A flotilla of "inflatable bladders" that helped map deep ocean currents has led oceanographers at Southampton University to discover defects in global weather forecasting computer models.
Southampton's oceanography centre, host to the world ocean circulation experiment, unveiled this month the first global maps of sub-surface ocean currents.
The new data, which challenges the assumptions about activity in the seas on which global warming forecasting models are based, will help improve the models' accuracy, says project director John Gould.
"Ocean currents play a vital role in monitoring the stability of the earth's climate system," Dr Gould says. "Our main objective is to create better computer models of ocean circulation."
The team's mission is to map ocean currents at depth rather than at just surface level. "We did not have the capability to monitor circulation at depth when the project started, but now we do," Dr Gould says.
The new map, which covers the the Pacific Ocean, was made possible by "neutrally buoyant floats" called Autonomous Lagrangian Circulation Explorers. The ALACE technology, which Dr Gould describes as "a sort of sub-surface weather balloon", was developed after the mapping project began in the early 1980s.
The bladders, which were developed in the United States, are less compressible than sea water. When the balloon's pressure level matches that of the sea, they remain at a stable depth - in this case set at about 900 metres. Each balloon has a small hydraulic pump that inflates a bladder to change the float's volume. Every two weeks, the pumps inflate to bring the balloons to the surface and the balloon transmits its position to a satellite.
Between 1990 and 1996, 303 ALACEs were deployed in the Pacific Ocean. Many balloons, which have a six-year life-span, are still operating, and the project is due to end in 2002. But the preliminary results have been revealing.
"No one has ever done this before," said Dr Gould, who was this week appointed director of the International Project Office of the CLIVAR project on climate variability and predictability, part of the world climate research programme at Southampton's oceanography centre.
"We have compared the computer models (of climate change) with our observations of the real world. The models have all underestimated the inherent variability of the ocean. The finding will be valuable for validating our computer models."