Indications of a Y o unger Dryas marine transgression in inner Hardanger, West Norway
STEIN KJETIL HELLE, KARL ANUNDSEN*, SOLVEIG AASHEIM & HAFLIDI HAFLIDASON
Helle, S. K., Anundsen, K., Aasheim, S. & Hafiidason, H.: Indications of a Younger Dryas marine transgression in inner Hardanger, West Norway. Norsk Geologisk Tidsskrift, Vol. 77, pp. 101-11 7. Oslo 1997. ISSN 0029-196X.
Litho- and biostratigraphy of sediment cores from three peatbogs situated at the head of the Hardangerfjord, western Norway, have been investigated with respect to late glacial relative sea-leve! changes. The palaeoenvironmental record, with emphasis on the diatom stratigraphy, reveals a late glacial marine transgression. Based on age and amplitude estimates, the transgression is closely identified with the Late Weichselian (Younger Dryas) transgression, which previously has only been documented at the outer coast of western Norway. The highest late glacial shorelines (marine limit) both at the head of the Hardangerfjord and at the outer coast of western Norway probably developed during the transgression maximum, suggesting a similar age of the inland and coastal late glacial marine limits. In order to accommodate the recorded sea-leve! changes in inner Hardanger with the established sea-leve! history of coastal areas, an earlier deglaciation of the Hardangerfjord than hitherto believed is proposed.
S. K. Helle & H. Haj/idason, Department of Geo/ogy; S. Aasheim, Botanical Institute, University of Bergen, A/legt. 41, N-5007 Bergen, Norway.
Introduction si1s in the sediment cores. The resu1ts from this study challenge previous reconstructions of relative sea-level Late glacial sea-level changes in the outer coastal areas of changes and ice-sheet models for the Late Weichselian western Norway are mainly known from investigations of and early Holocene in the Hardangerfjord area, and shed lake deposits. In his pioneering works, Fægri (1940, 1944) new light on the isostatic uplift history of the in found evidence of a Late Weichselian transgression in land region, close to the main water divide of western western and southwestern Norway based on pollen-dated Norway. ingression and isolation horizons in lake-sediment cores. Later works (Anundsen 1977, 1978; Anundsen & Fjeldskaar 1983; Krzywinski & Stabell 1984) confirmed Physiography and geological setting this transgression event, although some controversy exists on duration and timing of cu1mination of the transgres The peatbog basins are situated at the head of the sion. The transgression represents a significant sea-level Hardangerfjord, more than 160 km from the coast (Fig. rise with an amplitude in the order of 12-15 m, cu1minat 1). One basin Iies on the southern side of the fjord ing in Y ounger Dryas time ('the Y o unger Dryas trans (Bu-113, Fig. 2) on a 1.5 km wide, nearly flat valley gression', Anundsen 1978, 1985). The Younger Dryas shou1der which forms a bold promontory in the Eidfjord. shoreline is thus documented to Iie higher than any older The other two basins Iie in a wide, gentle-s1oping, north late glacia1 shoreline a1ong the entire coast from Sotra and east facing valley-side on the west side of the U1vikfjord south to Jæren (Fig. l) (Anundsen 1985). Further inland (Vambheim-119 and Vambheim-128, Fig. 3). All three along the fjordsof western Norway raised marine deposits basins, including their catchment areas, are surrounded are prominent along the fjord tributaries and bear witness by Precambrian bedrock, main1y foliated gneisses and to a substantial 1and-uplift after the deg1aciation. Along quartz diorites (Qvale 1981; Sigmond et al. 1984). the Hardangerfjord, however, directly dated evidence of The Late Weichselian ice-front oscillations at the outer late glacial sea-1evel changes is rare. The sea-level history part of the Hardangerfjord area have been described by and course of deglaciation along the fjord is therefore still a number of authors (Undås 1963; Holtedahl 1967; insufficiently known. Pollestad 1972; Aarseth & Mangerud 1974; Holtedahl Based on the stratigraphic results from three peatbog 1975; Genes 1978; Sindre 1980). There is a general agree basins from inner Hardanger, a shore-level displacement ment in these works that the mid and inner parts of the curve for the Late Weichse1ian and early Holocene is Hardangerfjord were ice-covered during the Y ounger constructed. Changes in palaeosalinity in each basin, i.e. Dryas. Until recently, debate has mainly focused on the the transitions between fresh, brackish and marine sedi outer parts of the fjord, addressing whether the ice-front ments, are reconstructed from the evidence of microfos- reached its maximum position at Halsnøy in late Younger Dryas (e.g. Aarseth & Mangerud 1974), or whether it had an even farther westward extension in the
• Deceased September I l, 1995. early Y o unger Dryas (Sindre 1980). 102 S. K. Helle et al. NORSK GEOLOGISK TIDSSKRIFT 77 ( 1997)
Fig. l. Location map of the inner Hardangerfjord area. A detailed map of the investigated localities is shown in Fig. 2 (Bu-113) and Fig. 3 (Vambheim-119 and Vambheim-128). The location of the four highest-lying terraces in inner Hardan ger are shown; the Tunheim terrace 125 .6-129.2 m a.p.s.l. (Tu) in Ulvik, the terraces at Bjotveit 128 m a.p.s.l. (B) and Tveisme 125 m a.p.s.l. (T) in Ullensvang and the Erdal terrace 122 m a.p.s.l. (E) in Eidfjord. Inset is a key map of South Norway (S =Sotra, B =Bømlo, J =Jæren).
Fig. 3. Detai1ed maps of Vambheim-119 and Vambheim-128 with drilling pro files. Profile A-A' in Vambheim-119 is shown in Fig. 7. CD= lowest basin threshold Q)= other basin outlet (!)= 11 O mm drilling si te.
Quaternary morphological elements in inner Hardan ger were mapped by Anundsen & Simonsen ( 1967). The shore-levels in inner Hardanger have been described pre vious1y by a number of authors (Rekstad 1911; Kaldho1 1941; Simonsen 1963; Anundsen 1964; Undås 1964; Anundsen & Simonsen 1967; Holtedahl 1975; Hamborg 1983). The marine limit in the Ulvikfjord area is defined by the prominent Tunheim terrace situated between 125.6 and 129.2 m above present sea-level (a.p.s.l.)· (Kaldhol 1941) (Fig. 1). The marine limit along the Eidfjord is defined by the terraces at Bjotveit (128 m a.p.s.l.), Tveisme (125 m a.p.s.l.) and Erdal (122 m a.p.s.l.) (Holtedahl 1975) (Fig. 1). Until now, only two 14C dates were available as to the age of early postglacial sea-levels in inner Hardanger. A freshwater gyttja from immediately above the marine to lacustrine transition in a sediment co re from Bu-113 (the same basin as in this study) is dated to 9720 ± 330 BP, and gives the age of the isolation of this basin from the fjord (Anundsen & Simonsen 1967). In addition, a 14C date on wood (Ju niperus communis) found in the lower part of a terrace in Fig. 2. Detailed map of Bu-113 with drilling profiles. The profile A-A' -A" is Eidfjord with a surface at approximately 23 m a.p.s.l., shown in Fig. 4. dating, however, is interpreted as an approximate age of petrographical characteristics. Samples with a known a sea-level 100 m a.p.s.l. in inner Hardanger (Rye 1969). volume ( 13-18 cm3) were prepared in accordance with Hamborg ( 1983) constructed a shoreline diagram for Meldgaard & Knudsen (1979). Some hand-picked ash the Hardangerfjord in which the shore-levels along the particles were also geochemically analysed by a JEOL- fjord were dated on the basis of regionally interpolated 6400 scanning electron microscope connected to a TRA ages. Based on the constructed shorelines, Hamborg COR NOR THERN 5600 spectrometer in accordance proposed a deglaciation model for the Hardangerfjord with the procedure described by Haflidason ( 1983). and suggested a stepwise ice-front retreat between 10,200 and 9600 BP along the main fjord and its tributaries. Quantitative diatom analysis Field and laboratory methods The succession of diatoms is, in the present study, the most important parameter for reconstructing the paleo The lithostratigraphy of each basin was obtained from environment of the basins, although it has been inter sediment cores using a Russian-type corer and a 110 mm preted in conjunction with other environmental in piston corer. Coring with the Russian corer was accom dicators. In each diatom sample, when possible, at least plished along longitudinal and transverse profiles of the 300 diatom valves have been counted, as recommended basins with co ring intervals between l O and 20 m. In by Palmer & Abbott ( 1986; cf. Galehouse 1971 ). How order to determine the exact depth to bedrock and reveal ever, because of low concentrations of valves in the lower the bottom morphology of the basin in detail, steel rods part of the sediment sequences, the number of counted were sledge-hammered through the most coarse and