September 21, 2005

An e-Science project is enabling the use of near real-time Met Office forecasts to predict the drift of a person who has fallen overboard. It will be demonstrated at the e-Science All Hands meeting in Nottingham on September 21 at 14:30-16:30.

Ocean currents and surface winds push the person away from his or her original location. SARIS, a software package developed by BMT Cordah Ltd, a subsidiary of British Maritime Technology Ltd (BMT), helps the Coastguard to speed up rescue time by running a computer prediction of the person’s drift pattern. It uses a pre-stored database of tide information and requires the user to input wind data manually.

Now, however, an e-Science project, funded by the Natural Environment Research Council, has adapted SARIS to access near real-time Met Office forecasts of ocean currents and winds, thereby providing more accurate and complete data to the SARIS prediction. It is hoped that this will yield more accurate trajectories of drift and hence reduce the time to rescue, perhaps saving lives.

Met Office forecasts combine real observations with computer simulations to give the most accurate results possible. Professor Keith Haines and colleagues at the Reading e-Science Centre have developed a web service known as GADS (Grid Access Data Service) to allow the results of these forecasts to be accessed on demand over the Internet. BMT then adapted the SARIS software to obtain data from GADS to produce the new prototype system.

“The forecast data are large and complex,” explains Dr Jon Blower, one of the Reading team, “so the GADS Web Service lets you download just the data you need and presents it to you in your chosen format.” SARIS then integrates the forecast data with its own models to make its prediction of the position of the drifting person. So far, the use of forecast data in SARIS has been tested only in the lab. A next stage will be to test its predictive capabilities with a real object at sea.

This prototype system is to be developed further in the Met Office-led DEWS project (Delivering Environmental Web Services), a #2.2M DTI Inter-Enterprise Computing project. In DEWS, the e-Science techniques used in the SARIS demonstrator will be extended to other applications and made more robust for real use. DEWS will examine the integration of Met Office forecasts with a variety of data to provide more detailed and accurate information for maritime and health applications, including controlling oil spills at sea and predicting the occurrence of lung disease.

###
Contact:

Judy Redfearn
Engineering and Physical Sciences Research Council
44-776-835-6309
judy.redfearn@epsrc.ac.uk

This text derived from http://www.epsrc.ac.uk/default.htm

Recommend this Article to a Friend

Back to: News