Coastal morpho-stratigraphy and Holocene relative sea level changes at Tuapaat, southeastern Disko Island, central West Greenland MORTEN RASCH and NIELS NIELSEN Rasch. M . & Nielsen, N . 1995: Coastal morpho-stratigraphy and Holocene relative sea level changes at Tuapaat, southeastern Disko Island, central West Greenland. Polar Research 1 4 ( 3 ) , 277-289. The geomorphology of a coarse clastic coastal landscape at Tuapaat (69"24'N 52"36'W), southeastern Disko Island, central West Greenland, is described, and a coastal morpho-stratigraphy of the landscape is constructed. ''C ages o n marine shells. whale bones, peat and gyttja are used to construct relative sea level changes throughout the Holocene. The emergence of SE Disko Island occurred in the early part of the Holocene. The Holocene marine limit is situated ca. 80m a.s.1. Between 4.7 and 1.0 ka BP. the relatlve sea level approached the present sea level and it h d S probably been below present sea level between 4.7 ka BP and the present. The morpho-stratigraphy in the lowest part of the coastal landscape at Tuapaat suggests a complex late Holocene relative sea level history which includes at least 3-4 transgressions during the last ca. 2.5 ka. Morten Rasch, Arctic Station. Unioersity of Copenhagen, DK-3953 Godhauti. Greenland; Niels Nielsen, Institute of Geography, University of Copenhagen. (aster Voldgade 10. DK-1350 Copenhagen K , Denmark. Introduction Present knowledge about late Holocene relative sea level (RSL) changes in West Greenland is very limited. The glacio-isostatic rebound fol- lowing the WisconsinanfWeichselian glaciation has been documented by several emergence curves representing local regions (Weidick 1972b; Kelly 1985; Funder 1989). From these curves it is well established that the RSL approached the present sea level between ca. 3 and 4 ka BP. Only scattered information is available today gbout the RSL changes that occurred in the period from 3- 4 ka BP to the beginning of the present century, when tidal observations were initiated. However, it is well established that the RSL in some parts of West Greenland has been below the present sea level during the last ca. 3 ka, and that trans- gression(s) have occurred within the present mil- lennia (Kelly 1980, 1985; Funder 1989; Weidick 1993; Rasch & Nielsen 1994; Rasch et al. in press). The main objective of this study is to contribute to a greater understanding of late Holocene RSL changes in West Greenland. Special attention is given to RSL changes within the last millennia. Both traditional and new methods have been applied in the investigation. An age-altitude dia- gram based on Holocene dates from southeastern Disko Island has been constructed. and the chronological sequence of coastal landforms (the coastal morpho-stratigraphy) in a coarse clastic coastal landscape at Tuapaat (69"24'N 52"36'W) on the southern coast of Disko Island has been studied. In coarse clastic coastal landscapes, results of RSL changes are easily observed because regressions lead t o repeated simple beach ridge (swash ridge) formations, while trans- gressions lead to the formation of either washover ridges (transgressive barriers) or coastal cliffs (Carter & Orford 1984; Tanner 1988; Rasch & Nielsen 1994). As a result. individual landforms in a coarse clastic coastal landscape show evidence of the course of the RSL at the time of their origin, and the entire landscape represents a chronologically arranged archive of former RSL changes. Physical setting Disko Island is situated in Disko Bugt, central West Greenland, between 69'15" and 70"20'N and between 51"50'W and 55"OO'W (Fig. 1). The climate is polar maritime. The mean annual tem- perature in the town, Qeqertarsuaq, southeastern Disko Island, is about -4°C and the mean annual precipitation varies between 100 and 500 mm (Danish Meteorological Institute 1962-1985). The island lies within the zone of continuous 278 M . Rasch & N . Nielsen S t r a i t Flg. 1 . Disko Island. Old Danish names of towns are 1 given in brackets. permafrost, and at present approximately 20% of Disko Island is glacierised (Humlum 1988). The wave climate on southern Disko Island is characterised by wind waves from the east caused by katabatic winds blowing off the Inland Ice and by swell from the southwest and west caused by cyclones i n Baffin Bay. The net littoral drift is dominated by swell impact and is directed towards the east. The sea surrounding Disko Island is ice- covered from mid-January to mid-April, and the shores of Disko Island are affected by ice-foot formation from December to the end of June (Nielsen 1979). The bedrock geology of southeastern Disko Island consists of unlithified Cretaceous sand- stones overlaid by Tertiary basalt (Henderson et al. 1976). and the landscape is characterised by Tertiary plateaus that have been heavily dissected by glaciers during the Quaternary. The Tertiary plateaus are situated around 1OOOm a.s.l., and today plateau surfaces are almost completely cov- ered by ice caps. In the lowermost part of the terrain, i n the unlithified sandstones, gullies have blurred the glacial origin of the landscape. The border between Cretaceous