most diatom valves is 1ess than 300. The identification of each compact bottom sediments, down to bedrock. Finally, all taxon is based mainly on Schmidt et al. (1874-1959), potential basin outlets were thoroughly levelled in order Peragallo & Peragallo (1897 -1908), Hustedt (1930- to detect the lowest basin threshold and its accurate 1966), Hustedt (1930), Cleve-Euler (1951- 1955), Patrick height above sea-level. & Reimer (1966), Krammer & Lange-Bertalot (1986, The co res tak en with the piston corer ( 11 O mm co res) 1988, 1991), Simonsen (1987), Lange-Bertalot & Kram were brought to the laboratory for detailed analyses. In mer ( 1989) and Alles et al. ( 1991 ). the laboratory the cores were described lithologically and The diatoms are grouped into Hustedt's (1957) salinity photographed. The organic content in the sediments was classes. The salinity preferences of the species are based determined by loss-on-ignition and TOC (total organic mainly on Cleve-Euler (1951- 1955), Hustedt (1957) and carbon) analyses. The loss-on-ignition was calculated as Simonsen ( l962). In the succession diagrams each species a percentage of the loss in weight of dry samples after is assigned to its respective halobian ( salinity) class. being ignited for 4 hours at 550°C. The TOC content in Unidentified specimens were placed in the 'Unknown' the sediments was determined using a LECO EC-12 group. In the total diagrams, species with unknown Carbon Determinator. salinity preferences were placed in the 'Unknown halo Samples for diatom analyses where taken at approxi bian' group. Fragilaria spp. were included in the total mately 4 cm intervals along the 110 mm cores. In addi diatom-sum in the calculations (l' total), since they gen tion, samples from a number of Russian cores were erally bad a low occurrence in the spectra ( usually well checked for their content of diatoms in order to confirm below 30% of total sum). Each diatom zone is defined on the general dia tom stratigraphy in the 11 O mm cores. the basis of the most dominating taxa ( > 10%) (Pienitz Each sample was treated with 30% hydrogen peroxide et al. 1991). and boiled. After cooling ovemight, the samples were centrifuged for 2 minutes at a speed of 1200 r.p.m., and the suspended clay material was decanted. The residue was resuspended with distilled water and spread onto a Locality l: Bu-113 cover glass. The samples were dried onto the cover glass Bu-113 has a surface area of about 75,000 m2 and a on a hot plate and finally mounted to a slide using catchment area of about 765,000 m2 (Fig. 2). It has a Naphrax mounting medium. well-defined rock threshold, with height levelled to 112.9 A number of samples were prepared for pollen analysis m a.p.s.l. according to the procedure described in Fægri & Iversen ( 1975). In addition, some samples were prepared for analyses of foraminifera according to Meldgaard & Lithostratigraphy Knudsen ( 1979). The chronostratigraphy of the sediment co res has been The detailed lithostratigraphy of the 110 mm core is outlined based on 14C dates, pollen analysis and analysis given in Table l and Fig. 5. The main lithological units of volcanic ash. The content of volcanic ash particles in of the basin, as revealed from the Russian co ring profiles, the lower part of the sediment sequence was quantified is summarized in Fig. 4. These units can easily be iden and classified in accordance with morphological and tified in the stratigraphy of the 11 O mm co re. 104 S. K. Helle et al. NORSK GEOLOGISK TIDSSKRIFT 77 (l 997) Table l. Sediment description of the IlO mm core from Bu-11 3. layer K and J and is marked by a distinct increase in the loss-on-ignition values (F ig. 5). Depth (cm The two lowermost sediment layers in the basin (a bo ve Layer below surface) Description a bottom layer of diarnicton interpreted as till) comprise K 360-373 Dark brown fine detritus gyttja sand y silt and sand y silt with gra vel, respectively ( corre 373-376 Olive-grey gyttja-rich sandy silt with grave! sponding to layers A and in the 110 mm core, Table l 376-385 Grey coarse sand with grave! B l H 385-393 Grey sandy silt with grave! and Fig. 5). The layers above are thinner and consist of G 393-413 Dark grey sandy silt. al ternating beds of sand, silt and sandy gra vel ( corre Distinct lower boundary sponding to layers C-J). All these layers can be traced F 41 3-421 Grey coarse sand with grave!. Distinct lower boundary along the central part of the basin. The organic-domi E 421-428 Dark green- grey sandy silt with grave!. nated sediments a bo ve consist of fine- and coarse detritus Distinct lower boundary gyttja and peat. D 428-450 Dark green- grey sandy silt. Upward-increasing content of sand c 450-460 Green-grey sand with grave!. Upward-fining from coarse to fine sand. Upward-decreasing content of grave!. Biostratigraphy Distinct lower boundary The basin is divided into six diatom zones (Diatom zones B 460-486 Dark green- grey sandy silt with gravel A 486-51 2 Dark blue-grey sand y silt I-VI, Fig. 5). In the lowermost layers, A to the lower part of layer D (spectra nos. 35 to 24), the diatoms are either absent or toa sparse to count and have not been The deepest, central part of the basin Iies in the included in the diatom zones. Above layer K the diatom northeast, where it has a trough-formed depression. Here content has not been quantified, but qualitative analyses the sediment thickness is over 8 m. The sediments in the reveal only freshwater diatoms. The content of pollen, basin can be divided into two main units; an upper unit spores, dinofiagellate cysts and foraminifera in the lower dominated by organic sediments, and a lower minero most stratigraphical units is shown in Fig. 6. A pollen genic unit. The transition between these two units is diagram for the upper organic-rich part of the basin distinct and can easily be followed along the entire coring stratigraphy is presented and discussed in Anundsen & profile. In the 110 mm core the organic-rninerogenic Simonsen (1967). The lowermost spectrum in Anundsen sediment transition corresponds to the boundary between & Simonsen's pollen record is characterized by high 110 mm 116 c ore A' i- l l l li l l l l l l l l 11 l l l l l l l l li l l l l l l l l l l l l l l l l l l l l l l l l l l l li l l ui110 110 .,; E ,. 2' 3' s· 106 O'� 106 1 • 90' o 100 200 300 (m) Organic dominated sediments Minerogenic sediments � Sandy silt Sphagnum peat �Sand ' silt and f7"':\l gravel -layers �(LayerA) � Bedrock (Layers D-J) � C oarse detritus r:::-:1S and with � gytt)a_ � Glaciotectonized L.:.:.Jgrave! �sediments (Layer C) l l l l • Drilling points � Fine detritus (Russian corer gytt)a_ r:-:L":lSandy silt with & � grave! (Layer Till? steel rods) (Layer K) l:..:..:...:J D B) Fig. 4. Lithostratigraphy in Bu-113 (profile A-A'-A" in Fig. 2). The profile has a vertica1 exaggeration of 10.1:1. NORSK GEOLOGISK TIDSSKRIFT 77 (1997) Marine transgression, Hardanger, West Norway 105 Betula and NAP (non-arboreal pollen) values and is The ingression contact in the sediments, however, is correlated with the uppermost pollen spectrum in our probably at the lithological transition between layers F pollen record. and G, i.e. slightly below the lower boundary of Diatom zone Ill. Of the other microfossils in this interval, di Diatom zone I (442-421 cm, spectra nos. 23-17). -The noflagellate cysts dominate (Fig. 6). The diatom assem lowermost dia tom zone occurs in layers D (sandy silt) blage characterizes a marine littoral environment. and E (sandy silt with gravel). The diatom valves appear in low numbers in this zone, and the valves generally Diatom zone IV (384-376 cm, spectra nos. 6-5). - The bear signs of strong corrosion. Of the l3 different taxa upper and lower boundaries of Diatom zone IV coincide identified in this zone, Trachyneis aspera is by far the with the lithological boundaries of layer I ( coarse sand most dominating taxon, representing more than 50% of with grave!). The concentration of diatom valves and the the total diatom sum in the lowermost spectra. T. aspera num ber of intact valves are high in this zone. Zone IV is is characterized as polyhalobian, meioeuryhaline, with an dominated by freshwater forms, mainly the halophobous optimum at salinity of 30%0 or higher (Simonsen 1962). Frustulia rhomboides var. saxonica, Tabellaria flocculosa Because of its robust character, T. aspera is probably and Eunotia incisa. The zone consists mainly of olig somewhat over-represented in the sediments compared to otrophic taxa, but also includes the eutrophic Epithemia the original diatom assemblage in the water column. sorex var. sorex and the mesotrophic Stauroneis phoeni Among the other strongly represented taxa in this zone is centeron var. phoenicenteron. The diatom assemblage in