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. ..

.. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 6 .. .. . . .. .. .. .. .. .. .. .. .. . .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. ..

Figure 7. Line drawing of the seismic profile across line GG U 82-12. The interpreted seismic sequences ( 1-12) by Larsen (1990) are shown. See Figure 2 for Iocation (modified from Larsen, 1990). (Butt et. al.)

NORWEGIAN JOURNAL OF GEOlOG Y

NW


SE

Continental Slope

I Unit Ill Glacial

lceland Plateau Edge

---�--�� -2 Unit li }

Preglaclal � Unit l

3 V.E. = 10 o km 20

Figure 8. Line drawing of the interpreted seismic sequence along line 90600. See Figure 2 for location ( modified from Vannes te et al. , 1995 ). (Butt et.al. )

Sequence 11 apparently represents a retrogradational episode followed by another progradational phase, Sequence 12 (Figure 7). However, this cannot be seen in the sedimentary record of Site 987 at its present sampling resolution.

Sedimentary deposit along line 90600 (Figure 8), which reaches a thickness of about 1000 m on the outer shelf and upper slope, was divided into three units. The lower two units (Unit I and Unit Il) were interpreted to be preglacial and the topmost Unit Ill to be of glacial origin (Vanneste et al. 1995). Unit Ill was further subdivided into three sub-units and each sub-unit was related to the main evolution stages in the Plio-Pleistocene glacia! development of the North Atlantic as proposed by Jansen et al. (1988). A shelf progradation by �45 km accompanied by minor aggradation was estimated during the deposition of Unit Ill.

Correlation of line 90600 to the sedimentary record of Site 987 is difficult as the southern and northern flanks of the fan exhibit distinctly different architecture (Dowdeswell et al. 1997). Besides, due to lack of dating along line 90600, any correlation will at best be speculative. Data from the present work as well as Larsen et al. (1994) indicate glacial advances across the shelf at much earlier times than 2.6 Ma proposed by Vanneste et al. (1995). The base of Unit Ill could therefore be older than the Late P liocene as was assumed by Vanneste et al. (1995).

Sediment sources Looking at the present-day scenario along the East Greenland Margin, sediments in the Scoresby Sund Fan can be sourced from: a) drainage of the Scoresby Sund area lying to the west (Figure 2) and, b) northern parts of the margin through the East Greenland Current (Figure l). In the present work, the high smectite content is interpreted to represent influence from volcanic sources whereas increasing amounts of K-feldspar (the dominant feldspar in most sedimentary rocks) together with kaolinite and illite are taken to reflect contribution from sedi-

mentary sources. Composition of clasts (>l cm) as reported by Jansen et al. (1996) (Figure 6) is used in a similar manner. However, the composition of sediments and clasts by itself cannot be used to make concrete suggestions regarding provenance as the rock types in the immediate vicinity of Site 987 are also found in NE Greenland and their input via the East Greenland Current cannot be ruled out. Mineralogy of sediments and clasts is therefore used in conjunction with earlier work by Larsen (1990) and Vanneste et al. (1995).

Unit V displays subtle yet consistent upward trends in mineralogy though the principal relationships among the various minerals remain the same i.e. plagioclase remains the dominant feldspar and smectite remains the major clay (Figure 4). Total clay and quartz show an overall upward decreasing trend while the opposite is true for the other minerals. Based on geometry of the seismic sequences along the southern flank (Figure 7), Vanneste et al. (1995), propose that the oldest sediments were derived from erosion of a volcanic plateau that probably forms an extension of the Iceland Plateau located further to the southeast. The consistently high smectite in Unit V may be used to suggest a similar origin for the Unit V sediments. The area was tectonically active during the Tertiary with seafloor spreading along the initially subaerial Kolbeinsey Ridge starting in the Early Miocene (Larsen 1990). The Kolbeinsey Ridge, which became submarine in the late Middle Miocene/early Late Miocene (Larsen 1990), together with the Iceland Plateau further south constitutes a potential source for the Late Miocene Unit V sediments. The upward increasing trend in other minerals suggests that the source area was being progressively modified ( expanded?) possibly including terrestrial regions of East Greenland.

There is a drastic change in relative XRD percentages of minerals across the upper boundary with Unit IV but there is no change in the character of the trends (Figure 4). The trends in Unit IV are more or less the same as in Unit V but the magnitude of shifts is much larger. However the overall grain-size is also much coarser in this unit and the changes in mineralogy could be a function of

14 F. A. Butt et al.

change in grain-size (Henning Dypvik, pers. comm. 2000). In order to overcome the grain-size effect (even if partly), the ratios between the relative percent age counts of the different minerals were calculated (Figure 5). The ratios indicate that K-feldspar does indeed increase both with respect to quartz and plagioclase. This, together with increasing trends in the illite and kaolinite relative counts, indicates input from a sedimentary source. This is supported by the occurrence of sedimentary clasts (>l cm) in the interval (Figure 6). In the case of debris flow units, the sedimentary clasts are more likely to indicate input from Jameson Land compared to NE Greenland as the debris flow units represent sediments transported downslope from the shelf.