sandstones and Ter- tiary basalts is situated ca. 400 m a.s.1. at Tuapaat. During the Wisconsinan/Weichselian glacia- tion, only a small part of Disko Island was covered by Inland Ice (Frich & Ingolfsson 1990; Ingolfsson et al. 1990). The margin of Inland Ice probably lay a few kilometres north of the present shoreline on southern Disko Island, and an extensive ice- shelf possibly occurred west of Disko Island (Bennike 1994). The Wisconsinan/Weichselian extent of the local glaciers on the island is almost unknown. It is, however, known that large valley glaciers occurred in the fiords on the western coast of the island (Donner 1978; Ingolfsson et al. 1990). The occurrence of glaciers in the fiords on western Disko Island and the altitude of the Holocene marine limit o n the island (60-90 m a.s.1.) suggest extensive local glaciation. During the retreat of the Inland Ice ca. 10 ka BP, the RSL rose to the Holocene marine limit. Generally, the Holocene marine limit on Disko Island increases towards the southeast from ca. 6 0 m a.s.1. o n northwestern Disko Island to ca. 90 rn a.s.1. on southeastern Disko Island (Ingolfsson et al. 1990). At Tuapaat, t h e Hol- ocene marine limit is situated at 80 m a.s.1. Between 10 ka BP and 3 ka BP, the RSL fell continuously from the Holocene marine limit to Coastal morpho-stratigraphy and Holocene relative sea level changes 279 the present sea level (Donner & Jungner 1975; Frich & Ing6lfsson 1990; Ingolfsson et al. 1990; Bennike et al. 1994). During this period coarse clastic beach ridge plains developed along the shores of Disko Island (Rasch & Nielsen 1994). During the last >4 ka, the RSL in Disko Bugt has been very close to or below the present sea level (Donner & Jungner 1975; Frich & Ingdlfsson 1990; Ing6lfsson et al. 1990). As a result, subaerial deposits containing information about marine events in this period are sparse and confined to the terrain immediately above sea level. The scarcity of information concerning marine events, including RSL changes. during the last 3-4 ka is illustrated by the very low number of radiocarbon dates on marine fossils from this period (Fig. 2). Study area - geomorphology The research site for the field investigation is a small cuspate foreland near the mouth of the Tuapaat River on the southern coast of Disko Island about 40 kilometres east of the town Qeqertarsuaq/Godhavn (Fig. 1). The interior of the foreland is mainly built up of simple beach ridges (swash ridges) consisting of rounded to well-rounded pebbles and cobbles (Tuapaat means well-rounded stones in Inuit-language). The pattern of the beach ridges indicate that they have been formed during littoral drift from the west (Figs. 3 , 4 and 5 ) . The mean elevation of the foreland is about 3 m a.s.1. A washover ridge (transgressive barrier) separates the interior of the foreland from the sea. The washover ridge is significantly higher and wider than the simple beach ridges in the interior part of the foreland (Fig. 6). Towards the west, the interior simple beach ridges of the cuspate foreland are superimposed by an alluvial fan (Figs. 5 and 7). Landward, the foreland is bounded by a fossil coastal cliff which leads to marine terraces at higher elevations. The front of the cliff is dissected by two larger and several smaller gullies. These gullies have pro- duced the yellow sandy sediment for the alluvial fan that covers the coastal landscape in the west- ern part of the foreland. The landscape immediately above the cuspate foreland consists of gently sloping and vegetated marine terraces with scattered exposures of coarse clastic beach sediment. The landscape has been extensively adjusted by solifluction, and it is therefore difficult to discern the detailed geo- morphology of individual beach ridges even in the exposures in the vegetation. The marine terraces continue to the Holocene marine limit at 80 m a.s.1. Methods An age-altitude diagram based on 33 Holocene dates o n terrestrial and marine material from the southeastern part of Disko Island and the islands of Kitsissut (south of Disko Island) has been constructed (Fig. 7). The Holocene marine limit is consistently situated ca. 80-90 m a.s.1. in the area represented by the diagram (Funder 1989; *I 15 1 0 1 2 3 4 5 6 9 10 1 1 Time (ka BPI Fig. 2. Frequency distrlbution of radiocarbon dated shells and whale bones from Disko Bugt. The total number of I4C dates is 87. Only four dates exist from the period after 4 ka BP Sources for dates: Weidick (1968, 1972a. 1972b, 1973,1974), Donner & Jungner (1975), Donner (1978). Bojsen & Frederiksen (1980), Ing6lfsson et al. (1990), Frich & Ingolfsson (1990), Bennike et al. (1994), and the present work. 280 M . Rasch & N . Nielsen I Ingcilfsson ct :ti. IUYO). The shaded area in the diagram r e p r e w i t 9 the band i n which the RSL cur\'e must lie (higher than dated marine material and I m w r than dated terrestrial/lacustnne material). The diagram differs from previously published emergence curves from Disko Bugt ( D o n n e r 6: Jungner 1075: Frich & Ingolfsson 1990: Ingolfsson et al. lY90) by only including dates from a restricted area with a uniform alti- tude o f the Holocrnc