Diploneis bambus. This species is characterized by dicates lacustrine environments, probably with acid, Simonsen (1962) as polyhalobian, meio- to mesoeury meso-oligotrophic conditions. Close to the transition to haline, i.e. a taxon with optimum at salinity over 30�00, Diatom zone V there is a distinct occurrence of but generally it has a wider salinity tolerance interval foraminifera, mainly the planktonic Neog/obigerina than T. aspera. Apart from T. aspera, which has a pachyderma (sin.) (Fig. 6). At the same level, the poly bimodal, tychoplanktonic habitat, the other taxa in this halobous diatom Nitzschia punctata var. apiculata has its zone have a periphytic, non-planktonic habitat. Gener first appearance. This probably reflects that, in the latest ally, the polyhalobian taxa represent between 50 and part of this lacustrine phase, sea-water periodically spills 76% of the total assemblage in this zone. Despite the low over the basin threshold, implying a sea-level close to the numbers of diatoms they typify a marine littoral environ basin threshold. ment. Of the other microfossils in this interval, dinoflag ellate cysts are the dominating forms. Diatom zone V (376-374 cm, spectrum no. 4). - This zone occurs in layer J (gyttja-rich sandy silt) and is Diatom zone Il (420-409 cm, spectra nos. 16-14). - A defined on the basis of only one spectrum. The diatom distinct change to a freshwater diatom assemblage takes assemblage in this zone has a very high diversity, with 39 place at the transition to layer F. Oligohalobous-indiffer different taxa. The zone is rich in intact diatom valves. ent and halophobous forms dominate Diatom zone Il, It is dominated by the salinity-tolerant oligohalobous representing between 76.9% (spectrum no. 14) and 87.4% indifferent Cocconeis placentula var. placentula ( > 40%). (spectrum no. 16) of the total diatom assemblage in this Simonsen (1962) has characterized C. placentula var. zone. One of the most dominating taxa is the halo placentula as oligohalobous, pleioeuryhaline, i.e. with phobous Frustulia rhomboides var. saxonica. This species salinity-tolerance limits between O and 20�00• According is characterized as a periphytic (non-planktonic) form, to Hustedt (1957) C. p/acentula var. placentula has an and lives on the sediment surface ( epipelic) or loosely optimum in alkaline, eutrophic conditions. The occur connected to higher plants (Round et al. 1990). Ingmar rence of Epithemia adnata var. adnata and E. sorex var. (1973) described F. rhomboides var. saxonica as a pioneer sorex (both alkalibiontic with optimum at pH > 7) also form in acid, oligotrophic lakes. An acid, lacustrine bears witness to eutrophic, alkaline conditions. The poly environment is also indicated by the appearance of halobous Nitzschia punctata var. apiculata and the oligo Tabel/aria flocculosa, Eunotia exigua var. exigua and halobous-halophile Navicula cari var. cincta is also Navicula subtillisima (e.g. Schroeder 1939). present in this zone. Both species are extremely salinity tolerant. Diatom zone V, containing salinity-tolerant Diatom zone Ill (408-385 cm, spectra nos. 13-7). - freshwater forms together with brackish and marine Diatom zone Ill occurs in layers G (sandy silt) and H forms, indicates an alkaline, brackish lake, probably with (sandy silt with grave!). The diatoms appear in low sea-water periodically spilling over the basin threshold. A numbers in this zone. It is characterized by the reappear marine impact on the lake in this period is also indicated ance of the marine forms Trachyneis aspera, Navicula by the occurrence of foraminifera, mainly Neoglobigerina latissima, Dip/oneis didyma and D. bambus, the same taxa pachyderma (sin.), and dinoflagellate cysts (Fig. 6). that dominated Diatom zone I. The transition between Diatom zone Il and Diatom zone III Iies slightly above Diatom zone VI (373-363 cm, spectra nos. 3-1). - The ( 4 cm) the lithological transition between layers F and G. transition to Diatom zone VI coincides with the litholog- 106 S. K. Helle et al. NORSK GEOLOGISK TIDSSKRIFT 77 (I997) Bu-113 112,9 m a.p.s.l. LOSS·ON· VOI.CANIC ASH IGNITION PARTICLES 3 % number/cm Diatom barren Fig. 5. Lithology, 14C dates, loss-on-ignition, volcanic ash particles and diatoms in Bu-11 3. The data are from a IlO mm core positioned as shown in Fig. 4. A sediment description of the core is presented in Table l. ical transition to layer K (fine-detritus gyttja). The zone Chronostratigraphy is dominated by freshwater ( oligohalobous-indifferent and halophobous) forms. The oligohalobous-indifferent Radiocarbon dates, pollen analysis and volcanic ash form Cocconeis placentula var. placentula, which domi provide the basis for a chronostratigraphic framework nated Diatom zone V, has a high representation in the for the recorded sea-level changes in Bu-113 (see, how lowermost spectrum also in this zone ( approximately ever, Discussion, below). Four radiocarbon dates ( one 40% in spectrum no. 3), but disappears in the uppermost conventional and three AMS) have been carried out in two spectra. The halophobous forms reappear in this the sediment record of Bu-113 (Table 2 and Fig. 5). In zone, mainly represented by Tabellaria floccu/osa and T. addition, one radiocarbon date from the basin was avail fenestrata. These taxa were also prominent in Diatom able from Anundsen & Simonsen ( 1967); a bulk sample zones Il and IV, and indicate a lacustrine environment. of gyttja conventionally dated to 9720 ± 330 BP (T-585). Diatom zone VI represents the last lacustrine phase and The stratigraphic level radiocarbon dated by Anund the final isolation of Bu-113 from the fjord. sen & Simonsen ( 1967) is correlated with the layer Bu-113 112,9 m a.p.s.l. I P= IAP + (%)INAP IP+X(%) Trees �hnå>olli Herbs Aquatics Spores Algae Dlnoftag. cysts XI x .. .. E �+ Foramlnl- �c: ., E -.::: " § E E O.. o,. ;; '5. 2 øra � � Chrono- .." 2 i� stratigraphy .!! il 1å � '!! .!! j l =.!!! 5 De lj o o pth � "l5 O.l- (11/cm � ..... i i � 21� ! !li } 0.. � 'i-� (cm)� Cl) HH Clltii 25 50 5 10 10�� 55!l 5 20I tO 20 o 20 20 tO 50 50 i 5 360 fj K Holo- p,... l Freah ta cene boreal � - - �!l 5 tJ:: � - - - 75 �_950 � � l � Fresh � � 83 48 � Late 387 � 55 - r 400 Wølch· 99 - 18 Marine G � sellan - - � l-- 15 29 43 -Freth 23 8 12 � = � - l-- 1-"- � =. 28 9 48 Marine D 450 Fig. 6. Pollen, spores, algae, dinoflagellate cysts and foraminifera in the lower part of the stratigraphy in Bu-11 3. Correlation with Simonsen's (1980) local pollen assemblage zone (paz) la is shown and a proposed chronostratigraphic framework is outlined. NORSK GEOLOGISK TIDSSKR!Ff 77 (1997) Marine transgression, Hardanger, West Norway 107 Olll".OO -"' - :::; : -... - _,., - _,., - Fig. 5 (continued). boundary K/J in our core (Fig. 5). Their date is syn Small traces of volcanic ash particles have been found chronous with our date of the K/J layer boundary in the Late Weichselian part of the stratigraphy (Fig. 5). (9665 ± 125 BP; T-9619A), and gives the age of the final Petrographical and geochemical analyses (for details, see isolation contact in Bu-113. An early Preboreal age of Helle 1993) show that the ash particles correspond to the this layer boundary is also indicated by the pollen record rhyolitic fraction of the Vedde Ash tephra described by of Anundsen & Simonsen ( 1967). Mangerud et al. (1984). The Vedde Ash Bed has recently Our pollen record is from the lower late glacial sedi been redated to about 10,300 14C years BP (Bard et al. ments in the basin and is stratigraphically below Anund 1994, Birks et al. 1996). We suggest that the marine layer sen & Simonsen's pollen record (Fig. 6). Based on these G, which contains the maximum peak of Vedde Ash in pollen records and on an AMS 14C date of layer J our record, is of late Y ounger Dryas age (see, also, (10,065 ± 125; Ua-2586), we place the Late Weichselian Discussion, below). Holocene transition at the boundary between layers I The remaining two AMS 14C dates from Bu-113 are and J (Fig. 5). from plant fragments found in sediment layers I and F. Table 2. Radiocarbon dates. Laboratory b Basin• reference Material" 14C years BP o'3C'Iood Bu-11 3 TUa-416 Plant fragments 7,595 ± 80 -27.7* Bu-11 3 TUa-415 Plant fragments 9,550 ± 110 -27.7• Bu-113 T-9619A Gyttja 9,665 ± 125 -25.5 Bu-11 3 Ua- 2586 Gyttja 10,065 ± 125 Vambheim-119 T-10148 Piece of wood (Pinus) 8,300 ± 100 -26.1° Vambheim-119 Ua- 2584 Gyttja 8,400 ± 95 Vambheim-119 T-9618A Gyttja 10,935 ± 135 -30.0 Vambheim-128 Ua-2585 Piece of wood (Pinus) 6,150 ± 180 Vambheim-128 T-9654A Gyttja 8,485 ± 120 -32.1 • Refers to the basin from which the dated sample was collected. b Laboratory designations: T=The Radiological Dating Laboratory in Trondheim, Norway (conventional dates); TUa=The Radiological Dating Laboratory in