Bulk mineralogy and grain-size distribution in Units Ill and V are essentially similar (Figure 4) and this may indicate similar depositional regimes for the two units. The upward tren ds in clay minerals in Unit Ill (Figures 4 and 5), however, suggest that Unit Ill sediments were derived from a varied source. The difference in source region is also indicated by numerous clasts (>l cm) of sedimentary origin in Unit Ill as opposed to Unit V (Figure 6).

The most noticeable and highly fluctuating mineralogical changes, particularly in clay minerals, occur in Unit I. The greatly lowered values of smectite with respect to illite and kaolinite are taken to indicate the dominant role of sedimentary areas as source regions. In situ alteration of smectite is improbable in view of the shallow depth of Unit I and thermal gradient (95 °C/km) measured at the site (Jansen et al. 1996). A diverse source area is also indicated by the clasts (>l cm) representing igneous, metamorphic and sedimentary rock types. Vanneste et al. (1995) proposed a drastic increase in sediment input from Scoresby Sund region during later phases of fan development along the southern flank of the Scoresby Sund Fan. Keeping this in view, the sedimentary signal in Unit I may be related to increased sediment sup p ly from Jameson Land.

Summary and conclusions The sedimentary succession at Site 987 is seen to consist of three dominantly hemipelagic phases with varying input from ice-rafting (Units l, Ill and V) and two debris flow units (Units Il and IV).

The existence of ice on continental Greenland is indicated since -7.5 Ma (Unit V) but environments at this time were dominantly hemipelagic. Sediments in this lowermost unit were chiefly derived from a volcanic source, probably the Iceland Plateau and/or the KolbeinseyRidge.

The first major glacial advance across the Scoresby Sund took place at ca. 5 Ma (Unit IV). Sediment input from Jameson Land in addition to volcanic sources is indicated by the presence of sedimentary clasts.

NORWEGIAN jOURNAL OF GEOLOG Y

A return to dominantly hemipelagic environment took place around 4.5 Ma (Unit Ill). However, a more extensive and varied drainage basin than the earlier hemipelagic p hase (Unit V) is inferred.

Another major glacial advance across the shelf at 2.58 Ma is represented by another debris flow unit (Unit Il). This phase is related to the onset of major Northern Hemisphere glaciations at 2.6 Ma.

The last ca. 2 m. y. are characterized by al terna ting warmer and colder environments where periods of hemipelagic sedimentation have been interrupted by deposition ofiRD (Unit I). Based on evidence from Vanneste et al. (1995) and abundant sedimentary clasts, Jameson Land is the likely dominant sediment source.

Si te 987 data also indicate that the Plio-P leistocene glaciations in the central East Greenland region were more limited in extent than the earlier glacials. In contrast, several ice advances, extending all the way to the shelf break, have been identified on the Svalbard-Barents Sea Margin during the same period. Differences in gladal thermal regimes are suggested to be a possible reason for this con trast. Another factor that is relevant is the proposed shift in the depocentre to the southern flank of the Scoresby Sund Fan. As it is located on the northern flank, Site 987 may not contain the P leistocene record in its entirety.

Acknowledgements. This work was carried out as part of a doctoral research grant provided by the Research Council of Norway (NFR). Laboratory work was partly funded by the Norwegian Polar Institute. Dr Svend Funder, Dr Julian Dowdeswell and Dr Jochen Knies are gratefully acknowledged for their critical reviews which helped to improve the manuscript. Staff of the chemistry and sedimentology laboratories at the Department of Geology, University of Oslo, particularly, Berit Løkenberg and Mufak Naoroz deserve a special thanks for their help. Adil Manzoor, Randi Hendrickson and Marianne Jacobsen are thanked for their assistance in carrying out the analyses.

References Berger, W. H., Bonneau, M. -C. & Parker, F. L. 1982: Foraminifera on

the deep-sea floor: lysodine and dissolution rate. Oceanol. Acta 5, 249-258.

Berger, W. H. & Jansen, E. 1994: Mid-Pleistocene Climate Shift - The Nansen Connection. The Polar Oceans and Their Role in shaping the Climate. Geophysical Monograph 85, 295-311.

Brindley, G. W. & Brown, G. 1980: Crystal Structures of clay minerals and their x-ray identification. Mineralogical Society London, 495 p.

Broecker, W. S. & Takahashi, T. 1978: The relationship between lysocline depth and in situ carbonate ion concentration. Deep Sea Res. 25, 65-95.

