Description: Late Quaternary sediment yields from the Isfjorden drainage area (7327 km**2), a high arctic region on Svalbard characterized by an alpine landscape, have been reconstructed by using seismic stratigraphy supported by sediment core analysis. The sediments that accumulated in the fjord during and since deglaciation can be divided into three stratigraphic units. The volumes of these units were determined and converted into sediment yield rates averaged over the drainage basin. During deglaciation, 13 to 10 ka, the sediment yield was ~860 tons(t)/km**2/yr. In the early Holocene it decreased to 190 t/km**2/yr, and then increased to 390t/km**2/yr during the late Holocene Little Ice Age. When normalized to the approximate glacierized area, these rates correspond to a sediment yield of ~800 t/km**2/yr . Sediment yield from non-glacierized parts of the drainage is estimated to be 35 t/km**2/yr. At times when ice advanced to the shelf edge, sediment was scoured from the fjord and deposited on the outer shelf and in a well-defined deep sea fan. Between 200 ka and 13 ka, 328 km**3 of sediment accumulated here, corresponding to a mean sediment yield rate of 335 t/km**2/yr. This is broadly consistent with calculations based on the above rates of sediment yield in glacierized and non-glacierized areas, and on estimates, based on glacial geology, of the temporal variation in degree of glacierization over the past 200 kyr. These figures indicate that much of the glacigenic sediment on the shelf and slope was eroded from the uplands of Svalbard by small glaciers during interstadials and interglacials. The sediments were temporarily stored in the fjord prior to redeposition on the shelf and slope during ice sheet advance. Taken into consideration, such redisposition of pre-eroded material will reduce estimates of primary ice sheet erosion rate.

Description: Data on glacial erosion have been compiled and synthesised using a wide range of sediment budget and sediment yield studies from the Svalbard-Barents Sea region. The data include studies ranging in timescale from 1 to 10**6 yr, and in size of drainage basin from 101 to 105 km**2. They show a clear dependence of sediment yield on the mode of glacierization. Polar glaciers erode at rates comparable to those found in Arctic fluvial basins, or about 40 t/km**-2/ yr or 0.02 mm/yr. In contrast, rates of erosion by polythermal glaciers are 800-1000 t/km**2/ yr (or ca 0.3-0.4 mm/yr), while rates from fast-flowing glaciers are slightly more than twice this: 2100 t/km**2/yr (or 1 mm/yr). Similar rates are also found for large glacierized basins like those in the southwestern parts of the Barents Sea. In contrast to the situation in fluvial basins, in which sediment yield typically decreases with increasing basin size, the tendency in glacierized basins is for erosion to be independent of basin size. In studies of sediment yield from glaciers it is sometimes difficult to distinguish between material actually dislodged from the bedrock by glaciers and material dislodged by other processes in interglacial times and simply transported to a depocenter by a glacier. Our data suggest that pulses of sediment resulting from advance of a glacier over previously-dislodged material last on the order of 10**3 yr, and result in inferred erosion rates that are approximately 25% higher than long-term average rates of glacial erosion. The maximum sediment fluxes from the large Storfjorden and Bear Island drainage basins occurred in mid-Pleistocene. The onset of this period of high sediment yield coincided with the onset of the 100 kyr glacial cycle. We presume that this was the beginning of a period of increased glacial activity, but one in which glaciers still advanced and retreated frequently. During the last two to four 100 kyr cycles, however, sediment yields appear to have decreased. This decrease may be the result of the submergence of the Barents Sea. Glacier erosion would be much higher for a subaerial Barents Sea setting than it would be for a present day subsea Barents Sea. A classical question in Quaternary Geology is whether glaciers are more erosive than rivers. We surmise that if factors such as the lithology and the available potential energy (mgh) of the precipitation falling at a given altitude, whether in liquid or solid form, are held constant, then glaciers are vastly more effective agents of erosion than rivers.