Trondheim, Norway/The Svedberg Laboratory in Uppsala, Sweden (AMS dates); U a=The Svedberg Laboratory in Uppsala, Sweden (AMS dates). c All dated material is collected from 110 mm cores. In Bu-11 3 and Vambheim-128, samples are from the same 110 mm core (i.e. from the same drilling site in the basin. See text for details). In Vambheim-119 samp1es with 1aboratory references T-10148 and Ua-2584 are from a 110 mm core from position S (Fig. 3), whereas T-9618A is from a 110 mm core from position P (Fig. 3). The gyttja samples consist of 2-3 cm thick sediment slices (except for Ua- 2584 which is from about 4 g gyttja) from which the NaOH soluble fraction was dated. In the remaining samples the NaOH insoluble fraction was dated. TUa-415 and TUa- 416 are accelerator dates of small plant fragments sieved from 2 cm thick sediment slices. Plant fragments were too small to be identified. d The o 13C values marked by asterisks are not measured, but assumed values given by the da ting la bora tory. 108 S. K. Helle et al. NORSK GEOLOGISK TIDSSKRIFT 77 (1997) The two dates are inverted and gave the ages 9550 ± 110 easily be followed along the entire coring profile. In the (TUa-415) and 7595 ± 80 (TUa-416), respectively. Con 110 mni core the organic-minerogenic sediment transi tamination of young carbon from rootlets is a possible tion corresponds to the boundary between layers J and K explanation of the inverted dates. Rootlets cannot be and is marked by a distinct increase in the loss-on-igni excluded as the dated plant fragments were too small to tion values (Fig. 8). be identified. If the chronostratigraphic framework out The lowest sediment layer in the basin (abo ve the lined above is correct both dates must be too young. impenetrable bottom layer of diamicton interpreted as till) consists of thinly bedded sandy silt and sand with grave!. This layer, which corresponds to layer A in the 11 O mm co re (Ta ble 3), is found along the central and Loca1ity 2: Vambheim-1 19 deepest part of the basin. Above, two layers of thinly This basin has a surface area of about 7500 m2 and a bedded and fine-laminated sandy silt and fine sand (lay catchment area of about 200,000 m2 (Fig. 3). It has a ers B and C, respectively) can be followed from the well-defined rock threshold at the northeastern boundary deepest part of the basin and towards the basin-edge. of the basin, levelled to 118.6 m a.p.s.l. The unit above consists of alternating layers of sand and silt, occasionally with some gra vel horizons ( correspond ing to layers J to Din the 110 mm core, Table 3 and Fig. Lithostratigraphy 8). This unit of layers can be followed along almost the entire profile A-A' (Fig. 7). Above, the sediments consist Detailed lithostratigraphy of the 11O mm core is given in of alternating layers of fine and coarse detritus gyttja. Table 3 and Fig. 8. The main lithological units of the Uppermost is a thick sequence of Sphagnum peat. basin, as revealed from the Russian coring profiles, are summarized in Fig. 7. These units are readily identifiable in the stratigraphy of the 110 mm core. Biostratigraphy The basin is deepest in the south with sediment thick ness over l O m. The sediments in the basin can be The basin is divided into three diatom assemblage zones divided into two main units; an upper unit dominated by (Diatom zones I-Ill) and cover spectra nos. 22 to l organic sediments and a lower minerogenic unit. The from sediment layers K to D (Fig. 8). In the three transition between these two units is distinct and can lowermost layers, A to C, the diatoms are either absent A' 124 l l l l 124 110mm 110 mm core core (Pos. P.) (Pos. S) 120 120 .; .,; 116 116 E 112 112 o 40 80 120 160 (m) Organic dominated sediments Minerogenic sediments Sand silt and Very thinly Sphagnum peat � ' Et=.! grave! -layers � bedded sandy (Layers D-J) silt and sand Coarse detritus � with grave! � gytt]a_ L2....JBedrock � (Layer A) � Laminated and Fine detritus � Till? . � very thinly � gytt]a l l l l l Drilling points bedded fine sand D (Layer K) (Russian corer & and sandy silt steel rods) (Layers B-C) Fig. 7. Lithostratigraphy in Vambheim-1 19 (profile A-A' in Fig. 3). The profile has a vertical exaggeration of 4.2: l. NORSK GEOLOGISK TIDSSKRIFT 77 (1997) Marine transgression, Hardanger, West Norway 109 Table 3. Sediment description of a 110 mm core (position P) from Vambheim- The oligohalobous-indifferent and halophobous forms 119. which dominate in Diatom zone I disappear in Diatom Depth (cm zone Il. Instead, zone Il is overtaken by the poly Layer below surface) Description halobous Diploneis didyma, Navicula latissima, Nitzschia punctata var. apiculata and Trachyneis aspera. According K 550-565 Dark brown silty fine detritus gyttja J 565-569 Blue-grey sand y silt with gra vel to Simonsen ( 1962), Diploneis didyma and Nitzschia I 569-582 Blue-grey sandy silt. punctata var. apiculata are salinity tolerant, and can live Upward-decreasing content of sand in salinities between 30-0.2/"00 and 30-8/"00, respectively. H 582-588 Grey coarse sand. Distinct lower boundary G 588-591 Blue-grey sandy silt. Both live on the sediment surface ( epipelic) in the littoral Upward-decreasing content of sand zone. Trachyneis aspera is not so tolerant and occurs in F 591-610 Grey sandy grave!. salinity between 35 and 17/"00 (Simonsen 1962). The di Distinct lower boundary E 610-616 Blue-grey sandy silt with grave! atom assemblage in zone Il displays a marine littoral D 616- 628 Grey coarse sand with grave! environment. c 628-642 Fine-laminated blue-grey sandy silt and grey There is a distinct appearance of Pinus pollen in this fine sand. 2-3 mm thick parallel lamina B 642-700 Very thinly bedded blue-grey sandy silt and zone, with values up to 80% of the total pollen sum (Fig. grey fine sand. 1-2 cm thick beds 9). The Pinus pollen could possibly originate from a A 700-737 Very thinly bedded blue-grey sandy silt and long-distance source. The high values are probably due grey sand with grave!. 1-2 cm thick beds to marine over-representation (Florin 1945). Dinoftagel late cysts also occur in zone Il, represented mainly by Peridinium faeroense. According to Rochon & de Vernal or they occur in very low numbers and have not been (1994), this is a cold marine form. included in the diatom zones. Above layer K the diatom content in the sediments has not been quantified, but Diatom zone Ill (564-557 cm, spectra nos. 2-1). - The qualitative analyses reveal only freshwater diatoms. The transition to Diatom zone Ill coincides with the litholog content of pollen, spores and dinoftagellate cysts in the ical transition to layer K ( silty fine-detritus gyttja). The lowermost stratigraphical units is shown in Fig. 9. zone is rich in diatoms, and the valves are generally well preserved. The marine forms disappear and the oligo Diatom zone I (627-569 cm, spectra nos. 22-7). - The halobous-indifferent and halophobous forms return. lowermost diatom zone occurs in sediment layers D to I, Zone Ill is characterized by an increase of the halo bracketed within the lithological boundaries of these phobous Brachysira brebissonii f. brebissonii and Frustu layers. The number of diatom valves in this zone is lia rhomboides var. saxonica. Both have an optimum at generally high ( except for the dia tom barren spectrum low pH and oligotrophic conditions. Together with the no. 19). The halophobous Frustulia rhomboides var. sax oligohalobous-indifferent Fragilaria construens var. con onica dominates with up to 40% of the total diatom struens, they dominate in zone IlL As a whole, the assemblage in the zone. F. rhomboides var. saxonica is diatom assemblage in Diatom zone Ill corresponds characterized as a pioneer form in acid, oligotrophic closely to the diatom assemblage in zone I and indicates lakes (Ingmar 1973). The high number of well-preserved acid, oligotrophic lacustrine conditions. Of the other valves of this species, with its delicate structure, indicates microfossils in this zone, the aquatic plant Nymphaea that it is autochthonous in the sediments, and thus and the green algae Pediastrum and Botryococcus are the closely reftects the sedimentary environment of the lake. most prominent (Fig. 9). Diatom zone Ill represents the Among the other prominent taxa in this zone are the last lacustrine phase and the final isolation of Vamb halophobous Brachysira brebissonii f. brebissonii and heim-119 from the fjord. Tabellaria flocculosa, and the oligohalobous-indifferent Fragilaria construens f. construens and Achnanthes pusilla var. pusilla. The diatom assemblage in Diatom zone I, Chronostratigraphy which covers a 58 cm long interval of the core with only minor diatomological changes, indicates a prolonged pe Three radiocarbon dates ( two conventional and one riod with acid, oligotrophic lacustrine conditions. The AMS) from Vambheim-119 (Table 2 and Fig. 8) were diatom barren spectrum no. 19 probably reftects a short carried out with the purpose of dating the final isolation period with very low productivity in the water column. contact in the basin. Gyttja from the lower boundary of At the top of this zone, near the transition to the marine layer K (i.e. the isolation contact), taken from a core Diatom zone Il, there is a small increase in the occur from the central part of the basin (position P, Figs 3 and rence of dinoftagellate cysts (Fig. 9). 