Channell, J. E. T., Smelror, M., Jansen, E., Higgins, S., Lehman, B., Eidvin, T., & Solheim, A. 1999: Age models for glacial fan deposits off East Greenland and Svalbard (ODP Site 986 and Site 987). In Rayrno, M. E., Jansen, E., Blum, P. & Herbert, T. D. (eds.): Proceedings of the Ocean Drilling Program, Scientific Results 162, 149-166.

College Station, Texas. Dowdeswell, J, A., Elverhøi, A. & Spielhagen, R. 1998: Glacimarine

sedirnentary processes and facies on the Polar North Atlantic margins. Quaternary Science Reviews 17,243-272.


Dowedeswell, J. A., Kenyon, N. H., Elverhøi, A., Laberg, J. S., Hollender, F. -J., Mienert, J. & Siegert, M. J. 1996: Large-scale sedirnentation on the glacier-influenced Polar North Adantic margins: longrange side-scan sonar evidence. Geophysical Research Letters 23, 3535-3538.

Dowdeswell, J. A., Kenyon, N. H. & Laberg, J. S. 1997: The glacierinfluenced Scoresby Sund Fan, East Greenland continental margin: evidence from GLORIA and 3.5 kHz records. Marine Geology 143, 207-221.

Elverhøi, A., Dowdeswell, J. A., Funder, S., Mangerud, J. & Stein, R. 1998: Glacial and Oceanic History of the Polar North Adantic Margins: An overview. Quaternary Science Reviews 17, 1-10.

Faleide, J. l., Solheim, A., Fiedler, A., Hjelstuen, B. 0., Andersen, E. S. & Vann este, K., 1996: Late Cenowic evolution of the western Barents Sea-Svalbard continental margin. Global and Planetary Change 12

(1-4), 53--74.

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 Green/and. Geology of Canada vol. l, 743-792. The Geological Survey of Canada.

Funder, S., Hjort, C. & Landvik, J. Y. 1994: The last glacial cycles in East Greenland: an overview. Boreas 23 ( 4), 283-293.

Funder, S., Hjort, C., Landvik, J. Y., Nam, S. -1., Reeh, N. & Stein, R. 1998: History of a stable ice margin - East Greenland during the middle and upper Pleistocene. Quaternary Science Reviews 17,

77-123.

Funder, S. & Larsen, H. C. 1989: Quaternary Geology of the shelves adjacent to Greenland. In Fulton, R. J. (ed): Quaternary Geology of Canada and Green/and. The Geology of North America vol. K-1,

769-772. Geological Society of America. Hebbeln, D., Henrich, R. & Baumann, K. -H. 1998: Palaeoceanography

of the last interglacial/glacial cycle in the Polar North Adantic. Quaternary Science Reviews 17,125--153.

Henrich, R. 1989: Glacial/interglacial cycles in the Norwegian Sea: Sedimentology, paleoceanography, and evolution of late Pliocene to Quaternary northern hemisphere climate. In Eldholm, 0., Thiede, J. & Taylor, E. (eds.): Proceedings of the Ocean Drilling Program, Scientific Results 104, 189-232. College Station, Texas

Henrich, R. & Baumann, K. -H. 1994: Evolution of the Norwegian Current and the Scandinavian lee Sheet during the past 2.6 m.y.: evidence from ODP Leg 104 biogenic carbonate and terrigenous records. Paleogeography, Paleoclimatology, Paleoecology l 08, 75-94.

Henrich, R., Wagner, T., Goldschmidt, P. & Michels, K. 1995: Depositional regimes in the Norwegian-Greenland Sea: the last two glacial to interglacial transitions. Geologische Rundschau 84, 28-48.

Huber, R., Meggers, H., Baumann, K. -H. & Henrich, R. 2000: Recent and Pleistocene carbonate dissolution in sediments of the Norwegian-Greenland Sea. Marine Geology 165, 123--136.

Jansen, E., Blei!, U., Henrich, R., Kringstad, L. & Slettemark, B. 1988:

Paleoenvironmental changes in the Norwegian Sea and the northeast Adantic during the last 2.8 m.y.: Deep Sea Drilling Project/Ocean Drilling Program sites 610,642,643 and 644. Paleo-ceanography 3, 563--581.

·

Jansen, E., Raymo, M. E., Blum, P. & the Leg 162 Shipboard Scientific Party 1996: Proceedings of the Ocean Drilling Program, Initial Reports 162, College Station Texas, 1182 p.

Jansen, E. & Sjøholm, J. 1991: Reconstruction of glaciation over the past 6 Ma based on Norwegian Sea records. Nature 349, 600-603.

Kellogg, T. B. 1980: Paleoclimatology and paleo-oceanography of the Norwegian and Greenland seas: glacial-interglacial contrasts. Boreas 7, 61-73.

