Relatively little data on the distribution of Antarctic icebergs were available prior to 1980. The published literature included size data of about 5000 icebergs, and position data of 12 000 icebergs. There were indications that the size data were biased in favour of larger icebergs.
A programme of systematic iceberg observations was therefore initiated by Norsk Polarinstitutt in 198! through the SCAR Working Group on Glaciology. This programme is based on standard “blue” forms distributed to all ships going to Antarctica. The icebergs are recorded every 6 h and in Five length groups: 10–50, 50–200, 200–500, and 500–1000 m, and those over 1000 m are described individually.
The amount of data has increased greatly from the start in 1981–82. The position of 70 000 icebergs, including 50 000 that had been size classified, were on file at Norsk Polarinstitutt by December 1985, and the data set is growing rapidly. Most ships travelling to and from Antarctica now participate in collection of the data. (Fig.1 shows the locations of the icebergs sighted.)
Fig. 1.
Location of iceberg observations under the programme initiated in 1981. Main ship tracks are clearly reflected. The average observation represents 14 icebergs.
The size distribution of the classified icebergs observed under this programme up to December 1985 is given in Table I:
Table I
The “standard size” (length, width, and thickness) is based on our observations from three Antarctic expeditions which carried out dedicated iceberg studies. Many icebergs are of course not right-angled parallelepipedal in shape, but this is a good approximation for most of the larger icebergs.
The data are based both on visual sightings and on radar observations. Duplicate observations from a ship moving at slow or zero speed are as far as possible eliminated, both during observation, and by critical appraisal before the data are filed. The data editing also includes evaluation of data quality, especially in connection with radar observations, and comparison of positions and dimensions of the large icebergs in order to reduce to a minimum repeated observations from different vessels of icebergs >1000 m. These account for most of the iceberg mass (see Table I).
Consideration of iceberg-distribution patterns and the observed area of the Southern Ocean, and of duplicate observations, indicates more than 300 000 icebergs south of the Antarctic Convergence, with a total ice mass of about 1016 kg. Consideration of mean residence times indicates an annual iceberg production from the continent of 23–1015 kg, which is considerably higher than most other recent estimates. This also suggests that the Antarctic ice sheet is in balance.
The data indicate large regional differences in iceberg sizes, the most noticeable being between the two sides of the Antarctic Peninsula, and between the Amery Ice Shelf/ Prydz Bay area and the remainder of East Antarctica. These differences are probably mainly related to different calving sites.
About one-third of the observed icebergs are over the continental shelf of Antarctica. The total under-water area of these icebergs is two orders of magnitude less than the under-water area of the Antarctic ice shelves. The annual total iceberg melting and its effect on the water masses over the continental shelf has been calculated from ocean-water temperature variations at 200 m depth and estimated melt rates. This turns out to be an order of magnitude less than the annual effect of melting sea ice. The iceberg data considered here are probably under-represented with respect to the smallest sizes, and they do not include icebergs that have become <10 m. Inclusion of these ice bodies would increase the total melt.
Connecting Antarctic Cross-Slope Exchange with Southern Ocean Overturning