7), was conventionally dated to 10,935 ± 135 BP (T- 9618A) (Fig. 9). From the same stratigraphical position, Diatom zone Il (568-565 cm, spectra nos. 6-3). - Di but from a 11O mm core taken in a more marginal part atom zone Il occurs in layer J (sandy silt with gravel). of the basin (position S, Figs 3 and 7), one conventional The zone is rich in diatoms, but the valves bear signs of dating (T-10148) and one AMS dating (Ua-2584) gave being corroded. 8300 ± 100 BP and 8400 ± 95 BP, respectively (Table 2). 110 S. K. Helle et al. NORSK GEOLOGISK TIDSSKRIFT 77 {1997) Vambheim-119 118,6 m a.p.s.l. Polyhalobous Fig. 8. Lithology, 14C date, loss-on-ignition and diatoms in Vambheim-119. The data are from a 110 mm core from position P in Fig. 7. A sediment description of the core is presented in Table 3. Thus, the radiocarbon dates diverge significantly and with values as high as the early Holocene Pinus rise (Fig. give no clear evidence of the age of the isolation contact. 9). The Pinus maximum in layer J lies stratigraphically However, by comparing our pollen record (Fig. 9) with below Simonsen's lowest pollen zone. Simonsen's ( 1980) pollen stratigraphy for inner Hardan In peatbog sediments at Kroken in Lusterfjorden in ger and with other pollen records from western Norway, inner Sogn the early Holocene Pinus rise was dated to some inferences of the age of this lithological transition approximately 8600 BP (Vorren 1972). In the same can be made. The lowest pollen zone in Simonsen's record, Vorren found a Pinus maximum in a marine layer pollen record, zone la of Preboreal age with dominance below, dated to before 10,520 ± 450 BP. Studies on the of Betula, Juniperus and Sa/ix, is recognized in our pollen Late Weichselian vegetation history at the coast of record (Fig. 9). The early Holocene Pinus rise, which southwestern Norway revealed a Pinus maximum during defines the transition between Simonsen's local pollen Y ounger Dryas (P aus 1988). Paus interpreted the Pinus zones la and lb, and is dated to approximately 8900 BP, maximum as a result of declining local pollen production is found in our pollen record about 20 cm above the during Y o unger Dryas and a relative increase of long lower boundary of layer K. Below, in the marine Jayer J distance transported Pinus pollen from northwestern Eu in our record, there is another distinct Pinus maximum rope. Although they are marine over-represented the Vambheim-119 118,8mLp.a.l. IP•IAP+INAP (%) IP+X(%) Trees Shrubs Herts Aquetlcl Spcq• AlgaaDinoflag.cysta l c p; �- Gl c: E :il l . c: Chrono- CL .. .!! !!l .!! !l l Dopth E stratigraphy l i � � (cm if � ;i � j �� i Ill! 0.. .B w w ) � ! 1 J �� u IP ·l ! 50 50 20 20 10 20 10 10 10 1 I10_J 10 1010 o 10 1010 20 1010 10 50 J 530 39 J } l Boreal .J � Ill ! 08� Hole>- 1b � � �7 cene l Pre- 1a Fresh 1 11 19 .,.,. ""' / �j 08 ,...""' - � �5 96 - � !;:- ;; � Dryu ;.... - � i � � Late � 1 Weich------� 21 24 selian Fresh � �93 - 600 F - - - - � f- � 6 7 =- f:.. =- 41 � o ��195 �5 8 Fresh - re- - - f-- - ��i 10 n r--r- Fig. 9. Pollen, spores, algae and dinoflagellate cysts in Vambheim-119. Correlation with Simonsen's (1980) local pollen assemblage zone (paz) la and lb is shown and a proposed chronostratigraphic framework is outlined. NORSK GEOLOGISK TIDSSKRIFT 77 (1997) Marine transgression, Hardanger, West Norway 111 Oligohalobous - indifferent Halophobous Unknown halobian l li ,; [ U� > -� l" i � �� ! 1 jz .'j�ffg�l !. IL l i il l! l! li!l lt li 1l ��tml � li! -- 1� 1� �� 1-1 � 1.: �� � l!lm ! li l z l1 z11 1 ,_; 10 l 1o l1o l 10 30 l 1o l1o l10 l1o l1o l1o l1o ! 10 l 10 l1li�o 11��o 110 l 10 ,. l 10 l 10 l 10 l 10 l 10i� l1Il0 l,; l 10 J ... .. �.J::: l ':: Fig. 8 (c ontinued). Pinus pollen in sediment Iayer J in our record could Lithostratigraphy corresponding1y originate from a Iong-distance source in The detailed lithostratigraphy of the 11O mm core is Y ounger Dryas. given in Table 4 and Fig. 10. The main lithological units Very small traces of volcanic ash particles have been of the basin, as revealed from the Russian coring profiles, found below the J/K Iayer boundary. The occurrence of can easily be identified in the stratigraphy of the 11O mm tephra in the sediment record has not been quantified in core. The sediments of the basin can be divided into two detail, but the concentration is significantly lower than in main units; a Iower rninerogenic sediment unit which Bu-113. However, petrographical and geochemical analy consists of al terna ting Iayers of sand and silt ( corre ses (for details see Helle 1993) show that the ash particles sponding to Iayers A-E, Fig. 10 and Table 4) and an correspond to the rhyolitic fraction of the Vedde Ash upper, organic-rich sediment unit (layers F, G and H). A tephra, recently redated to about l O 300 BP (Bard et al. �ttja-ri�h gravel horizon (layer F) separates the organic 1994, Birks et al. 1996). nch sedtments above from the minerogenic sediments Based on this the Y ounger Dryas-Pre bo real transition below. in the sediments is suggested to occur somewhere be The basin has a maximum depth of only 3 m in the tween the J/K sediment Iayer boundary and the early southeast. The shallowness of the basin makes the sedi Holocene Pinus rise (Fig. 9). The isolation contact ment column vulnerable to erosion and resedimentation. (layer boundary J/K) is probably of late Younger Dryas The gravel horizon (layer F) could possibly represent age (about 10,300 BP or later) (see, also, Discussion, such an erosion surface in the sediment sequence and below). thus indicate a hiatus. Due to Iow organic content in the sediments we were not able to 14C date the ingression contact in the basin (layer boundary 1/J, Fig. 8). Apart from our inference that it is Late Weichselian, its exact age is not known. Table 4. Sediment description of the 110 mm core from Vambheim-128. Depth (cm Layer below surface) Description Locality 3: Vambheim- 128 H 200-212 Dark brown silty fine detritus gyttja G 21 2-217 Olive-green gyttja-rich silt This basin has a surface area of only 750 m2 and is F 21 7-218 Olive-green gyttja-rich gravet situated proximal to a relict river fan (Fig. 3). The E 21 8-244 Blue-grey sandy silt. Distinct Jower boundary catchment area is about 36,250 m2 and Iies within the D 244-246 Blue-grey silt. catchment area of Vambheim-119. The Iowest rock Distinct lower boundary threshold, situated in NW, was levelled to 128 m a.p.s.l. c 246-248 Blue-grey sandy silt B 248-249 Grey sand with gravet The basin threshold has about the same height as the Distinct lower boundary highest terrace in Ulvik, the Tunheim terrace, which is A 249 -274 Fine-Jaminated blue-grey sandy silt and grey situated between 125.6 and 129.2 m a.p.s.l. (Kaldhol fine sand. 1-2 mm thick silt lamina, 2-3 mm thick sand lamina 1941). 112 S. K. He/le et a/. NORSK GEOLOGISK TIDSSKRIFT 77 (1997) Vambheim-128 128 m a.p.s.l. Oligohalobous-indifferent Halophobous halobian c .EI.E o !Z UJ � ::;; Vl > z ::::> 1 ".. Loss- o "t> ijl "' I > on- o � ·c; � .c ei UJ I E CL ignition 0.. .c� o 8: Vl > � ·� ., .. � � a: g .11 € � � % < i -� � � i= z -� -� l! I I � �::::> " 3 � :;; I 1'i UJ li a; 3 ...... � ... � .c c 0.. � 0.. ::;; illl c UJ ...... UJ o J 111 .t ...: .... li: o H o � IIIIIIIIII::::::::::J: :> � i l all lall � 30 50 90 10 30 5_Q 70 J10 10 li J10 10 lj 10 30 10 10 1_0 �o l i 209 3 17 209 -- 213 ,_ - r 213 317 IG �- - � -l r r--• • 40 m �485±120 !Marine Ill�� �l m 2 226 226 3 230 230 1 E 234 • 234 - Dia tom barren 2<2 9 242 • - - �:: :�::7 �g :' i:: 254 + 12 254 - l c 260 · 13 260 - · ·(· ··· l U150±80 · · · :L ·272·1--- Fig. JO. Lithology, 14C dates, loss-on-ignition and diatoms in Vambheim-128. Layer A contains a few foraminifera (16 Neoglobigerina pachyderma). Layer E also has a limited foraminiferal fauna (28 N. pachyderma , 3 Orbulina un iversa and 2 Cass idulina reniforme). The data are from a IlO mm core positioned as shown in Fig. 3. A sediment description of the core is presented in Table 5. The transition between layers F and G is interpreted to be a hiatus (see text). Biostratigraphy Above layer H the diatom content has not been quantified, but qualitative analyses reveal lacustrine con In the lowermost spectra, which cover the minerogenic ditions with freshwater diatoms. part of the stratigraphy (sediment layers A to E), the diatoms are either absent or too sparse to count (Fig. 11). However, some foraminifera are present in the sediment Chronostratigraphy layers A and E (for details, see Helle 1993), comprising mainly the planktonic form Neog/obigerina pachyderma Two 14C dates ( one conventional and one AMS) have and the benthonic Cassidulina reniforme and Orbulina been obtained from Vambheim-128 (Table 2). The lower universa. Owing to the occurrence of foraminifera the boundary of the gyttja-rich sediment layer F was conven lower minerogenic part of the stratigraphy in Vambheim- tionally dated to 8485 ± 1 20 BP ( T-9654A) (Fig. 10). 