marine limit. I n addition. no interpretations concerning depth range of occurrence o f dated fossils a.ere applied in the ~- TUAPAAT, D I S K 0 ISLAND GEOMORPHOLOGY Beach ridge (recent) Beach ridge (fossil) Marine terrace Cliff Ic] Lake I Lagoon , Alluvial fan 200 rn I Fig. 3. Aerial photo of the studv area at 1 uapaat (Aerial photo- XXK-123. Copyright: Kort OR Matrikelstyrelsen, Denmark) construction of the diagram. Most of the animals normally found as fossils in marine deposits in West Greenland live in a wide zone (from the shoreline to water depths of tens of metres), and interpretations concerning depth range of occur- rence will therefore be very uncertain unless good information about stratigraphy and sedimen- tologv of t h e adjoining sediments is available ( D o n n e r & Jungner 1975; Andrews 1986: Sutherland 1987; Pirazolli 1991). A geomorphological sketch m a p of the coastal landscape at Tuapaat (Fig. 4) was prepared using F i g . 4 . Gcomorphological sketch map of the \tudy area at Tuapaat. Coastal morpho-stratigraphy and Holocene relative sea level changes 281 Fig. 5 . The study area at Tuapaat. The intenor of the cuspate foreland consists of coarse clastic simple beach ridges. In the background, towards the west, an alluvial fan covers the simple beach ridges (Fig. 6). The cliff on the right aide of the photo is approximately 7 m high. aerial photograph interpretation and field data. Levels of coastal landforms were measured in terrain profiles using electronic theodolite and echo-sounder (Figs. 8 and 9). Local datum was based on tidal observations on the location. Absolute and relative errors on altitude measure- ments are considered approximately 50.5 m and less than 2 0 . 2 m. respectively. A coastal rnorpho-stratigraphy of the cuspate foreland was interpreted from the geomorpho- logical sketch map and from the terrain profiles (Figs. 8 and 9). The term morpho-stratigraphy denotes a system for relative dating of landscape elements. A rnorpho-stratigraphic unit is a strati- graphic unit primarily identified by the surface form it displays. The applied principles of coastal Fig. 6. An alluvial fan superimposes the coastal deposits in the westernmost part of the study area. The alluvial fan is built up by sediments eroded from the marine terraces at higher elevations. I n the spring and early summer, the interior, eastern part of the foreland is flooded (notice the lake in left side of the photo). The lake is dammed by an ice core within the present beach ridge. The lake drains when the ice core melts in the early autumn. 282 M . Rasch & iV. Nielsen RELATIVE SEA LEVEL SOUTHEAST D I S K 0 AND KlTSlSSUT rn a.s.1. 90 €3 70i I 1 50 4 0 LEGEND: A PEATIGYTTJA 8 SHELLS WHALE ? t l l l l l l l i l l l 0 1 2 3 4 5 6 7 8 9 1 0 1 1 ka BP Fig. 7. Age-altitude diagram of “C dates from southern Disko Island and Kitsissut. The shaded area marks the zone in wluch the RSL curve must lie. Upward directed arrows point to the threshold altitude of lakes in which dated gyttja samples have heen collected. The downward directed arrow in the lower left corner of the diagram points to the expected RSL (Om a.s.1.) at the time when the whalebone in Profile 2 (K-6181) was deposited. The dotted vertical line represents dating 1-6239 (see Table 2 ) . The exact field altitude of this sample is unknown. I t is however known that the sample was collected above sea level. Sources for the dates: See Table 2. morpho-stratigraphy have been described in detail by Rasch and Nielsen (1994). Five different morpho-stratigraphic unit types were identified at Tuapaat (Table 1, Figs. 8 and 9). Three pits were dug within Profile 1 in the alluvial fan deposits in the western part of the study area (Figs. 4 and 8). The northernmost and the southernmost pits both reached the surface of the beach deposits underneath the alluvial deposits. Because of permafrost, it was not poss- ible t o excavate to the coastal basement in the middle pit. In the northernmost pit, three layers of peat were discovered just above the coastal deposits. The stratigraphy of this pit was logged (Fig. lo), samples of the different layers were analysed for grain size, and plant species within the peat layers was determined by Bent Fredskild, Museum of Botany, University of Copenhagen. Three different facies types were identified from the log. A whale bone was found embodied in the sedi- ments on the landward side of the present beach in Profile 2 in the eastern part of the foreland (Fig. 9). The whale bone and the peat layers from the northernmost pit in Profile 1 were radiocarbon dated (Table 2). T h e samples K-6179, K-6180 and K-6181 were dated at the Carbon-14 Dating Laboratory at the National Museum of Denmark, Copenhagen, while sample AAR-1422 was dated at the AMS I4C Laboratory at the University of Aarhus, Denmark. The date of K-6181 was corrected for isotopic fractionation by nor- malising t o dL3C = 0.0%0 PDB. K-6179, K-6180 and AAR- 1422 have been corrected by norrnal- king to 6I3C = - 2 5 . a 0 PDB. Results Age-altitude diagram The age-altitude diagram in Fig. 7 illustrates well the emergence of southeastern Disko Island in the early part of the Holocene. Between 9 and 7 ka B P , the RSL fell continuously with a rate of about 3 cm/yr from ca. 8 0 m a.s.1. to ca. 30m a.s.1. The width of the shaded area in the lower part of the diagram illustrates that present knowl- edge about the exact course of the emergence of southeastern Disko Island after ca. 7 ka BP is very limited. However, the lower part of the diagram indicates that the RSL reached present sea level in the period between 4 . 