On melting icebergs

THE feasibility of towing icebergs to their coasts and melting them on arrival to provide a supply of fresh water is being studied in various parts of the world. Saudi Arabia, Australia and California1 are amongst those considering such a project. One suggestion for melting the icebergs is to run each iceberg aground in water approximately 250m deep, the mean depth of Antarctic icebergs. A relatively shallow pen would then be built around the iceberg and it is conjectured that the melt water will rise, without much mixing, into the pen, from where it will be siphoned off for subsequent use. Neshyba2 claims, however, that melt water produced by icebergs may be responsible for a large amount of upwelling and mixing in the Antarctic's Weddell Sea. He asserts that as the melt water rises up the side of an iceberg, it entrains a sizeable quantity of warmer, saltier water from the environment. Neshyba calculates that an average-sized iceberg is thus responsible for a vertical volume transport into the upper layer of the Weddell Sea of 3.3 × 108 cm3 s−1. This is a considerable transport, and the raising of water and nutrients from depth by this mechanism would be of importance in determining the physical and biological properties of the Weddell Sea. The first suggestion, that fresh melt water rises to the surface, and Neshyba's claims that the melt water will have mixed significantly with its salty environment before reaching the surface, cannot both be correct. We show here that both may be in error.