128 is interpreted to be of marine origin. Fragments of pine found in sediment layer A were dated The diatom assemblages in sediment layers F, G and to 6150 ± 180 BP (Ua-2585). Neither pollen analysis nor H contain only freshwater forms and the number of ash particle analysis has been carried out in this basin. diatom fragments is extremely high. The high degree of The lowermost, organic-poor marine sediment sequence fragmentation can be explained in terms of high turbu (layers A to E) is, however, considered to be late glacial lence in the water column of a shallow lake (less than 2 deposits. We find support for this by comparing the m deep) situated proximal to a river outlet. Caloneis loss-on-ignition curve from this basin (Fig. l O) with the hebes dominate spectrum no. l in silty fine-detritus gyttja loss-on-ignition curves of the lowermost late glacial sedi (layer H) by more than 80%, whereas in spectrum no. 2 ment sequences of Bu-113 and Vambheim-119 (Figs 5 C. hebes represents less than l% of the total dia tom and 8). The date of pine fragments is thus considered to assemblage. This species is characterized as oligo give too young an age for layer A. The dated fragments halobous-indifferent with a benthic habitat. Tabellaria could have been contaminated with roots. fiocculosa dominates spectrum no. 2, with 45% occur A grave! horizon (layer F) between the lower late glacial rence in a gyttja-rich silt (layer G), but it has a low sediment sequence and the upper organic-rich Holocene occurrence in the spectra above and below. Spectrum no. deposits (layers G and H) probably represents a hiatus. 3 (gyttja-rich gravel, layer F) is dominated by Pinnularia Because of the possible hiatus the radiocarbon date of the spp. by more than 40%. The diatom assemblage in lower boundary of sediment layer F (8485 ± 120 BP) is sediment layers F, G and H indicates a lacustrine envi considered a minimum age for the earliest lacustrine ronment. deposits, and not the timing of basin isolation. NORSK GEOLOGISK TIDSSKRIFT 77 (1997) Marine transgression, Hardanger, West Norway 113 A B Basins with threshold only a few metres below marine limit: Basins wlth threshold more than 10 metres below marine limit: lnner Hardanger: The coast (Yrkje & Sotra): lnner Hardanger: The coast (Sotra): OUGO· .... HAL. OliGO- w HAL z OliGO- 00 w , a: 3! HAL. , o z mo m w o. Ol1GO -8 u. < HAL. og 9 ;;; �i < o .. (/) � J: (/) -' § �� :> i ow�§ o - :> < �· 0 ID 8 ID o c J: e �� i ID !!! � >-l- � o -' 9 z � � � a:Z < < o � -"' � -5 J: � 20 o. ... ... 1030 550 tso Fig. Il. The Late Weichselian transgression as documented by the diatom record from localities at the coast and inner Hardanger. O A. Total diagram for dia toms from (left to right): Vambheim-119 (tltis work), Meåstjørna (Braaten & Hermansen 1985) and Kvemavatn (Krzywinski & Stabell 1984). The last two basins are from Yrkje and Sotra, respectively. O B. Total diagram for diatoms from (teft to right): Bu-113 (this work), Klæsvatn and Herøyvatn (Krzywinski & Stabell 1984). The last two basins are from Sotra. The basins are situated a few metres below (A) and more than 10 m below (B) the local marine limit, respectively. Sea-level history inferred from the glaciation. On the contrary, the lower, thick lacustrine investigated basins facies in Vambheim-119 (sediment layers D to J) bear witness to long-lived lacustrine conditions during the In spite of a rather extreme 1o cation of the three investi early phase of the basin development. gated basins, situated at the head of a long, narrow After the initial (glacio-) marine invasion of Bu- 113 a fjord, we are confident that the biostratigraphic records, regression took place and the sea-level dropped below and thus the changes in pa1aeosalinity inferred from the basin threshold. The basin developed into an iso these records, actually do reflect sea-level changes (and lated lake and layer F was deposited, representing the not say salinity changes due to meltwater pulses). A first lacustrine facies recorded in Bu-113. comparison of diatom records with basin localities from the outer coast (Sotra and Yrkje) show striking similari The second marine facies in Bu-113 (sediment layers ties (Fig. 11). The compared basins are simi1arly situ G and H) was deposited as the sea again transgressed ated with respect to the loca1 marine limit (situated a the basin. During the transgression Vambheim- 119 was, few metres below and more than l O m below the local for the first time, invaded by the sea. The 1ower, thick marine limit, respectively). The marine and lacustrine lacustrine facies in Vambheim- 119 (sediment layers D to facies, as documented by the diatom records from l) is fo llowed by a marine facies (layer J). During the basins at the outer coast, are analogous to the diatom transgression-maximum, sea-level possibly also reached records of Bu-113 and Vambheim-1 19. The diatom the threshold of Vambheim-128 (sediment layers A to records thus most probably reflect corresponding sea E). The subsequent regression was probably rapid, in levet fluctuations (Fig. 12 a, b). terrupted only by a minor sea-level oscillation recorded Late Weichselian marine invasion of the lowermost as a lacustrine and a brackish facies in Bu-113 (sedi basin, Bu- 113, occurred with deglaciation of the ment 1a yers I and J, respectively). Hardangerfjord. The marine sediment layers D and E, Although it is not compatible with the established and probably also the diatom barren (glacio-marine?) g1acial chronology for the Hardangerfjord ( see Discus layers below, were deposited in this period. There are sion below) the transgression is assumed to have culmi no indications fr om the diatoms that sea-level also nated during Y o unger Dryas. The final isolation of reached Vambheim- 119 during the earliest phase of de- Vambheim-119 fr om the sea (layer boundary J/K) took 114 S. K. Helle et al. NORSK GEOLOGISK TIDSSKRIFT 77 (1997) a) m (a.s.l.) 130 • 14C dating; T-9654A (±1 a), lacustrine 125 ...._'"cdating; T-9619A (±1 a), isolation contact -'"cdating; Ua-2586 (±ta ), brackish Vambhelm-119 120 p e 14C dating (:!:1 a) of a piece of wood (Rye, 1969) '---""4cdating; T-9618A. 115 assunned to be too old Bu-113 P Pinus peak p (? mid-Younger Dryas) P Earty Holocene Pinus rise 110 V Maximum peak of Vedde-ash (? 10 300 BP) O Lacustrine facies 105 O Brackish facies IJ Marine facies 100 0 Late Weichselian Holocene l i l ....,__ Unknown'"C age •' l t 10 9 8 1 (nolto scale) 4 CkaB.P. b) m (a.s.l.) 50 lsotra (Stabell1982) ·Fig. 12. D a. Late glacial and early postglacial shore displacement in inner Hardanger. The leve! of the basin thresholds has not been cor rected for tilted uplift. The earliest part of the time-axis is not to scale and the height-axis is shortened. The lower diatom-barren part of the 25 stratigraphy of Bu-113 and Vambheim-119 is not included in the sea-leve! reconstructions. Vambheim-128 probably has a hiatus (dashed line) between the marine layers and the lacus trine layers above ( see text). The age indicators included are discussed in detail in the chrono stratigraphic sections. A 14C date (9680 ± 90 BP) on wood from a terrace in Eidfjord with its o surface 23 m a.p.s.l. is also included assumed to 4 represent the age of a sea-leve! of a bo ut l 00 m 12 10 8 6 4 2 O 1 CkaB.P. a.p.s.l. (Rye 1969). D b. Shore displacement curves from Bømlo (Kaland 1984) and Sotra (Stabell 1982). See Fig. l for location. place before the brackish episode of Bu- 113 (layer J), lished glacial chronology for western Norway. According which is dated to about 10 000 BP. Vambheim-119 was to previous works (e.g. Undås 1963; Holtedahl 1967; thus probably isolated in the late Younger Dryas (possi Pollestad 1972; Aarseth & Mangerud 1974; Holtedahl bly around 10 300 BP). The final isolation of Bu-113 is 1975; Sindre 1980; Hamborg & Mangerud 198 1) the mid dated to immediately before about 9700 BP (layer and ioner parts of the Hardangerfjord were ice-covered boundary J/K). during the Y o unger Dryas. Based on the evidence fr om ice-overrun