7 and 1.0 ka BP, and that it has probably been below present sea level for at least part of the period between 4 . 7 ka BP and the present. Coastal morpho-stratigraphy and Holocene relative sea level changes 283 Fig. 8 . Topographic profile from the western part of the cuspate foreland. The position of Profile 1 is shown in Fig. 4. (Profile 1). The morpho-stratigraphy of the profile has been logged above Profile 1B. The abbreviations used in the morpho-stratigraphic log are defined in Table 1. A detailed log of the stratigraphy in the northernmost pit (P3) is given in Fig. 10. PROFILE 1 A - TUAPAAT 1993 m a.s.1. m.s.1. -20 I i 1 0 250 500 750 1000 m PROFILE 1B I WR I Morpho-stratigraphy RBRP I CC I RBRP m a.s.1 Legend : = Beach deposil (Recent) w Beach deposil (Fossil) L5J Alluvial tan deposit El Peat 10- 5 - ................... H.H.W.S. - H.L.W.S. 0 - m a.s.1 P2 i 0 Legend = Beach deposil (Recent) w Beach deposll (Fossil) L5J Alluvial tan deposit El Peat ........................ H.H.W.S. ...* H.L.W.S. .. - - - - - - - - - - . ................. ....... *., 1 50 100 150 200 250 300 m T a b k 1. Charactcristics of the five morpho-stratigraphic unit types identified within the cuspate foreland at Tuapaat (compiled from Rasch & Nielsen 1994). Morpho-stratigraphic unit type Characteristics Indication of Code Simple beach ridge Washover ridge Narrow sharp-crested beach ridge with internal beds generally dipping 5'-10" towards the sea. Coastal progradation SBR Beach ridge with rounded crest and with internal beds following the surface of the ridge. Washover ridges are usually higher and wider than simple beach ridges. Typical internal bed slopes are less than 5". Coastal recession WR Gently sloping marine terrace with several more or less parallel simple beach ridges on the top. Coastal cliffs are usually more or less straight, coast parallel cliffs. Coastal deposits will always occur at the foot of a coastal cliff. Coastal recession cc Limit between morpho-stratigraphic units of very different age. Coastal recession MU Regressive beach ridge plain Coastal cliff Coastal progradation RBRP Morphologic unconformity 281 M . Rasch & ,V. ,Vielseti PROFILE 2 - TUAPAAT M o r p h o - s t r a t i g r a p h y RERP I C C ,RBRP, W R J m a . s I. 20 1 0 50 t o o 150 200 m ,t~orpho-strfitigraph?' The morpho-stratigraphy of the cuspate foreland at Tuapaat (Figs. 8 and Y) indicates four phases of shoreline movement. T h e marine terraces above ca. 10 m a.s.1. indicate coastal progradation ( R B K P ) . This was followed by coastal recession ( C C ) and resulted in the formation of the fossil coastal cliff at t h e northern edge of the foreland. Phase three was another period of shoreline pro- gradation ( R B R P ) which formed the simple beach ridge plain within the cuspate foreland. T h e tinal phase involved coastal recession ( W R ) and the development of the washover ridge at the shore- line o f the foreland. The washover ridge at the present shoreline consists of two separated ridges (Fig. 11). In Profile 1, i t is not possible to distinguish between the two ridges. probably because the inner ridge ha3 been superimposed by alluvial fan deposits (Fig. 8 ) . However. in Profile 2 a sharp boundary occurs between the outer a n d the inner ridges of the washover ridge (Figs. 9 and 11). O n the outer ridge the beach sediments a r e washed clean by wave5 and washovers, while lichens. comprised mostly of Rhizocurpoti jeriitlandiciirri with typical diameters of 2 cm. grow on t o p of the sediments of the inner ridge. Both the inner and outer ridges contain large volumes of sediment (Fig. 11) and n a r l I SAMPLE NO. 1920729-23 i 9 2 0 7 2 9 - 2 2 7930729-21 1130729-20 1930729-19 T930729-18 1920729-17 FLCRA Reworked p a r k 01 Cyperaceae Reworked Iresh water mosses Reworked lresh w a l e r m o s s e s LAB.NC. K-6180 K-6181 A A A - 1 4 2 2 AGE " C Y B P 390: 55 480' 75 980 2 60 Fig. 10. L o g of t h e c t r a t i g r a p h y in t h e n o r t h e r n m o s t pit (P.3) i n Protilc 1 (Fig. 8). G r a i n s i r e ( M , ) and s o r t i n g ( d , ) arc given i n F - v a l u e s . Facies 1 consist, of t h i n l e n t i c u l a r l a y c r s o f band a n d d t . R o o t h o r i z o n s o c c u r s c a t t e r e d within the facies. The facies type ha\ been i n t e r p r e t e d iis alluvial f a n deposits Facics :consists of;i p o o r l y w r t c d . \ l i t ? d i m c n t with r c w o r k c d parts ( i t pl,inr\ and n i o s \ e c T h e f a c w tvpc h a s h s e n i n t e r p r e t e d ;is lake s e d i m e n t \ o r i g i n a t i n g f r o m sinall p o o l \ in d c p r c 4 o n i in t h e al1uvi.d f a n . Fmrs 3 c o n s i s t s of \and with \c.ittered o c c u r i c n c c \ o l roundcd-\ccll r o u n d e d pehhles a n d cobble\. 