The amount and type of driftwood logs on raised beaches in Svalbard probably indicate variations in sea ice conditions through time in the Arctic Ocean. This idea is based on the fact that wood floating in water has a limited buoyancy. The maximum period of buoyancy for coniferous wood (Picea/spruce, Larix/larch, Pinus/pine) is between 10 and 17 months: for broadleaves (Betula/birch, Salix/willow, Populus/aspen) from 6 to 10 months. The buoyancy depends on many circumstances but foremost on the properties and basic density of wood when the drift starts. Two dominant drift features are evident in the Arctic Ocean. The Pacific Gyral has a clockwise circulation in the Beaufort Sea. There are large variations in the rate of drift of this circulation, but the average speed is probably 700-800 km/year. The Transpolar Drift Stream runs from the Siberian coast to Svalbard and north Greenland and the average drift rate 600 km/year may be close to the actual net speed of an ice floe or a log of driftwood. The passage from the Siberian coast to the Svalbard area takes, roughly, five years. A trunk with the best possible buoyancy sinks after a third of that time. However, because of rafting on or in the sea ice, driftwood logs land on the islands of Svalbard. Most of the wood on the eastern islands of the archipelago is coniferous wood of Siberian origin. There are at least two easily noticeable zones on the shores with raised beaches where the amount of driftwood is small. These zones probably represent climatic conditions with periods of summers; a high percentage of open water and a northerly position of the ice limit. Introduction and field work During the YMER-80 expedition in July and August 1980, field investigations were made in the Svalbard area (Schytt 1981). The research of the land parties was concentrated to studies of the present and Pleistocene glaciation of Svalbard. A field party with six members worked primarily with lake sediments in Nordaustlandet and another party on Storoya with glaciological investigations. A third group, Lars Brydsten, Anders Haggblom and Ann-Cathrine Ulfstedt started the field work on Glenhalvoya, Nordaustlandet. Brydsten and Haggblom continued the field investigations at Braastadskardet, the north coast of Hopen and later on Kapp Platen, Nordaustlandet. The icebreaker Ymer called at both Stor6ya and Murchison Bay (Kinnvika, the old polar research station from I.G.Y. 1957-59) on her way to Longyearbyen. Members of the landparties left Svalbard by the regular airline to Norway. Already after our investigations on Hopen in 1965 (cf. Hoppe et al. 1969) the interest was focused on driftwood. An interpretation of both the origin of wood and the varying amount of driftwood on the raised beaches seemed important. The field studies of beach morphology and Arctic driftwood were therefore included as a part of the land-based research during YMER-80. The icebreaker and research vessel Ymer was an excellent base for both the scientific work and the important transports. Driftwood in the Arctic The occurrence of driftwood along Arctic beaches was noted by the first polar explorers. The driftwood along the beaches of Iceland was described in detail as regards type, amount, economic usefulness and origin by Olavius in an economic report from Iceland (Olavius 1780). The biological context of the driftwood in northern Svalbard, particularly that from Ltagoya, was described by Pennant in 1784. This resource had been exploited by people of the Arctic cultures long before that. When Eurola (1971) published a summary of earlier studies concerning Arctic driftwood, he noted that one hundred years had passed since Agardh's (1869) classical work on driftwood appeared. Eurola also describes botanical analyses of the wood structure of 19 of his own samples from Agardh Bay in Spitsbergen. Common to previous studies of driftwood have been analyses of wood structure in order to Geografiska Annaler 64 A (1982) 1-2 81 This content downloaded from 157.55.39.144 on Wed, 07 Sep 2016 05:24:27 UTC All use subject to http://about.jstor.org/terms