marine sediments of Aller Holocene. In the inner and central parts of the Hardan Mangerud ( 1987) found evidence of two transgressions gerfjord only minimum age determinations for the timing in the Y ounger Dryas. of deglaciation have so far been available (e.g. Anundsen & Simonsen 1967; Rye 1969), thus pointing towards ( 4) A substantial tilting of the landmass in western early Holocene ice-free conditions. Norway took place at the beginning and early part of the However, the sea-leve! data in the present work indi Holocene. Differential uplift in this period is probably cate an earlier deglaciation of the head of the Hardanger responsible for most of the difference in altitude between fjord. Despite the inconsistency of the obtained radio the highest shore-levels in inner Hardanger and those at carbon dating results, the fo llowing arguments advocate the outer coast. The regression rate during this period in that the marine transgression of inner Hardanger took inner Hardanger was probably extremely high. Rye place during the Y o unger Dryas chronozone: ( 1969) estimated the sea-leve! in Eidfjord, inner Hardan ger, to be about 100 m a.p.s.l. at 9680 ± 90 BP (Fig. 12a). (l) The shore displacement curve fo r inner Hardanger A radiocarbon dating of shell found in a terrace with a suggests that the highest shorelines in the inner fjord surface at 68 m a.p.s.l. in Øystese, in the central part of branches developed during the transgression maximum the Hardangerfjord, gave 9420 ± 130 BP (B. G. An (Fig. 12a). Thus, shorelines developed prior to the trans dersen pers. comm. in Hamborg 1983). This indicates a gression ( deglaciation shorelines) are situated well below sea-leve! of approximately 70 m a.p.s.l. in mid-Hardan this maximum level. This is in contrast to previous gerfjord at this time. The high rate of uplift reduces the reconstructions of early postglacial shorelines in the likelihood of a marine transgression in inner Hardanger Hardangerfjord (e.g. Hamborg 1983) where the marine in the early Holocene. limits are defined to represent the initial entry of the sea Anundsen & Fj eldskaar (1983) and others concluded at the time of deglaciation. Our sea-leve! reconstructions that the Y o unger Dryas transgression can partially be are more in accordance with the late glacial sea-leve! attributed to lesser glacio-isostatic rebound during development at the outer west coast, where the Y ounger Y o unger Dryas. In inner Hardanger the transgression Dryas shoreline, developed during the transgression max probably took place in this glacio-isostatically quiescent imum, is documented to Iie higher than any older shore period. This implies that little or no differential uplift (til t lines (Anundsen 1985). ing) occurred between the outer coastal areas of western Norway and inner Hardanger during Younger Dryas. (2) An estimate of the amplitude of the transgression in inner Hardanger can be calculated as the height differ In order to accommodate the recorded sea-leve! ence between the lowest-lying basin, Bu-113, and the changes in inner Hardanger with the established sea-leve! highest nearby marine terraces; the terraces at Bj otveit history of the outer west coast we propose that an open, and Tveisme (Fig. 1). The terraces are situated about ice-free corridor fr om the coast to the head of the 128 and 125 m a.p.s.l., respectively (Holtedahl 1975), Hardangerfjord existed during Y ounger Dryas time. This yielding a height difference of 12- 15 m. If the sea-leve} suggests a more complex ice-front position than hitherto was above the basin threshold of Vambheim- 128 during understood, possibly with icelobes descending from the the transgression maximum (as indicated by the occur nearby mountains and into the fjord tributaries without rence of planktonic fo raminfera fo und in this basin), entirely filling the main fjord. the amplitude of the transgression could be close to During the Allerød the margin of the main ice sheet in 15 m. However, the distance between Vambheim and western Norway withdrew several tens of km inland Bu is about 3.6 km normal to N-S trending isobase (Andersen et al. 1995). The Hardangerfjord was proba lines and the level of the basin thresholds are not cor bly deglaciated in this period, possibly with ice domes rected for an y differential ( tilted) uplift. The latter es ti left isolated on coastal mountains. The ice-overrun mate of 15 m of the amplitude must therefore be marine sediments of Allerød and Y ounger Dryas age considered a maximum. The Y ounger Dryas marine fr om the outer areas of the Hardangerfjord (Aarseth & transgression at the outer coast of western Norway is Mangerud 1974; Holtedahl 1975) could have been influ also estimated to have an amplitude of 12-15 m (Anund enced by a late Y ounger Dryas readvance of ice located sen 1985). on coastal mountains possibly separated from the main Y o unger Dryas inland ice sheet. (3) After the main transgression in inner Hardanger This scenario is more in accordance with the proposals there are indications of a minor sea-leve! fluctuation at of Kaldhol (1941) and Undås (1944). Without the sup Bu-113 dated to approximately 10,000 BP (Fig. 12a). A port of 14C dates they correlated the prominent ice-front similar fluctuation has also been recorded at the coast of deposits at the fjordheads of inner Hardanger with the western Norway. At Yrkje, North Rogaland, a minor Ra-moraines in eastern Norway. Also Simonsen (1963) sea-leve! fluctuation is reported to have taken place after and Anundsen ( 1964) denoted that the Eidfjord-Osa the main transgression, and before the early Holocene moraine ridge system at the head of the Hardangerfjord sea-leve! regression (Anundsen & Fj eldskaar 1983). At and in the nearby mountain plateau possibly could corre Gurskøy in Sunnmøre, NW Norway, Svendsen & spond to the Ra-moraines, thus suggesting that these 116 S. K. Helle et al. NORSK GEOLOGISK TIDSSKRIFT 77 (1997) could be Younger Dryas ice-front deposits. Later, how Fægri, K. 1940: Quartiirgeologische Untersuchungen im westlichen Norwegen. Il. Å ever, they concluded (Anundsen & Simonsen 1967) that Zur spiitquartiiren Geschichte Jærens. Bergen Museums rbok 1939-40, naturvitenskapelig rekke 7, 201 pp. the prominent moraine ridge system was attributed to a Fægri, K. 1944: Studies on the Pleistocene of westernNorway. Ill. Bømlo. Bergen Preboreal ice-front advance. Museums Årbok 1943, naturvitenskapelig rekke 8, 100 pp. The sea-level data in the present work show that the Fægri, K. & Iversen, J. 1975: Textbook of Pollen Analysis. Munksgaard, Capen hagen. 259 pp. course and timing of deglaciation of the Hardangerfjord Galehouse, J. S. 1971: Point counting. In Carver, R. E. (ed.): Procedures in are still in question. Future work, combining both sea Sedimentary Petro/ogy, 385-407. John Wiley, New York. levd and glacial-geological data fr om coastal and inland Genes, A. N. 1978: Glacial geology of the island Stord, west Norway. Norsk Geologisk Tidsskrift 58, areas, rnay solve the contradictory puzzle of the Y ounger 33-50. Haftidason, H. 1983: The marine geology of Eyjafjiirdur,North Iceland: sedimen Dryas ice in the Hardangerfjord. tology, petrographical and stratigraphical studies. Unpublished thesis, Univer sity of Edinburgh. 281 pp. Hamborg, M. 1983: Strandlinjer og isavsmelting i midtre Hardanger, Vest-Norge. Acknowledgements. - We express our deepest gratitude to the late Professor Karl Norges geologiske undersøkelse 387, 39-70. Anundsen. A supervisor and colleague, but most of all a dear friend, has passed Hamborg, M. & Mangerud, J. 1981: En rekonstruksjon av isbevegelser under away. He will be sadly missed. siste istid i Samnanger og Kvam, Hordaland, Vest-Norge. Norges geologiske We !hank Statoil and the Swedish Nuclear Power Inspectorate for financial undersøkelse 369, 77-98. support. Thanks to the European Commission through the ENAM (European Helle, S. K. 1993: Strandforskyvning i indre Hardanger, Vest-Norge. En lito- og North Atlantic Margin) programme (Haftidi Haftidason). We also thank: Tom biostratigrafisk undersøkelse av 3 myrbassenger i Ulvik og Ullensvang. Unpub Schistad, Leif Harald Flatekval, Kai Morten Flatekval, John Erik Skare, Njål lished thesis, University of Bergen. 