'This facics typc has been i n t e r p i e l e d n\ h e a c h d i i n c n t s Coustul tiiorpho-Jtrutigrupliy atid Holocetie rekutiue sea leuel cliatigrs 285 Tuhlr 2. Radiocarbon dates from the southeastern part of Disko Island and Kitsissut. Where possible. dates on marine material have heen corrected for isotopic fractionation by normalising to 6°C = O.mr PDB. while dates on terrestrial niatcrial have heeti corrected for isotopic fractionation by normalizing to 6'3C = -25.0% PDB. ~ Lac no. 3 10 10 10 5 10 7 3 6 5 6 5 5 3 6 6 1 3 2 3 3 2 2 4 9 8 11 11 9 9 6 6 1 Ldh n 0 K-5774 ~ - 6 1 8 0 ~ - 6 1 8 1 K-6179 K-5777 Hel-945 AAR-1422 UA-1787 K-5775 K-3908 K-5778 K-3909 K-5779 1-6239 K-5776 K-3666 K-3691 K-3505 Hel-904 Hel-905 Hel-903 Lu-3040 Hel-906 HcI-907 Lu-3038 K-4568 RCD-21 K-3660 K-3667 Hel-7210 K-4567 AAR-5 K-3665 __ N Idt ~ 69"27' 69024' 6Y24' 69"24' 69"16' 6929' 69"24' 6Y04' 69"27' 69"17' 69"16' 69"17' 69"16' 69"16' 69"27' 69"17' 6Y15' 69"32' 6927' 69"29' 69"26' 69"26' 69"29' 69"29' 69"26' 69"18' bK"59' 6T40' 69"40' 6978' 69"IX' 69"17' h9"17' __ W long. 53"40' 52"37' 52"36' W37' 53"50' 53"38' 52"31' 53"32' 53"40' 53"28' 53"50' 53%' 53"50' 53"50' 53"40' 53"28' 53"35' 53"41' 53"40' 53"38' 53"42' 53"35' 53"38' 53"38' 53"27' 53%' 53"lY' 52"00' P o l ' 53"15' 53"15' 53"28' 53"28' ~ Material Gyttja Peat Whale Peat Gyttja Peat Peat Peat Gyttja Gyttja Gyttja Gyttja Gyttja Shells Gyttja Gyttja Shells Gyttja Shells Shells Shells Shells Shells Shclls Shells Shells Shells Shells Gyttja Shells Shells Shells Gyttja Elevation m a.s.1. 13.5' 2.9 3.1 3.2 112 0.3 3.5 5.2 12.5' 82 102 82 8.5' ?3 13' 82 5-7 20-30 18.7 3.3 6.8 45 40.3 21.6 10-12 69 62 40 100 69 28 82 82 Age. "C yr5 B P 370? 55 390 2 55 480? 75 670 -t 75 890-t 65 Y70 ? 110 9XUt- 60 1610 ? 100 1780t- 55 2050t- 80 2330t- 75 3900+ 90 3930% 85 4685 t- 120 4730 % 115 53102 95 5990t- 95 6750 + 105 681)O t- 150' 6840 t- 140 8050 ? 170' 8170% 80 8280 + 170' 8340 % 170' 8390 2 80 8620 2 120 8690k 90 X700 t- I20 8950 t- 125 9060 2 120' 9220 ? 130 9240 2 250 10180 t- 155 b ' T % PDB Reterence -20.1 -27.7 -14.5 -27.7 -24.3 -28.2 -20.6 -24.2 -22.3 -24,s -24.1 -23.5 -24.7 - I .o -22.8 2.1 I .5 1 .8 1.1 2 . 0 4.5 0.9 0.8 0 . 3 - 12.9 1.7 0.6 -23.7 1 .n Bennikr 1994 This work This work This work Bennike 1994 Donner 1978 This work Bennike et al. 1994 Bennike 1994 Ingolfsson et al. 1990 Bennike 1994 Ingdfsson et al. 1990 Bennikc 1994 Weidick 1973 Bennike et al. 1990 Ingolfsson ct al. 1990 Ingnlfsson ct a l . 19YO Foged 1989 Donner 1978 Donner 1978 Donner 197X Ingdtsson et al. 1990 Donner 1978 Donner lY7X Ingolfsson et al. 1990 Frich B Ingolfsson 1990 Bennike et al. 1994 lngdfsson et al. 1990 Ing6lfsson et al. 1990 Frich B lnp6lfsson 1990 Frich & lngnlfsson 1 Y Y O Ingdlfsson et al. I Y Y O Ingolfsson et al. 1Y90 ' Datc was not corrected for isotopic fractionation by authors of tirst publication. Date has been corrccted to b"C = 0.0% PDB. ' Threshold of lake situated at 16.5 m a d . ' A field altitude of 19m a.s.1. was quoted by Weidick (1973). According to Frich & Ingdfsson (1990) this altitude is incorrcct. have small dunes developing o n the surface. It is unlikely that the ridges a r e ephemeral coastal landforms resulting from single larger storms. T h e ridges more probably represent permanent fea- tures developed at different times as a result of coastal recession. The basal peat (Fig. 10, 3 . 0 6 3 . 0 8 rn a.s.1.) in the northernmost pit (P3) in Profile 1 was dated t o 980 * 60 BP (AAK-1422). T h e date implies that the fossil coastal cliff behind the cuspate foreland, and at least the inner part of the simple beach ridge plain of the cuspate foreland, developed before 980 5 60 BP T h e KSL did not reach the present sea level before 4.7 ka. Accordingly, the fossil coastal cliff developed during the period between 4.7 and 1 ka BP T h e whale bone that was found in the inner beach ridge of the washover ridge was dated to 480 2 75 BP (K-6181). This result implies that the inner beach ridge of the washover ridge was active in the 15th century, that the simple beach ridge plain of the cuspate foreland was developed before the 15th century, and that the o u t e r beach ridge of the washover ridge was formed after the 15th century. 286 M . Rasch & N. Nielsen Discussion There has been some confusion about the shape of Holocene RSL curves of Disko Bugt. Donner and Jungner (1975) used a best fitting second degree polynomia for their construction of a RSL curve representing the entire Disko Bugt. The curve shows emergence in the period between deglaciation and about 1.5 ka BP followed by sub- mergence in the period between 1.5 ka BP and the present. According to their curve, the RSL was below present sea level between ca. 3 ka BP and the present. In later attempts to construct RSL curves for the entire Disko Bugt. it has been presumed that the RSL fell continuously with a decreasing rate throughout the Holocene, and that the RSL did not fall below present sea level (Frich & Ingolfsson 1990: Ingolfsson et al. 1990). The present study supports Donner and Jungner’s idea of a RSL history characterised by emergence in the early part of the Holocene