Glacial inception and Quaternary mountain glaciations in Fennoscandia
Introduction
The last Fennoscandian deglaciation is well constrained. Much less is known about the preceding stages in the glacial history, since subsequent glaciers overrode and destroyed a large portion of both stratigraphic and geomorphological records. However, this is true only for terrestrial data; deep-sea marine records remain unaffected by glacial erosion. This review compares stratigraphic records, both marine and terrestrial, with geomorphological data in an attempt to characterize regional ice caps and ice fields in the Scandinavian mountains. The primary focus of this paper is glacial inception and its relation to mountain glaciations (MGs). The second objective is to explore the geomorphological impact of MGs in and close to the Scandinavian mountain chain. Marine-, near shore-, terrestrial stratigraphic data, and geomorphological data are used to guide this review.
MGs are defined as glaciers that are both dynamically and climatologically dependent on an elevated topography. In practice, this implies that MG constitute cirque glaciers, valley glaciers and ice fields (Østrem, 1974; Sugden and John, 1976), or a succession of all these glacier types. As soon as the glacier becomes independent of topography it is defined as an ice cap or ice sheet. The best modern analogues of the glaciers envisioned in this paper are probably the Patagonian ice fields.
Glacial inception has long been a subject for discussion. The classical model (Flint, 1943), emphasizes “a highland origin, windward growth” based on observations in North America. However, Enquist (1918) showed that cirque formation was more prominent on the lee-side of the Scandinavian mountains due to wind drift of snow. Andrews et al. (1970) invoked a similar lee-side effect to explain differences in glacierization levels on Baffin Island. A third model was proposed by Ives et al. (1975) involving an albedo feedback effect on large polar plateau areas potentially generating an almost “instantaneous glacierization”. A model experiment by Payne and Sugden (1990b) showed that topography potentially causes a strong climatic feedback resulting in accelerated glacier growth. The only thing that seems clear is that topography plays an important role for glacial inception, and that there is a delicate relationship between climate, topography, and glacierization.
MGs based in the Fennoscandian mountains seem to have persisted throughout a very large portion of late Cenozoic time (Porter, 1989; Mangerud et al., 1996; Kleman and Stroeven, 1997; Thiede et al., 1998). Kleman and Stroeven (1997) calculated that mountain-centered ice caps and ice fields have persisted for approximately 50% of the last 2.75 million years (Ma). However, Mangerud et al. (1996) estimate that medium sized mountain-based ice sheets have partly covered Scandinavia for as much as about 90% of the last 2.6 Ma. This difference stems from the use of different proxy records for ice volume, and dissimilar criteria to separate the modes of glaciation. The potential role of MGs is nevertheless evident no matter which model is adopted. Indeed, Rudberg (1988), Rudberg (1992), Porter (1989) and Kleman and Stroeven (1997) argue that MGs had a very large geomorphological impact in areas close to the mountains. Scandinavian examples of large-scale landforms that may have been shaped by MGs are fjords, piedmont lakes, and U-shaped valleys (Rudberg, 1992).
There are three ways of investigating specific ice-sheet configurations: the stratigraphic approach, the geomorphological approach and the ice-sheet modelling approach. The stratigraphic approach emphasizes investigations at one or a few key localities. Stratigraphic investigations often yield a chronology but a poor control on spatial configuration of the ice sheet (e.g. Mangerud et al., 1996). Geomorphological investigations can be applied at the ice-sheet scale and are thus well suited for spatial studies. The problem with this method is that different glacial events and configurations can not be dated other than in relation to each other, and assigning an absolute chronology to a sequence of geomorphological landforms is often difficult. One attempt to overcome this problem is to match the landform record to a well dated proxy for global ice volume (Mangerud et al., 1996; Kleman et al., 1997; Kleman and Stroeven, 1997; Boulton et al., 2001). This is a powerful approach that appears useful for at least the last glaciation. The major drawback of this methodology is that it is unclear how well a proxy for global ice volume corresponds to timing and volume of local or regional ice-sheet configurations. In short, it is often difficult to combine chronological with land form records; landform assemblages may thus “float” considerably in time, and chronological data may be detached from the overall ice-sheet pattern.