87 pp. Fodnes and Heine Helland for field assistance; Leif Harald Flatekval and Trond Holtedahl, H. 1967: Notes on the formation of fjords and fjord-valleys. Geo Kui for analysis of pollen; Josef Kusior for developing and improving the drilling grafiska Annaler 49A, 188-203. equipment; Else Hansteen Lier and Jane Ellingsen for help with the illustrations; Holtedahl, H. 1975: The geology of the Hardangerfjord, West Norway. Norges Ole Tumyr for assistance on the electron microcope. Thanks to Jon Eiriksson and geologiske undersøkelse 323, Bjørg Stabell for their valuable comments on the manuscript. We also owe thanks 1-87. O to Edward L. King who helped improve the English in the manuscript. Hustedt, F. 1930-1966: Die Kieselalgen Deutschlands, sterreichs und der Schweitz. In Rabenhorst, L. (ed.): Kryptogamen-Flora von Deutschland, Osterreich und der Schweitz. Akademische Verlagsgesellschaft, Vol. l-Ill. Manuscript received February 1996 Leipzig. Hustedt, F. 1930: Bacillariophyta (Diatomeae). In Pascher, A. (ed.): Die SujJ wasserj/ora Mi/te/europas, Heft 10. Fischer, Jena, 466 pp ( +875 fig.). Hustedt, F. 1957: Die diatomeenflora des FlujJ-Systems der Weser im Gebiet der References Hansestadt Bremen. Abh. Naturw. Ver. Bremen 34, 181-440. Ingmar, T. 1973: Sjiiavsniiringar från aktualgeologiska synspunkter. En oversikt. Aarseth, I. & Mangerud, J. 1974: Younger Dryas end moraines between Hardan Lundqua Report 3, 48-90. gerfjorden and Sognefjorden, western Norway. Boreas 3, 3-22. Kaland, P. E. 1984: Holocene shore displacement and shorelines in Hordaland, Alles, E., Niirpel-Schempp, M. & Lange-Bertalot, H. 1991: Zur Systematik und western Norway. Boreas 13, 202-242. Okologie charakteristischer Eunotia- arten (Bacillariophyceae) in elektrolytar Kaldhol, H. 1941: Terrasse- og strandlinjemålinger fra Sunnfjord til Rogaland. 206 men Bachoberliiufen. Nova Hedwigia 53, 171-213. pp. Hellesylt. Andersen, B. G., Mangerud, J., Sørensen, R., Reite, A., Sveian, H., Thoresen, M. Krammer, K. & Lange-Bertalot, H. 1986: SujJwasserflora von Mitte/europa, Band & Bergstrørn, B. 1995: Younger Dryas ice-marginal deposits in Norway. 2/ Bacil/ariophyceae. Teit Naviculaceae, 876 pp, 206 plates. Fischer, Stutt Quaternary International 28, 147-169. l, l. gart. Anundsen, K. 1964: Kvartærgeologiske og geomorfologiske undersøkelser i Krammer, K. & Lange-Bertalot, H. 1988: SujJwasserflora von Mi/te/europa, Band Simadalen, Eidfjord, Måbødalen, Hjølmodalen og tilstøtende fjellområder. 2/2, Bacil/ariophyceae. 2. Tei/ Epithemiaceae, Bacil/ariaceae, Surirellaceae, 596 Unpublished thesis, University of Bergen. 81 pp. pp, 187 plates. Fischer, Stuttgart. Anundsen, K. 1977: Sediments, pollen and diatoms from two basins in south Krammer, K. & Lange-Bertalot, H. 1991: SujJwasserj/ora von Mitteleuropa, Band western Norway. Reports from the Department of Geo/ogy, University of 2/3, Bacillariophyceae. 3. Tei/. Centrales, Fragilariaceae, Eunotiaceae, Trondheim, the Norwegian Institute of Technono/gy I. 43 pp. 576 pp, Anundsen, K. 1978: Marine transgression in Younger Dryas in Norway. Boreas 166 plates. Fischer, Stuttgart. 7, 49-60. Krzywinski, K. & Stabell, B. 1984: Late Weichselian sea leve! changes at Sotra, Anundsen, K. l 985: Changes in shore-level and ice-front position in Late Hordaland, western Norway. Boreas 13, 159-202. Weichselian and Holocene, southern Norway. Norsk Geografisk Tidsskrift 39, Lange-Bertalot, H. & Krammer, K. 1989: Achnanthes-eine Monographie der 205-225. Gattung mit Definition der Gattung Cocconeis und Nachtriigen zu den Navic Anundsen, K. & Fjeldskaar, W. 1983: Observed and theoretical late Weichselian ulaceae. Bib/iotheca diatomologica 18, l- 393, l 00 plates. shore-leve1 changes related to glacier oscillations at Yrkje, southwest Norway. Mangerud, J., Lie, S. E., Furnes, H., Kristiansen, I. L. & Lørno, L. 1984: A In Schroeder-Lanz, H. (ed.): Late- and Postg/acia/ Oscil/ations of Glaciers: Younger Dryas Ash Bed in Western Norway, and its possible correlations with Glacial and Periglacial Forms, 133-170. A. A. Balkema, Rotterdam. tephra in cores from the Norwegian Sea and the North Atlantic. Quaternary Anundsen, K. & Simonsen, A. 1967: Et Preborealt breframstøt på Hardan Research 21, 85-104. gervidda og i området mellom Bergensbanen og Jotunheimen. Årbok for Meldgaard, S. & Knudsen, K. L. 1979: Metoder til indsamling og oparbejdning Universitetet i Bergen. Matematisk Naturvitenskapelig Serie 7, 42 pp. af prøver til foraminifer-analyser. Dansk Natur - Dansk skole, Årskrift 1979, Bard, E., Arnold, M., Mangerud, J., Paterne, M., Labeyrie, L., Duprat, J., 48-57. Melieres, M.-A., Sønstegaard, E. & Duplessy, J.-C. 1994: The North Atlantic Palmer, A. J. M. & Abbott, W. H. 1986: Diatoms as indicators of sea-leve! atmosphere-sea surface gradient during the Y o unger Dryas climatic event. change. In van de Plassche, O. (ed.): Sea-leve/ Research; a Manual for the Earth and Planetary Science Letters 126, 275-287. Col/ection and Evalution of Data, 457- 487. Geo Books, Norwich. Birks, H. H., Gulliksen, S., Haftidason, H., Mangerud, J. & Possnert, G. 1996: Patrick, R. & Reimer, C. W. 1966: The diatoms of the United States, vol. I. New radiocarbon dates for the Vedde Ash and the Saksunarvatn Ash from Monographs of the Academy of Natura/ Sciences of Philadelphia 13, 1-688. Western Norway. Quaternary Research 45, 119-127. Paus, Aa. 1988: Late Weichselian vegetation, climate, and floral migration Braaten, A. M. & Hermansen, D. 1985: En lito- og biostratigrafisk undersøkelse at Sandvikvatn, North Rogaland, southwestern Norway. Boreas 17, 113- av marine og limniske sedimenter i Yrkje, Nord-Rogaland. Unpublished thesis, 139. University of Bergen. 205 pp. Peragallo, H. & Peragallo, M. (1897-1908): Diatomees marines de France et des Cleve-Euler, A. 1951-1955: Die Diatomeen von Schweden und Finn/and. Kungliga districts maritimes voisins. Micrographie-Editeur. Grez-sur-Loing. 491 pp. Svenska Vetenskapsakademiens Handlingar, Vol. I-V, Stockholm. Pienitz, R., Lortie, G. & Allard, M. 1991: lsolation of lacustrine basins and Florin, M.-B. 1945: Skiirgårdstall och 'strandskog' i viistra Siidermanlands pol marine regression in the Kuuijuaq area, northern Quebec, as inferred from lendiagram. Geo/ogiska Foreningens i Stockholm Forhand/ingar 67, 511-513. diatom analysis. Geographie physique et Quaternaire 45, 155-174. Follestad, B. A. 1972: The deglaciation of the southwestern part of the Folgefonn Qvale, H. 1981: Ulvik, Berggrunnsgeologisk Kart 1316 Il - M l: 50,000, peninsula, Hordaland. Norges geologiske undersøkelse 280, 31-34. foreløpig utgave. Norges geologiske undersøkelse. NORSK GEOLOGISK TIDSSKRIFT 77 (1997) Marine transgression, Hardanger, West Norway 117 Rekstad, J. 1911: Geologiske iakttagelser fra nordvestsiden av Hardangerfjorden. Simonsen, R. 1987: Atlas and Catalogue of the Diatom Types of Friedrich Hustedt. Norges geologiske undersøkelse Årbok 1911, No. 2. 62 pp. Vols. 1-3, 542 pp. + 772 plates. Cramer, Berlin-Stuttgart. Rochon, A. & de Verna!, A. 1994: Palynomorph distribution in Recent sediments Sindre, E. 1980: Late Weichselian ice-front oscillations in the Hardanger from the Labrador Sea. Canadian Journal of Earth Sciences 31, 115-127. Sunnhordland District, West Norway. Continental ShelfInstitute (!KU), publ. Round, F. E., Crawford, R. M. & Mann, D. G. 1990: The Diatoms. Bio/ogy and no. 102. 16 pp. Morphology of the Genera. Cambridge University Press. 741 pp. Stabell, B. 1982: Changes in diatom flora in late Quaternary western and southern Rye, N. 1969: Einergrein av preboreal alder funnet i israndavsetning i Eidfjord, Norwegian marine and freshwater sediments: response to basin isolation from Vest-Norge. Norges geologiske undersøkelse Årbok 1969, 33-36. the sea. In The Response of Diatom floras During Late Quaternary Shore Line Schmidt, A., Scmidt, M., Fricke, F., Heiden, H., Muller, O. & Hustedt, F. Disp/acement in Southern and Western Norway, 193-323. Dr. scient. thesis. (1874-1959): Atlas der Diatomaceenkunde. 472 plates. R. Reisland. Ascher University of Oslo. leben. Leipzig. Svendsen, J. I. & Mangerud, J. 1987: Late Weichselian and Holocene sea-levet Schroeder, H. 1939: Die Algenftora der Mulde. Pftanzenforschung 21, 1-88. history for a cross-section of western Norway. Journal of Quaternary Science 2, Sigmond, E. M. 0., Gustavson, M. & Roberts, D. 1984: Berggrunnskart over 113-132. Norge. M 1:1 million - Norges geologiske undersøkelse. Simonsen, A. 1963: Kvartærgeologiske undersøkelser i indre Hardanger. Unpub Undås, I. 1944: Sørfjordbygdene i seinglacial og postglacial tid. Festskrift for lished thesis, University of Bergen. 61 pp. Ullensvang Hagebruks/ag. Simonsen, A. 1980: Vertikale variasjoner i Holocene pollensedimentasjon i Ulvik, Undås, I. 1963: Ra-morenen i Vest-Norge, 78 pp. J. W. Eide, Bergen. Hardanger. AmS Varia 8, 1-68. Undås, I. 1964: When were the heads of the Hardangerfjord and the Sognefjord Simonsen, R. 1962: Untersuchungen zur Systematik und Okologie der Boden ice-free? Norsk Geografisk Tidsskrift 19, 291-295. diatomeen der Westlichen Ostsee. Internationale Revue der Gesamten Hydro Vorren, T. O. 1972: Interstadial sediments with rebedded interglacial pollen from biologi, Systematiche Beiheft l, 144 pp. Akademie-Verlag, Berlin. inner Sogn, West Norway. Norsk Geologisk Tidsskrift 52, 229-240.