Dynamic sea-level change during the last deglaciation of northern Iceland MATS RUNDGREN, OLAFUR INGOLFSSON, SVANTE BJORCK, HUI JIANG AND HAFLIDI HAFLIDASON Rundgren, M., Ingolfsson, zyxwvutsrq O., Bjiirck, S., Jiang, H. zyxwvu & Haflidason, H. 1997 (September): Dynamic sea-level change during the last deglaciation of northern Iceland. zyxwvut Boreas, Vol. 26, pp. 201-215. Oslo. zyx ISSN 0300-9483. Born A detailed reconstruction of deglacial relative sea-level changes at the northern coast of Iceland, based on the litho- and biostratigraphy of lake basins, indicates an overall fall in relative sea level of about 45 m between 11 300 and 9100 BP, corresponding to an isostatic rebound of 77 m. The overall regression was interrupted by two minor transgressions during the late Younger Dryas and in early Preboreal, and these were probably caused by a combination of expansions of local ice caps and readvances of the Icelandic inland ice-sheet margin. Maximum absolute uplift rates are recordcd during the regressional phase between the two transgres- sions (10 000-9850 BP), with a mean value of c. zyxwvu 15 cm . I4C yr-' or 11-12 cm cal. yr-'. Mean absolute uplift during the regressional phase following the second transgression (9700-9100 BP) was around 6 cm I4C yr- I, corresponding to c, 3 cm 'cal. yr-I, and relative sea level dropped below present-day sea level at 9000 BP. Mats Rundgren and Hui Jiang, Department zyxwvut oj Quciternary Geology, Lund University, Tornmagen 13, S-223 6 . 3 Lund, Sweden; Olafur Ingdjkson, Earth Sciences Centre, University of' Goteborg, S-413 81 Giireborg, Sweden; Sounte Bjirck, Geological Insiitute, Univer.riry of Copenhagen, gster Voldgude 10, DK- 1350 Copenhagen, Denmark; Haflidi Hafiiduson, Geological Institute, University of Bergen, AI&aten 41, N-5007 Bergen, Norway; received 8th zyxwvutsrqpo Angusr 1996, accepted 30th June 1997. Raised beaches and marine terraces are common fea- tures along the coast of Iceland, indicating high rela- tive sea level in connection with the last deglaciation. It was recognized early that this high sea-level stand was followed by a period with sea levels lower than at present (Birdarson 1923), but it was not at that time possible to date this low-level event. Later studies of submerged peat deposits around the island have con- firmed an event of low sea level (Thorarinsson 1956; Jonsson 1957; Th. Einarsson 1961), and radiocarbon dates (throughout this paper, BP refers to uncali- brated I4C years) on submerged peat deposits in southwestern Iceland indicate that relative sea level fell below the present some time between 9500 and 9000 BP (Thorarinsson 1956; Kjartansson et al. 1964; Thors & Helgadottir 1991; Ingolfsson et al. 1995). Marine seismic reflection data from the same area (Thors & Helgadottir 1991), suggest that relative sea level was 30-35 m below the present level in the early Holocene. Similar data from Eyjafjordur, northern Iceland (Thors & Boulton 1991) indicate that relative sea level in that area reached -40 m in the Late Weichselian, followed by a transgression to -20 m in the early Holocene. In a recent paper, Ingolfsson et zyxwvuts LII. (1995) presented a stratigraphically controlled curve of relative sea-level displacement for southwestern Iceland. Their data suggest a fall in relative sea level of at least 45 m in the period 10 300-9400 BP, corresponding to a mean absolute uplift rate of 6.9 cm . I4C yr-'. Ingolfsson et al. (1995) argued that this rapid uplift most likely can be ascribed to a combination of rapid Preboreal de- glaciation and low asthenosphere viscosities below Iceland. The purpose of the present investigation is to analyse in detail the uplift patterns connected with the deglaciation of northern Iceland using the well-estab- lished technique of reconstructing relative sea-level changes from isolated lake-basin stratigraphies (e.g. Hafsten 1960; Bjorck & Digerfeldt 1982; Anundsen 1985). With this method it is possible to detect tempo- ral variations in isostatic rebound during deglaciation and possibly identify shorter periods of subsidence, enabling a connection to advances of the Icelandic ice sheet. The Skagi peninsula in northern Iceland (Fig. 1) was chosen as a suitable area for the investigation because (1) it carries numerous lakes at varying alti- tutes, (2) it is known to have been deglaciated rela- tively early, i.e. at least in the Aller~d period (Bjorck rt nl. 1992; Rundgren 1995), (3) it displays a distinc- tive pollen stratigraphy, which can be used for corre- lation between lake successions (Rundgren 1995), and (4) it features a series of pronounced raised beach ridges indicating that the area has experienced earlier relative sea levels at least 65 m higher than at present (Moriwaki 1990). Methods Our strategy was to collect sediments from a number of lakes at varying elevations between the marine limit and present sea level, and then to use diatom analysis and lithostratigraphic methods to identify marine, brackish and freshwater phases in these sediment suc- cessions. If transitions between sedimentary environ- ments could be dated, it would be possible to reconstruct relative sea-level changes in relation to