One attempt to merge the spatial approach with the chronological approach is by numerical ice-sheet modelling. However, the most advanced glaciological models still suffer inadequate description of several physical processes. Most models are validated against geological data (e.g. Hughes, 1998; Näslund, in press); precautions must therefore be taken to avoid circular arguments if models are used to explain stratigraphic or landform records. In this paper, recent modelling efforts are reviewed and compared to field data.
Section snippets
Timing of mountain-centered glaciations in Fennoscandia—the sedimentary record
The fundamental pacing of glacial cycles is to a large extent known through work on deep-sea data (e.g. Hays et al., 1976; Martinson et al., 1987; Imbrie et al., 1993). However, mechanisms controlling this pacing are debated. The finer detail of the last few glacial cycles is well established through the use of marine (Bond et al (1992), Bond et al (1993)) and ice core records (Johnsen et al., 1995). Terrestrial records have also furthered the understanding of the last glacial cycles (e.g.
Spatial patterns of mountain centered glaciations—the geomorphological record
Early geomorphological work was to a large extent focused on the last deglaciation because the prevailing paradigm was that few if any glacial traces could survive the last glacial stage. At the turn of the last century some field geologists found traces of a relatively large mountain glaciation pre-dating the deglaciation (Svenonius, 1899; Sjögren, 1909; Tanner, 1914). This “glaciation” was difficult to fit into the deglaciation paradigm and was generally left without much further attention.
Modelling mountain centered glaciations
Glaciological models are key tools with which we can improve our understanding of ice sheets and the mechanisms that drive them. Ice-sheet models can yield time-slice “output” for continental ice sheets. This contrasts with stratigraphical and geomorphological approaches where chronological control is restricted to key localities (c.f. Kleman and Borgström, 1996). In modern models, different types of data and knowledge are merged, i.e. they are based on the physics of glaciers; they may be
Discussion and conclusions
Local to regional ice sheets have occupied the Scandinavian mountains throughout a large part of the late Pliocene and Pleistocene. Calculations of this cumulative MG duration range from 50% (Kleman and Stroeven, 1997) to 90% (Mangerud et al., 1996). In Kleman and Stroeven (1997), a marine ice volume proxy record was sliced at certain thresholds into three categories. One threshold delineates that cirque-type glaciers dominated, another isotope threshold defines full-scale glaciations. In
Acknowledgments
Constructive reviews by Jan Lundqvist and an anonymous referee significantly improved the paper. Discussions with Clas Hättestrand, Peter Jansson, Arjen Stroeven, Johan Kleman, Karna Lidmar-Bergström, Stig Jonsson, Ninis Rosqvist, Christer Jonasson, and Jens-Ove Näslund are greatly acknowledged. Ahlmann's fund, Lagrelius’ fund and Mannerfelt's fund provided resources for aerial photos and satellite images.
References (121)
- et al.
Mid-Pleistocene cosmogenic minimum-age limits for pre-Wisconsinan glacial surfaces in southwestern Minnesota and southern Baffin Islanda multiple nuclide approach
Geomorphology
(1999) - et al.
Palaeoglaciology of an ice sheet through a glacial cyclethe European ice sheet through the Weichselian
Quaternary Science Reviews
(2001) Origin and consequence of cyclic ice rafting in northeast Atlantic Ocean during the past 130,000 years
Quaternary Research
(1988)- et al.
The last interglacial–glacial cycle in NE Fennoscandiaa nearly continous record from Sokli (Finnish Lapland)
Quaternary Science Reviews
(2000) - et al.
A time dependent glaciological model of the Weichselian ice sheet
Quaternary International
(1995) High-resolution modeling of the advance of the Younger Dryas ice sheet and its climate in Scotland
Quaternary Research
(1999)- et al.
The Eem stable isotope record along the GRIP ice core and its interpretation
Quaternary Research
(1995) - et al.
Glacigenic debris flows on the North Sea trough mouth fan during ice stream maxima
Marine Geology
(1998) - et al.
Preglacial surface remnants and Quaternary glacial regimes in northwestern Sweden
Geomorphology
(1997) - et al.
Late Weichselian submarine debris flow deposits on the Bear Island trough mouth fan
Marine Geology
(1995)