followed by submergence in the late Holocene. In the Nuuk area in the southern part of West Greenland, Norse ruins that now occur below sea level demonstrate that the RSL in that area has been lower than at present (Weidick 1976). No such unequivocal indication of lower than present relative sea level has been observed on south- eastern Disko Island. However, in Disko Bugt, frequent occurrence of Eskimo ruins that have been partly destroyed by coastal erosion con- Fig. J l . The washover ridge that encompasses the cuspate foreland at Tuapaat consists of two separated ridges. The inner ridge is covered by the lichen, Rhizocarpon jemflarrdicum, with typical diameters of 2 cm. N o lichens occur on the outer ridge. Notice the small dunes on top of the ridges. vincingly suggest that one or more transgressions have occurred since habitation of the ruins (Mathiassen 1934). The theory is strongly sup- ported by the fact that washover ridges and coastal cliffs occur frequently along the present shores of southern Disko Island (Rasch & Nielsen 1994). In addition, a Dorset-Culture ( 6 0 0 4 BC) midden that was found below the mean high water level in the southeastern part of Disko Bugt is difficult to explain without presuming that the RSL has been below present sea level ( J . F. Jensen, Museum of Qasigiannguit pers. comm.). The proposed late Holocene submergence in Disko Bugt may have resulted from isostatic adjustment to late Holocene (Neoglacial) glacier advances (Kelly 1980; Funder 1989). Alterna- tively, the proposed Holocene RSL changes on southeastern Disko Island (emergence in the early part of the Holocene followed by late Holocene submergence) might have resulted from the migration of a collapsing fore-bulge. RSL-curves showing emergence followed by submergence occur frequently in the literature concerning RSL changes in the outer part of formerly glaciated areas (e.g. Forman 1990; Quinlan & Beaumont 19-31), and convincing accordance between such RSL-curves and theoretical RSL-curves has been achieved in geophysical models that include the migration of a collapsing forebulge o n a visco- elastic earth (Clark et al. 1978; Quinlan & Coastal morpho-stratigraphy and Holocene relative sea level changes 287 Beaumont 1981; Andrews & Peltier 1989; Liverman 1994). Only trends of the RSL history can be inter- pre'ted from age-altitude diagrams. The inves- tigation of the morpho-stratigraphy in the lowermost part of the coastal landscape at Tua- paat does, however, allow a discussion of possible smaller fluctuations of the RSL in the latest part of the Holocene. Changes in the shoreline position (recession/progradation) at a given location must result from either changes of the RSL, shifting of the wave climate or changes of littoral sediment sources. Changes in shoreline position caused by shifting wave climate or changes of the littoral sediment sources will be expected to have only local significance, while changes in shoreline posi- tions caused by RSL changes will be expected to have significance over larger areas and to occur on coasts with different exposure and sediment sources. Therefore, when indications of synchron- ous and similarly directed shoreline movements are observed at locations with different exposure and sediment sources, it is most probable that the shoreline movements have resulted from chang- ing RSL. Coastal landscapes with a morpho-stratigraphy very similar to the morpho-stratigraphy of the cuspate foreland at Tuapaat are common on Disko Island (Fig. 12). Therefore, it is suggested that the morpho-stratigraphy at Tuapaat, and at the other Disko Island study sites, reflects shore- line displacements caused by RSL changes. This suggestion implies that the proposed coastal recessions at Tuapaat in the period between 4.7 and 1.0 ka BP, before and after the 15th century, all resulted from transgressions caused by rising RSL. Unfortunately, the Tuapaat field obser- vations do not allow for an exact determination of the relative sea level at the transgression maxima. Weidick (1993) has suggested that transgression maxima occurred in the coastal part of West Greenland in the 14-15th century and in the middle of the present century. This interpretation is based on archaeological observations of the Fig. 12. Topographic profiles of the lowermost part of the coastal landscape at different sites on Disko Island (compiled from Rasch et al. in press). All profiles have been measured perpendicular to the shoreline. The morpho-stratigraphy of the profiles has been logged above the profiles. The abbreviations used in the morpho-stratigraphic log have been defined in Table 1 . m 3,s I. SAOQARLIIT IlORLlll N 60'44'N 54'32'W m.s.1. TUAPAAT 68'24" 52'35'W 0 50 IW m 0 50 1W 150 m 288 M . Rasch & N . Nielsen spatial relation between the present shoreline and Thule-culture Eskimo ruins of different ages (Mathiassen in Gabel-J~rgensen & Egedal 1940) and on early tidal observations (Saxov 19.58). Recently, supporting evidence for this theory have been found i n a study of morpho-stra- tigraphy and chemical stratigraphy within a coastaflagoonal landscape system at Saqqarliit Ilorliit on western Disko Island (unpubl. data) (Fig. 1). At Saqqarliit Ilorliit, high sodium and sulphur content and relatively coarse grain sizes in lagoon sediments from the 14th century and the middle of the present century might suggest transgression maxima in these periods. Based on the Saqqarliit Ilorliit data it might further be suggested that transgressions have occurred early in the period between 2.5 and 0.7 ka B P and i n the 1 6 1 7 t h century. The morpho-stratigraphy at Saqqarliit Ilorliit is illustrated in Fig. 12. The transgression early in the period between 2.5 and 0.7 ka BP led to the formation of the coastal cliff, while the transgressions in the present millennia led to the development of the washover ridge immediately landward of the shoreline and to marine influenced sedimentation in lagoons behind the washover ridge. Considering the errors on radiocarbon dates, t h e Tuapaat results agrees well with the pre- liminary results from Saqqarliit Ilorliit. The coastal cliff at Tuapaat may have formed during the transgression that was dated to the period between 2.5 and 0.7 k a BP at Saqqarliit Ilorliit. The date of 980 2 60 BP on the peat sample from the northernmost pit (P3) in front of the fossil coastal cliff at Tuapaat only implies that the cliff was formed before ca. 1.0 ka BP Consequently, the dating reduces the possible time interval for the transgression to the period between 2.5 and 1.0 ka BP. The inner beach ridge of the washover ridge at Tuapaat was probably active during the 14-15th century transgression maximum, and the outer beach ridge of the washover ridge might have formed during the 1 6 1 7 t h century transgression and/or during the transgression culminating in the present century. An age of at least 500 years on the washover ridges constituting the present beaches in Disko Bugt also agree well with the implications of a date of 380 2 80 BP (1-16414) on a whale bone that was found in the present beach ridge on the western part of Nuussuaq peninsula, northwestern Disko Bugt (Bennike et al. 1994; 0. Bennike; Geological Survey of Denmark pers. comm.). This whalebone was probably also deposited during or after the transgression cul- minating in the 14-15th century. A morpho-stratigraphic sequence similar to that at Tuapaat has also been described from Siniffik ca. 20km west of Tuapaat (Fig. 12) (Rasch & Nielsen 1994). Based on the sizes of lichens, Rhizocarpon jemtlandicum, on the beach ridges at Siniffik and on the oldest buildings in the town of Qeqertarsuaq, it was suggested that the relict coastal cliff (Fig. 12, CC) at Siniffik developed during a transgression in the middle of the 15th century. As the morpho-stratigraphic sequences at Siniffik and Tuapaat are almost identical, it is probable that the individual mor- pho-stratigraphic units at the two sites developed synchronously. This implies that the fossil coastal cliff at Siniffik was formed before 0.7 ka BP. Conclusions Between 9 and 7 ka BP the RSL on southern Disko Island fell continuously with a rate of about 3 cm/ y r from ca. 8 0 m a.s.1. (the Holocene marine limit) to ca. 30 m a.s.1. The RSL reached present sea level between 4.7 and 1 .O ka BP Present knowl- edge of the course of the RSL curve on southern Disko Island between 7 and 1 ka BP is very limited due to the scarcity of dates on marine material from this period. It is suggested, however, that the RSL history of southern Disko Island was one of emergence in the early part of the Holocene followed by submergence in the late Holocene. The suggestion implies that the RSL has been below present sea level during the period between 4.7 ka BP and the present. Indications of at least three late Holocene trans- gressions were found at Tuapaat. The exact timing of the transgressions is unknown, though several data suggest that the first transgression occurred early in the period between 2.5 and 1.0 ka BP, that the second transgression culminated in the 14-15th century, and that the third transgression culminated in the middle of the 20th century. In addition, one transgression may have occurred in the 16-17th century. Acknowledgements. - The field investigations at Tuapaat were funded by the Danish Natural Research Council and were carried out during the summers of 1993 and 1994. Logistics for the field work and transportation to the field site were provided by the Arctic Station in Godhavn/Qeqertarsuaq. Thanks are extended to B. Fredskild for determination of plant species Coastal morpho-stratigraphy and Holocene relative sea level changes 289 in peat samples, to 0. Humlum and J . Desloges for helpful comments on the manuscript, t o P. Unger, K. Kjaer, T . Olsen and P. Olsen for assistance in the field and to J . Jdnsson and 0. B. Berthelsen for technical help with the figures. References Andrews. J . T . 1986: Elevation and age relationship: raised marine landforms in glaciated areas: Examples based on Canadian Arctic data. Pp. 67-95 in van de Plassche, 0. (ed.): Sea leuel research: a manual for the collection and eualuation of data. Geobooks, Nonvich. Andrews, J . T . & Peltier, W. R. 1989: Quaternarygeodynamics in Canada. Pp. 541-572 in Fulton, R . I . (ed.): Quaternary geology of Canada and Greenland. Geological Survey of Canada, Geology of Canada 1. Bennike, O . , Hansen, K. B . , Knudsen. K. L., Penney. D . N. & Rasmussen, K . L. 1994: Quaternary marine stratigraphy and geochronology in central West Greenland. Boreas 23. 