King, E. L., Sejrup, H. P., Haflidason, H., Elverhøi, A. & Aarseth, l. 1996: Quaternary seismic stratigraphy of the North Sea Fan: glacially-fed gravity flow aprons, hemipelagic sediments and large submarine slides. Marine Geology 130, 293-316.


Larsen, H. C. 1990: The East Greenland Shelf. In Grantz, A., Johnson, L. & Sweeney, J. F. (eds.): The Arctic Ocean Region. The Geology of North America vol. L, 185--210. Geological Society of America.

Larsen, H. C., Saunders, A. D., Clift, P. D., Beget, J., Wei, W., Spezzaferri, S. & ODP Leg 151 Scientific Party 1994: Seven million years of glaciation in Greenland. Science 264, 952-955.

Mienert, J., Hollender, F. -J. & Kenyon, N. H. 1995: GWRIA survey of the East Greenland margin: 70°N to 80°N. In Crane, K. & Solheim, A. (eds.): Seafloor Atlas of the Northern Norwegian-Green/and Sea. Norsk Polarinstitutt. Meddelelser 137, 150--151.

Mienert, J., Kenyon, N. H., Thiede, J., Hollender, F. -J. 1993: Polar continental margins: studies off East Greenland. EOS Transactions of the American Geophysics. Union 7 4, 224, 234, 236.

Moore, D. M. & Reynolds Jr., R. C. 1997: X-ray diffraction and the identification and analysis of clay minerals. Oxford University Press, Oxford, 378 p.

Nam, S. -l., Stein, R., Grobe, H. & Hubberten, H. 1995: Late Quaternary glacial-interglacial changes in sediment composition at the East Greenland continental margin and their paleoceanographic implications. Marine Geology 122, 243-262.

Poulsen, N. E., Manum, S. B., Williams, G. L. & Ellegaard, M., 1996:

Tertiary dinoflagellate biostratigraphy of Sites 907, 908, and 909 in Norwegian-Greenland Sea. In Thiede, J., Myhre, A.M., Firth, J.V., Johnson, G.L. & Ruddiman, W.F. (eds.): Proceedings of the Ocean Drilling Program, Scientific Results 151, 255--287. College Station, Texas.

Ruddiman, W. F. & Raymo, M. E. 1988: Northern hemisphere climate regimes during the past 3 Ma: possible tectonic connections. Philosophical Transactions of the Royal Society of London B 318, 411--430.

Shackleton, N. J., Backman, J., Zimmerman, H., Kent, D. V., Hall, M. A., Robersts, D. G., Schnitker, D., Baldauf, J. G., Despairies, A., Homrighausen, R., Huddlestun, P., Keene, J. B., Katenback, A. J., Krumsieck, K. A. D., Morton, A. C., Murray, J. W. & WestebergSmith; J. 1984: Oxygen isotope calibration of the onset of ice rafting and history of glaciation in the North Adantic region. Nature 307, 620-623.

Solheim, A., Andersen, E. S., Elverhøi, A. and Fiedler, A. 1996: Late Cenowic depositional history of the western Svalbard continental shelf, controlled by subsidence and climate. Global and Planetary Change 12 (1-4), 135--148.

Solheim, A., Faleide, J. l., Andersen, E. S., Elverhøi, A., Forsberg, C. F., Vanneste, K., Uenzelmann-Neben, G. & Channell, J. E. T. 1998:

Late Cenozoic seismic stratigraphy and glacial geological development of the East Greenland and Svalbard-Barents Sea continental margins. Quaternary Science Reviews 17, 155--184.

Vanneste, K. 1995: A comparative seismic stratigraphy study of major Plio-Pleistocene glaciogenic depocentres along the polar North Adantic margins. Unpublished Ph.D thesis, University of Gent, 291

p. Vanneste, K., Uenzelmann-Neben, G. & Miller, H. 1995: Seismic

evidence for long-term history of glaciation on central East Greenland shelf south of Scoresby Sund. Geo-Marine Letters 15, 63--70.

Vorren, T. 0., Blaume, F., Dowdeswell, J. A., Laberg, J. S., Mienert, J., Rumohr, J. & Werner, F. 1998: The Norwegian-Greenland Sea continental margins: morphology and late Quaternary sedimentary processes and environments. Quaternary Science Reviews 17,

273-302.

Wolf-Welling, T. C. W., Cremer, M., O'Connell, S., Winkler, A. & Thiede, J. 1996: Cenowic Arctic Gateway paleoclimatic variability: indications from changes in coarse-fraction composition. In Thiede, J., Myhre, A. M., Firth, J., Johnson, G. L. & Ruddiman, W. F ( eds.) Proceedings of the Ocean Drilling Program, Scientific Results 151,515--568. College Station, Texas.

Evolution of the Scoresby Sund Fan, central East Green ...

Documents

Transcript of Evolution of the Scoresby Sund Fan, central East Green ...