194-215. Bojsen, T . & Frederiksen, L. 1980: Kystmorfologi ved Ivisarqut og Narssarssuk. Pp. 31-64 in Engraf, A , , Nielsen, J., Nielsen, N. & Humlum, 0. (eds.): Arktisk Geomorfofologi, Enoks Haon. DBko 1978. Kobenhavns Universitets Geografiske Institut, Geonoter 5. Carter, R . W. G. & Orford. J . 1984: Coarse clastic barrier beaches: A discussion of the distinctive dynamic and morpho- sedimentary characteristics. Mar. Geol. 60, 377-389. Clark, J . A , , Farrel. W. E. & Peltier, W. R . 1978: Global changes in postglacial sea level: A numerical calculation. Q. Res. 9 , 265-287. Donner, J . & Jungner, H . 1975: Radiocarbon dating of shells from marine Holocene deposits in the Disko Bugt area, West Greenland. Boreas 4 , 25-45. Donner. J . 1978: Holocene history of the west coast of Disko, central West Greenland. Geogr. Ann. 60A, 63-72. Forman. S . L. 1990: Post-glacial sea level history of north- western Spitsbergen, Svalbard. Geol. SOC. A m . Bull. 102. 1580-1590. Frich, P. & Ingolfsson, 0. 1990: Det holoczne sedimen- tationsmiljd ved Igpik samt en model for den relative land- h a n i n g i Disko Bugt omridet, Vestgronland. Dan. Geol. Foren. h s k r . 1987-89, 1-10, Funder, S . 1989: Quaternary geology of the ice-free areas and the adjacent shelves of Greenland. Pp. 743-792 in Fulton, R. J. (ed.): Quaternary geology of Canada and Greenland. Geological Survey of Canada, Geology of Canada 1. Gabel-Jergensen, C. C. A . & Egedal, J . 1940: Tidal obser- vations made at Nanortalik and Juliannehaab in 1932-1934. Medd. Gr(n1. 107(2). 47 pp. Henderson, G., Rosenkrantz, A. & Schiener, E. J . 1976: Cretaceous-Tertiary sedimentary rocks of West Greenland. Pp. 341-362 in Escher, A . & Watt, A . S . (eds.): Geology of Greenland. Gronlands Geologiske Undersogelse, K ~ b e n h a v n . glacier initiation line altitude: A methodological approach to the study of rock glaciers. Arc. Alp. Res. 20(2), 160-178. Ing6lfsson, 0.. Frich, P., Funder, S . & Humlum, 0. 1990: Paleoclimatic implications of an early Holocene glacier advance on Disko Island, West Greenland. Boreas 19. 297-311. Kelly, M. 1980: The status of the Neoglacial in western Greenland. Grgnl. Geol. Unders. Rapp. 96. 14 pp. Kelly, M . 1985: A review of the Quaternary geology of western Greenland. Pp. 461-501 in: Andrews, J . T . (ed.): Quaternor?. enoironments, Eastern Canadian Arctic, Bafjirt Bay and Western Greenland. Allen and Unwm, Boston. Liverman. D. 1994: Relative sea level history and isostatic rebound in Newfoundland, Canada. Boreas 23. 217-230. Mathiassen, Th. 1934: Contribution to the archaeology of Disko Bay. Medd. Grgnl. 93(2). 192 pp. Nielsen, N. 1Y79: Ice-foot processes. Observations of erosion of a rocky coast, Disko, West Greenland. Zeitsch. Geomorph. Pirazzoli, P. A . 1991: World atlas of Holocenesea level changes. Elsevier. Amsterdam. 300 pp. Quinlan. G. & Beaumont, C. 1981: A comparison of observed and theoretical postglacial relative sea level in Atlantic Canada. Can. J . Earth Sci. 18, 11461163. Rasch, M. & Nielsen, N. 1994: Holocene relative sea level changes indicated by morpho-stratigraphic sequences: Disko island. West Greenland. Geogr. Tidrskr. 94, 37-45. Rasch, M . , Christiansen, H . H., Hansen, B. U., Hasholt. B., Humlum, 0.. Jakobsen, B. H . & Nielsen, N. in press: Greenland landscape elements as indicators of rapid environ- mental change. In Berger, A. R. (ed.): Internationa[ Work- shop of Geological Indicators of rapid environmental change. Corner Book, New Foundland. July 11-17, 1994. A. A . Balkema Publishers, Rotterdam, Netherlands. Saxov, S . 1958: The uplift of Western Greenland. Medd. Dan. Geol. Foren. 13, 51&523. Sutherland, D. G . 1987: Dating and associated methodological problems in the study of Quaternary sea level changes. Pp. 16S197 in Devoy, R . J . N. (ed.): Sea surfacestudies: a global oiew. Croom Helm, London. Tanner, W. F. 1988: Beach ridge data and sea level history from the Americans. J . Coast. Res. 4 , 81-91. Weidick, A. 1968: Observations on some Holocene glacier fluctuations in West Greenland. Medd. Grdnl. 165(6). 202 PP. Weidick, A. 1972a: C" Dating of survey material performed in 1971. Grdnl. Geol. Unders. Rapp. 45, 5 6 6 7 . Weidick. A . 1972b: Holocene shore-lines and glacial stages in Greenland - an attempt to correlation. Grgnl. Geol. Unders. Rapp. 41. 39 pp. Weidick, A . 1973: C L 4 dating of survey material performed in 1972. Grdnl. Geol. Unders. Rapp. 55, 6 6 7 5 . Weidick, A . 1974: CI4 dating of survey material performed in 1973. Grdnl. Geol. Unders. Rapp. 66, 42-45. Weidick. A. 1976: Glaciation and the Quaternary. Pp. 431-458 in Escher, A . & Watt, A . S. (eds.): Geology of Greenland. Gronllands geologiske Undersogelse, Kobenhavn. Weidick, A . 1993: Neoglacial change of ice cover and the related response of the Earth's crust in West Greenland. Gronl. N.F. 23(3), 321-331. Humlum, 0. 1988: Rock glacier appearance level and rock Geol. Unders. Rapp. 159, 121-126.